The FeverBee Guide To Creating
Successful Community Strategies

In this guide you’ll learn:

  • The complete community discovery process.
  • How to track and use the right metrics. 
  • How to analyse and diagnose the problems of any community. 
  • How to select the right strategic approach for your community.





The Community Discovery Process


Organizational Analysis


Community Analysis


Community Results


Selecting The Right Strategy


Optimising A Community


Adapting Strategies


Repairing A Community


Relaunching / Closing A Community


Prioritising Your Solutions

If you have any questions at any point while reading this guide, feel free to ask us on:

Richard Millington
Richard Millington

Founder of FeverBee

1. Two Key Steps To
Successful Community Strategies


We’ve spent the last 14 years developing community strategies for the world’s largest (and often most complex) organizations. We’ve also trained close to 1300+ people in our community management academy.

This means we’ve seen hundreds of community strategies.

Some are lengthy 70+ page documents. Others are a deck of 10 slides or less. Some use common strategic planning frameworks, others use templates created by various industry resources.

Regardless of the length, format, and structure of the strategy, it’s become patently obvious the vast majority of community strategies will fail because they overlook the critical thing which makes a strategy succeed.

Two Key Steps To Develop A Strategy

Let’s strip strategy back to basics.

There are two basic requirements for a strategy to be successful. You have to diagnose the right problem(s) and prescribe the right solution(s). 

To build a successful community strategy, you must diagnose the problem and correctly prescribe the right solution.

The reason why most strategies have no chance of success is that they’re not trying to solve the right problems. 

It’s like a doctor prescribing medicine without first diagnosing the illness. She might get it right, but it’s unlikely. 

If you don’t know what problems you’re trying to solve, you can’t come up with a strategy to solve them. 

In fact, the solution part of the strategy doesn’t have to be all that great as long as it’s targeting the right problems. 

(Quick aside: While a problem implies a negative, this isn’t necessarily the case. There are also ‘good problems’ to have. For example, finding the best way of managing rapid growth is a ‘problem’ – but not an inherently negative one).

You Can’t Create An Effective Strategy By Template

In the past, you’ve probably prescribed popular tactics to improve your community. This might include changing platforms, implementing gamification programs, or expanding the MVP program.

But were you really sure what problem you were trying to solve? Or were you copying what others were doing? 

Did you really get the benefits you wanted from your tactics? 

You can spend months, years, or even an entire career doing busy work without ever really being able to point to a major achievement. This is what happens when you tend to jump to solutions without really diagnosing problems. 


Your community strategy has to be custom-designed for the unique situation your organisation finds itself in.

There Are Many Types of Strategies, Which Will You Select?

On the surface, it can seem that every major brand community is following the same approach. 

They select a good platform, launch a community, drive people to it, build a program for top members, and then moderate as best as they can. 

But these aren’t common strategies, they’re common tactics. Strategies are the overall approach to achieving a goal. 

Types of strategies include: 

  • Find the ‘sweet spot’ between what generates the most value for members and the organization? 
  • Using the 80/20 principle to double down on what’s most effective and remove the rest. 
  • Expanding the benefits the community offers to the organization (for example, to go beyond support and encompass ideas and advocacy)? 
  • Expanding the potential audience of the community to new demographics or categories of members. 
  • Being the indisputable leader in your industry for the topic (or becoming a cult brand) 

…and this isn’t even close to a comprehensive list. 

Developing the right strategy begins with recognising there’s an opportunity cost to everything you do. Time, budget, and reputational capital invested in one activity can’t be invested in another. Pursuing one strategy means you can’t pursue another. 

The only way to develop the right strategy is to first fully understand the problem(s) you’re going to solve. 

2. The Complete Community Discovery Process

Understand The Macro-Forces Which Shape Your Strategy

A community lives within a complex ecosystem that both supports and constrains what you can do.

The FeverBee Community Forces Model below shows you can only achieve the results which are within the constraints of the community experience you offer, the dynamics of the organization, and the broader macro-forces at work.

Your Results Depend Upon The Community Experience

Let’s begin with the inner layer. 

The results you can achieve are the direct result of the community experience you offer.

If the community experience is terrible, you won’t achieve good results.

If the community is poorly managed, offers a bad onboarding experience, doesn’t properly position itself well, is difficult to use, or runs an unsuccessful MVP program, it won’t achieve the desired results. 

Your Community Experience Is Constrained By Organization Dynamics 

The community experience you offer depends upon the dynamics of the organization.

The rate of growth impacts your ability to acquire members.

The technological constraints and capabilities impact the user experience you provide.

Risk tolerance affects how you position the community.

The budget and stakeholder support affects how many staff you can recruit etc…

The more an organisation supports and is willing to invest in the community, the better the experience will be. 

Organizational Dynamics Are Constrained By External Forces

The degree to which an organisation will support and invest in the community is constrained by external forces. 

This includes the popularity of the sector, laws and regulations, the state of the economy, changing audience demographics and desires etc…

As a simple rule of thumb, the better things are going the more an organisation will invest in a community. 

Layers Are Neither Static Nor One-Directional

Layers adapt to changes in surrounding layers. 

For example, if you start achieving better results, you will likely attract more members, making it easier to acquire more members and gain more stakeholder support.

Over time you might change your organisation’s risk tolerance by proving the community isn’t so scary after all.

Some impacts can skip layers. For example, if your entire audience is moving towards Reddit or YouTube to learn and share advice, and you’re developing a strategy which requires them to visit a community and share knowledge, that’s setting yourself up for an upward battle.

At the other extreme, successful organizations can also shape the macro environment. One of our clients, UiPath, plays an active role in shaping the entire robotic process automation industry.

The key point, however, is to take the right steps to do the following:

1) Evaluate each layer. What is the current status quo of each layer? Is it positive or negative towards you? What are the strategic implications of each layer for you?

2) Understand the trends of each layer. Are the results improving or deterioriating? Is the experience getting better or worse? Is the organization becoming more or less conducive to community success? Are the macro trends creating a more hostile or fertile environment for the community?

3) Understand the relationships between one layer and the next. What are major aspects of each layer which influence subsequent layers? How will the trends likely impact your work?

Once you have all this information, you can identify the key challenges you need to solve and develop a strategy to solve them.

Ultimately, your entire success depends upon the macro forces at work.

For example, stricter data privacy laws recently forced a client to migrate from one platform to another. This had profound repercussions for the experience, internal support, and much more.

Likewise, another client has a comparatively large community, but engagement is declining because the audience is increasingly turning to social media to engage and learn from one another. They can either watch engagement decline or make drastic changes.

A critical part of your success is aligning your strategy to the broader macro-forces in your industry.

If you find yourself on the wrong side of the macro-trends, your community is going to really struggle.

On the flip side, your community might be growing rapidly because you’re benefiting from prevailing trends (like the growth of the industry or customer base). Be mindful not to take the blame (or the credit) for things which are outside of your control.

Your community's success depends upon developing a strategy which navigates successfully through the environment.

Understanding The External Environment

If you’ve done any kind of business course, you’ve come across the trusty PEST Analysis.

As the acronym suggests, this covers the following factors

  • Political / Legal. Global forces, legislation changes etc…
  • Economy. Overall sector growth, costs, employment etc…
  • Social. Changing desires, demographic changes, new trends etc…
  • Technical. Product adoption rates, emerging technologies etc..

Like most frameworks, it’s only as good as the information you put in it. But it’s a good way to begin categorizing external forces (you can begin with this template if it helps).

Don’t use a generic analysis. Gather one that is specific to your industry.

For example, the threat of nuclear war might have increased recently, but that might not have a meaningful impact on your industry.

If you’re lucky, you might have a strategic planner in the organization who has done this already. You might be able to adapt to this. If not, you may have to do it yourself.

Below are the key things we would look out for:

Political Economical

Are there any new laws which will affect your industry? (e.g. sectors like banking, healthcare, etc…)

Are there any new laws which will affect technology? (e.g. data privacy/security)

Are there any new laws which will affect the management of the community? (e.g. section 230, Online Safety Bill etc…)

Are there any likely political scenarios which may affect how you engage the community

List the major organizational dynamics which are already being utilised as a strength (i.e. strong support, good budget, high-risk tolerance etc).

List the notable positive results of the community (or measurement of results)

Is your industry growing or shrinking in popularity?

Is your audience becoming richer or poorer?

Is the economy growing or shrinking?

Are costs of labor/platforms increasing or decreasing?

Is the industry mature or nascent?


Are the demographics of your sector/industry changing (influx/decline of key areas?)

Are the needs and desires of your audience(s) changing?

Are the habits of your audience(s) changing?

Are there any new social trends which will have an impact on your industry?

Are there any new technologies which are gaining popularity?

Are there any new technologies which may prove useful?

Is organic reach/acquisition becoming easier or harder?

Are there any noticeable trends in tools and technologies which members are using?

Some trends have no more of an impact than a stone skipping off a lake. Others will make ripples. And some will make everlasting waves. You have to figure out which is which.

Not every trend you identify is going to be a game-changer.

Some may have no more of an impact than a stone skipping off a lake. Others will make ripples. And some will make everlasting waves. You have to figure out which is which.

Try to spot the waves.

For each of these, I’d assign two scores.

  1. Positive/negative. On a scale of -5 to 5, estimate whether this is a negative or positive trend.
  2. Urgency. On a scale of 1 to 5, estimate the speed at which this trend will impact your community. This is subjective but use your best guess (or research). If the trend is quite far out, reduce the score to a 1. If it’s happening right now in a major way, list a 5.

Now you can multiply the metrics together to get an overall importance score.

You can see this in the table below (with the macro-trends listed from most urgently negative to most urgently positive).

You can skip the ones in the middle for now and focus on those with a high positive or negative score.

CategoryTrendPositive vs. negative
(-5 to +5)
Short, medium, or long-termImportance
PoliticalRising fines for organizations which fail to protect member data.-54-20
TechnologicalMaturing community platform market offering fewer options-32-6
EconomicalEconomic downturn placing pressure on organizations to reduce costs on non-essential spending.-32-6
PoliticalCountry-specific laws designed to protect members from harm (and end of safe harbour protections for hosting harmful content).-22-4
SocialRising popularity of Reddit, StackExchange, Discord, and FB groups as preferred channels of engaging communities.-22-4
TechnologicalDeclining organic reach-21-2
SocialSearch-driven approach to solving problems.313
EconomicalDownward costs of major community platforms.224
EconomicalIndustry is steadily growing.515
SocialRapid increase in customers from India and the Philippines.326
TechnologicalRise of new community intelligence tools336
TechnologicalGreater abilities to translate content into new languages248

In the example above, our macro analysis of the community might conclude there are three major negative macro trends to consider (protecting member data, maturing community platform market, and reducing costs) to accommodate and three major positive factors to consider (improved language translation, rise of community intelligence tools, new demographics).

Aside – it’s usually best to engage stakeholders in this process. They often have great knowledge about the broader industry.

Feel free to ask around in industry forums too and get the experiences of your peers (p.s. This is why it’s a great idea to participate in peer groups for your industry).

Audience Analysis

The better you understand the audience you want to engage in the community, the better your strategy will be. Technically the audience analysis falls within the ‘social’ factors of the PEST analysis. But it’s such a big topic we want to splinter it into its own section here.

Far too often, when research is undertaken it’s primitive at best.

The community manager might ask their members what they want, get a bunch of responses, and then begin working on those things.

There are two problems with this:

1) Members can’t explain what they really want. Members are far better at describing their problems than the solutions to those problems.

2) This research reflects only the needs of the top members. These are unsurprisingly the people most likely to respond to research requests.

To get useful information, we need to uncover the issues faced by both types of members.


Researching Community Participants

There are several great ways of researching your current community members. This includes:

1) Member interviews. These are scheduled one-to-one interviews with community members. These are usually hosted online and take around 30 minutes to complete (any longer and fewer people agree to them).

Resource: Common Member Interview Questions.

2) Member surveys. These are surveys which are sent to community participants. The critical thing is you segment your members by a criteria which is relevant (e.g. level of activity, tenure in the community, job roles, demographics etc..). This lets you review answers by member segments later.

Resource: Common Member Survey questions.

3) UX interviews. This is where you have member share their screen via a recorded Zoom call and set them tasks to do in the community and review the results. Every time you undertake this kind of research you will gather new insights. This will be critical for your user experience research later.

Resource: Rocket Surgery Made Easy

4) Behavioral Analytics. Finally, you can also use tools like Hotjar, Mixpanel, FullStory, Amplitude or many others to track what members are doing in your community. You can see which pages they visit, how they progress through them, and where they get stuck. You should really understand how the typical person progresses through your community. It’s probably not how you might think.

By the end of this research, you should have an incredible understanding of who your community audience is, how they progress through your community, the different member segments you have and what their goals are.

You can see this in this example:

Resource: Member Segmentation Template


Uncovering Real Motivations

It’s not easy to find the real needs and motivations of members.

You need to interview or survey the customer base (or broader audience). These are exactly the kinds of people who are less likely to respond. This means you might need to incentivise this research a little ($50 for their time usually does it).

This is also the best kind of research to do before you launch a new community.

The common mistake is to simply ask people what they want and then try to offer members precisely that. But this ignores that members might already have a place they go to satisfy that need. So we need to push deeper here. This means we usually want to know:

1) What are members’ needs and desires? To begin, try to understand what members do within the topic. What do they spend most of their time doing? What do they like and dislike? What are the challenges they face? What does their day to day look like? You can literally ask members to describe what they get up to and how they use products/services or engage with the topic.

Sometimes members can articulate needs, sometimes they can’t. Some challenges might be finding information, getting help and support, or more. Try to push members to be really specific. ‘Getting help’ is incredibly vague. What kind of help exactly do they need? When do they need that help? Where do they typically get stuck?

Don’t limit answers to just things you think a community can provide. Let members talk about their careers or anything related to their careers. Try to get a good breakdown of what members are dealing with at any particular time. This will help you create terrific personas later.

2) How serious is this need? A good question to ask here is ‘what happens if you can’t solve this problem?’. The reason we ask this question is to understand how important the problem is. For example, my ‘problem’ might be better connecting with my peers. But it’s not an urgent problem I’ll lose sleep about. However, if my laptop dies or I can’t load Zoom, I’m going to be rushing to fix it because it halts my entire work.

So try to determine on a scale of 1 to 5 how important this problem is to the individual. Perhaps also how frequently they have this need. This also helps you determine what kind of community to build.

3) How frequent is this need? Some needs are ever present. Others might arise occasionally. Most support-based communities, for example, are high priority but are only needed by members on a monthly or annual basis. However, the need for belonging and desire is ever-present – if not often as strong at any given time.

Frequency largely determines the retention rate of the community. The more frequent the need arises the more frequently members are likely to visit the community.

4) Where do they go to solve those challenges now? The next question is where we think about the ‘competition’. Where does the audience go to satisfy those needs today? It might seem like the ideal answer is ‘nowhere’ – but that often portrays an audience which doesn’t feel a need that strongly.

The purpose of this question is to determine what the competition is like. If people are already going to Google, Reddit, Facebook, customer support or influencers to satisfy those needs, what will the community offer which is any better?

5) Why do they prefer those channels to resolve those challenges? Failing to properly answer this question is a major reason why new communities fail. They simply don’t offer something that’s significantly better in the eyes of members than the approach they’re already taking to satisfy their needs.

For example, you might think a hosted community is better than a Facebook group – but you might find members simply can’t be bothered to visit a new place every day when they’re already on Facebook. Please, please, please, don’t underestimate how hard it is to change member behavior. Your community has to offer something that’s so much better than it’s a no-brainer for members.

This usually means much more convenient, exclusive, or emotionally impactful than any other channel.

6) Why don’t they go to a community? If you have an existing community, this is a good question to slip in at the end. Until this point, you really shouldn’t have talked about the community at all. You should only be speaking about the needs of your members. Believe it or not, this question is far less important than the others but it can throw-up some interesting answers such as flaws in the website, lack of awareness, or something.

Treat the answers here with caution. Often members can’t properly articulate why they would use one channel over another. The answers to the previous questions are far more important.

Once you begin this process, you should be able to develop an incredibly useful dataset of your audience.

Here’s a snippet:

Need / DesireLevel of desireHow frequent is this need?Where do you go?Why do you go there?Why not the community?
Solve product problemsSevereMonthlyI search on GoogleIt’s easiest place to get the best informationCommunity is harder to figure out. Search results aren’t as good.
Find peers in my industryModerateWeeklyLinkedInIt’s easier to search by job title.Haven’t seen this as an option.
Highlight problemsModerateEvery two to three monthsNowhereN/ADon’t see an option to do this within the community.
Get help to setup the softwareSevereToday (short-term)DocumentationIt’s the only place that has the setup guideCommunity information is scattered all over the place.
Find examples I can show my boss.MildMonthlyFollow industry veterans on TwitterPeople share the articles - it’s convenient.Top people don’t share articles in the community.

Now you can start to come up with really clear and precise problem statements which you need to solve. You can articulate this in statements like:

“Around 30% of the audience wants to connect with peers at their level to ask questions in a private place. At the moment they try to do this via LinkedIn because they can search for people by job titles.”

“75% of the audience gets stuck on a product challenge once a month and usually goes to Google to solve it. They go to Google because it’s easiest and they’re familiar with how it works.

Google typically surfaces a support article or a post from a random array of other channels – the quality of these is varied. It often takes a few clicks to find the right answer.”

Right here we’ve created two specific problem statements which let us predict the potential target audience, and how frequently they might visit, and it gives us some idea of how we can design our communities to be so much better.

This research is indispensable if you’re launching a new community. It will guide everything you do. This is also why you want to do the research before developing the platform.

Build Your Personas

Now you should have enough information to begin building out your member personas. Keep these practical.

You should know who your audience is, what they are looking for, what their goals are, who they want to connect with, what’s stopping them from achieving their goals and how to reach them.

Remember the general rule about audience research:

If you want more participation, research your members.

If you want more people, research your non-members.

Feel free to adapt our template to suit your unique situation below.

Resource: Member Persona Template

If you want more participation, research your members.
If you want more people, research your non-members.

3. Organizational Analysis

Uncover Organizational Dynamics

Not long ago I sat with a frustrated community manager who kept mentioning how ‘dumb’ and ‘stupid’ her organisation was for setting up a community in an especially challenging way.

Language like ‘dumb’ and ‘stupid’ isn’t very helpful in any situation – let alone community strategy. For starters, the organization was generating hundreds of millions of dollars a year in sales. That’s a strong sign it’s filled with highly capable individuals – not buffoons. Yet the setup of the community wasn’t conducive to success.

So what’s happening here?

The organisation wasn’t ‘dumb’ or ‘stupid’, it was simply operating under a unique set of constraints and priorities the community manager didn’t understand.

More specifically, it was constrained by limited experience with communities, a risk-averse culture, and more pressing priorities than community. The problem isn’t usually the organisation, it’s the failure to understand the organisation.

The problem isn't usually the organisation,
it's the failure to understand the organisation.

Deeply Understand The Organization Before Developing A Strategy

It’s very hard to develop an effective strategy unless you deeply understand the organization. You simply won’t know what is and isn’t possible. 

This isn’t something you can surface in a 30-minute discussion. It takes a lot of work engaging with various stakeholders to understand past decisions, priorities, risk tolerance and more.

For some of these areas, you have to use your own intuition to build a full picture. For others, you can ask a few quick questions to tease out the answers you want.


Strategic Priorities

Perhaps the first thing to understand is strategic priorities.

Forget about the community for a second and try to understand from each of the stakeholders you speak with what the overall objectives of the organization are and what are the objectives and priorities of each department you engage with.

Some questions here might include:

  • What is your role at [organisation]?
  • What are you trying to achieve this year?
  • What’s the key challenge to overcome?
  • What’s stopping you from achieving that goal today?
  • What metrics matter most to you?
  • What are your biggest concerns?

You should have an understanding of not only the organization’s strategic priorities but also why those are strategic priorities.

Try to understand the precise metrics or outcomes they care about in the short, medium, and long term. Understand how they know if they’ve achieved their goals and what metrics they’re accountable for.

You should be able to list the priorities not just for the organization but for each department.

It helps at this stage to build a stakeholder map which highlights both the goals and concerns of each person you speak with.

Risk Tolerance

It’s a good idea to understand the risk tolerance of the organization.

Some questions you want to answer include:

  • Does the organization exist within a highly regulated industry?
  • Does the organization have a strong fear of members saying something negative, misleading, or sharing confidential information?
  • Does the organization have reason to be concerned about unique security threats?
  • Can the organization test different ideas and initiatives? Or do things have to work the first time?

Don’t only try to answer the question, but try to get specific.

What are the exact types of posts or scenarios which would make people concerned?

Work with legal, public relations, marketing, customer support, and others to get a well-detailed list of their fears and how worried about risks they seem to be.

Generally speaking, the more risk-averse the organization is, the more challenging it is to build a community. More importantly, you need to proactively take steps to mitigate the risk.

Generally speaking, the more risk-averse the organization is, the more challenging it is to build a community. More importantly, you need to proactively take steps to mitigate the risk.

Technological Constraints and Capabilities

If you’ve been in this industry for a while, you’re probably had a situation where your organization had chosen a community platform which was less than ideal.

This is another one of those situations where it’s easy to dismiss the choice as ‘stupid’ rather than a natural consequence of the constraints and capabilities of the organization.

You want to have a strong idea of the organization’s technological constraints and capabilities. This includes answering questions such as:

  • What’s on the organization’s technological roadmap?
  • Does the community need to be deeply integrated or kept separate from the organization’s CRM (customer relationship management software) and website?
  • What are the infosec requirements for incorporating new technologies?
  • Does the organization have in-house capabilities to design and configure a platform?
  • Does the organization have in-house capabilities to implement a platform?
  • What is the procurement process like for purchasing a new technology?

Again, add or remove questions as you see fit.

The critical part of this process is understanding the technology possibilities and needs for whatever you plan to do.

Stakeholder Support / Alignment

The greater the depth and breadth of stakeholder support on a shared goal (or set of goals) the better shape the community will be in.

Before being able to identify the strategy, you should try four key questions:

  1. Do stakeholders understand what a community is?
  2. Do stakeholders have an agreed, shared, goal of the community?
  3. Do stakeholders know whether the goal is being achieved or not?
  4. Do stakeholders care about the goal?

Getting the answers to these questions isn’t usually too hard.

In your interviews with stakeholders you can simply ask:

  • How much do they know about the community?
  • What do they believe the goal of the community to be?
    What do they want the goal of the community to be?
  • How will they know if the goal is being achieved?
  • Has the goal been achieved so far?

Once you have the answers to these questions, you can compare them to gauge the depth and level of support.

Understanding the answers to these questions will play a major role in your strategy. You may decide, for example, your stakeholders need specific training and support.

Resource: 46 questions for uncovering community goals

Community Budget

Understanding community budgets differ greatly between new and existing communities.

Let’s begin with new communities.

In theory, people would approach building communities a little like how they approach buying a house. They would begin with an idea of what they can afford and look for a house within that range.

In practice, most organizations begin with only a limited idea of what a community will cost and gradually revise their estimates upwards as they receive new information. This makes it difficult to ascertain the budget directly.

You can often begin by finding out whether they have investigated any existing community platforms and staffing needs for a community.

Some questions here include:

  • How big is the community team?
  • What kind of platforms are you looking at?
  • How many people do you anticipate will participate?
  • How do resources for the community compare with other channels?

This often gives a good indication of what kind of budget they have in mind.

If they haven’t done either, you can begin outlining low, medium, and high-cost options with the associated benefits of each.

If they have an existing community, you can either find out directly what the budget is or simply multiply the number of staff by $100k to $155k (adjust for non-tech organizations) and use our community platform tool to estimate what package they’re on.

The key strategic implication here is you can’t expect a low-cost community option to jump to high-cost overnight. Plan accordingly.

Resource: Community Platform Comparison Tool

Community Team Capabilities.

Finally, it’s handy to review the experience of the community team. This can be a bit tricky.

Some people are quite defensive about being openly evaluated.

Some common questions here include:

  • What is your current role within the community team?
  • What kinds of roles did you have in the past?
  • What are the biggest challenges you face?
  • What’s stopping you from overcoming them?
  • What are some skills you wish you knew?

Another option is to use our community accelerator tool.

This will give an evaluation of each member of staff. However, this can be a challenge because people don’t like to honestly evaluate themselves.

Another approach is to understand the strengths and weaknesses of the team by measuring each of them against these benchmarks.

Here you’re trying to make sure the community team has the skill set to match the strategy you’re developing.

If you’re developing a strategy the community team doesn’t have the ability to execute, you’re going to struggle (and you will need to provide training).

Organizational Dynamics Summary

At this point you should be able to clearly evaluate the following:

  • Stakeholder alignment.
  • Whether the community is a priority.
  • Organization’s rate of growth.
  • Risk tolerance.
  • Community team capabilities.

Make sure to also add any other significant notes, strategic implications and an overall score of 1 to 5.

You can see an example of this below:

FactorScoreObservationStrategic Implications
Stakeholder Alignment4Majority of stakeholders agree community purpose is retention but are unclear how to measure it.

Product team feels the primary goal should be feedback.
Community metrics workshop to identify best metrics to track retention.

Also cover whether feedback can be included as a primary goal
Community Priority2Community is not a major priority amongst key stakeholders. It is given little priority in internal calls and key strategy documents.Internal workshop with key stakeholders on potential benefits of community.

Limited budget and support for major activities.
Budget2Community is not well resourced. Budget for 1 community manager and a mid-tier platform only.Select a platform on $50k or less budget. Community manager must focus on high-priority tasks only.

Build benchmarks showcasing how similar organizations budget for their community.
Risk Tolerance4High risk tolerance. Accepting both of criticism and the opportunity to learn as we go.Begin with a small pilot program to gather feedback.

Encourage members to speak openly and honest about problems they encounter (with feedback system to resolve them).
Community team capabilities2No community person in place at present. Expected to appoint a ‘community strategist’.Persuade they needs a ‘hands on’ community manager to get the community started.

Notice how when you do this research some of the solutions begin to naturally present themselves. Now we should be able to respond to the initial community manager who thought her organization was ‘stupid’.

They’re not stupid – they just have a certain set of constraints which we need to embrace.

4. Community Analysis

Community Management

Community management is a really broad term. But it essentially covers how the community team provides a great experience for members. The ways of understanding this vary from project to project, but the key things to look at are liveliness, organization, content, and events and activities.

1) Liveliness. This is a broad term but essentially relates to the level of participation within the community. Does the community feel like a lively place of fresh activity? Or does it feel like a ghost town? If you don’t have a strong, lively, community not much else matters.

The primary questions we try to answer here include:

    • Does the community attract organic, fresh, activity every day?
    • Do most posts receive a response within 24 hours?
    • Do 20% of accepted solutions receive an answer (if applicable?)
    • In a random selection of 10 posts, does the community team tag other members, welcome newcomers, solicit updates/responses, and engage in an empathetic manner (adapt as needed)?
    • Is there an overall sense of community amongst members?


2) Categorisation. This is a simple term related to the community’s organisation. We want to answer questions here:

    • Are post titles updated for search traffic when needed?
    • Are posts properly tagged and categorized within the community? (if tags are used?)
    • Are old posts frequently archived (or updated?)
    • Is spam quickly removed from the community?
    • Are there clear systems for escalating serious issues and resolving disputes?

3) Content. If the community hosts content (blog, knowledge base, documentation, other articles) we can evaluate this too. Key questions here might include:

    • Is the community integrated on the same platform as the knowledge base?
    • Does the community manager send a monthly/weekly newsletter featuring the best of the community?
    • Does the community team create popular knowledge articles or posts within the community?
    • Is the content written to achieve specific goals (SEO, user need/intent gaps, gaps in search experience etc…)?
    • Is content professionally designed and published?
    • Is content frequently shared and receives strong signals of response/endorsement to signify it’s achieving its goals?
    • Is there an easy-to-browse taxonomy of content?
    • Is content measured and evaluated – with further content/updates planned for most successful articles?


4) Events and Activities. If there are regular events and activities are part of the community then we can benchmark this too.

This often includes conferences, user group meetings, online webinars and other activities. Common questions here might include:

    • Do the events attract significant levels of attendees?
    • Is participation in the event high?
    • Is feedback gathered from the event and evaluated for future recommendations?
    • Do events have a special goal (share knowledge, educate, etc…?)
    • Is participation and attendance of events rising or falling?

At the end of this, we try to have clear benchmark scores to highlight where the community is today and what can be improved.

User Experience

The user experience covers everything which members experience from the moment they visit the interface of the community.

This covers aspects like navigation, taxonomy, aesthetics, liveliness, calls to action, SSO, and the mobile experience.

Your north star for this is usually convenience. Do members intuitively know what to do when they visit the community – and how does this compare with other experiences they might visit for the same goal?

There are two great things you can do here.


1) Benchmark the community against best practices.

Literally, go through the community experience yourself using our benchmarks and compare it to best practices.

Register to join it, post comments, search for information, and do the things you want members to do. You should be able to get a feel for what does and doesn’t work.

There is no definitive way of doing this, we typically aim to see how the community compares on navigation, aesthetics, liveliness, and calls to action, and how it shows on mobile.

You can see the typical set of questions we look to answer below.

1) Navigation. Navigation is usually the critical aspect of a community’s design. Key questions here include:

    • Is the community prominently linked to from the brand homepage?
    • Is the community linked to in the top navigation bar?
    • Is all the content within the community accessible within three clicks?
    • Are there sub-menu depth indicators? and/or breadcrumbs so people can find their way back?
    • Is community content/activities structured by popularity?
      Is the community structured by the user, product, intent, or sector? Does this taxonomy make sense?

2) Aesthetics. This covers how the community looks and feels. Typical questions to consider here include:

    • Is the content readable with clear contrast against the background?
    • Does aesthetics match the brand colours and look visually appealing?
      Is key content above the fold?
    • Is static copy kept to a minimum?

3) Liveliness. This covers whether the community feels like an active, lively, hub with good social density or not. Questions to consider here include: 

    • Does the homepage show what’s new and popular in the community?
    • Does the homepage show the latest discussions by category?
    • Does the community display liveliness statistics on the homepage?
    • Does the community feature a list of popular topics/questions above the fold?
    • Is static copy minimized in favour of dynamic/updated content?
    • Do visible questions appear to be getting responses?
    • Does the community appear to get fresh posts regularly from members?

4) Calls to action. This reviews whether it’s clear what members should be doing on each page. Questions to review here include:

    • Is the login/registration box clearly shown on the homepage and discussion pages?
    • Is the search box unmissable on the homepage?
    • Is the ‘ask a new question’ button unmissable on the homepage/discussion pages?
    • Is it clear where members can update their preferences?

5) Mobile. This section reviews how the community looks on mobile. This will be more important in some communities depending upon their level of traffic from mobile phones. 

    • Do navigation options collapse into a hamburger menu/icon on mobile?
    • Does the community fit well within a mobile framework?
    • Is text removed to necessitate a better browsing experience?
    • Are calls to action more prominently shown (i.e. full width?)
    • Does the mobile experience present fewer items with clear priority given to main areas?

This isn’t a comprehensive list of questions but provides a useful place to start considering how to rate each factor.

By the end, you should be able to give an overall benchmark to the community in each area.

2) UX Testing

In theory, you should do the UX testing at this stage. In practice, we do it earlier for the simple reason that it’s easier to schedule all of the audience research and analysis at once.

By the end of this process, you should have a very strong idea of the community experience you’re offering members, how members are experiencing it and some of the things you probably want to think about addressing if this is a priority in your strategy.

UX testing will give you a remarkable set of insights into how to improve the community and where members are struggling.

Acquisition and Onboarding

Onboarding covers the entire journey from the moment someone first hears about the community to become a top member of the community. We divide this into three stages.

1) Pre-registration. We begin by looking at how the community attracts new members in the first place. Key questions to answer here include:

    • Does the community have a fresh source of new members (incoming search visitors, incoming new customers, referrals from elsewhere)? Look at the data to see if the number of unique new visitors each month is increasing or decreasing.
    • Are non-members presented with a unique banner showcasing the primary benefit of the community and encouraging them to join?
      Is the registration box embedded within the homepage?
    • Is the registration button within the community unmissable to visitors?

2) Registration. Now we look at the process of registering the community. In most cases, this should take no more than a few seconds (but be mindful of some communities where it should take much longer). Common questions to answer here might include:

    • Does the registration form collect four or fewer items of data from members?
      Does the registration page offer single-sign-on (SSO)?
    • Are any error messages logical and easy to understand?
    • Does the confirmation email contain only a single link to complete registration?
    • Does the confirmation email arrive instantly?
      What % of members who begin the registration process complete it?

3) Post-registration. Theoretically, the post-registration process could cover everything that happens next. But we want to make it a little more practical here. So let’s cover the immediate next steps. Your questions here might include:

    • Does the registration process automatically popular a member’s profile based on the data provided?
    • Is context brought forward (i.e. are members taken back to the place where they left after registration?)
      Is there a clear place for newcomers to learn about the topic (not about the community!)
    • Do members receive a welcome email / direct messages from the community manager to get started?
      Are members presented with a different welcome banner after they register?
    • Do members receive a series of scheduled emails several days after registration?
    • Does the community avoid sending out multiple emails to members upon registration?
    • Is the onboarding process suitable for the type of community? I.e. if it’s support, are they taken to a place where they’re likely to get support? If it’s belonging, are they taken to connect with people they know?

You can use our onboarding resources to benchmark this and get a good overview of what is and isn’t working here.

By the end of this process, you should have a very strong idea of the community experience you’re offering members, how members are experiencing it and some of the things you probably want to think about addressing if this is a priority in your strategy.

Top Members’ Program

Most communities should have some form of program to support their best members. Sometimes this is an unofficial group the community team reaches out to on a regular basis, other times it’s a well-formalized and documented program. We usually try to answer questions like:

  • What percentage of questions are answered by top members?
  • Is the size of the program suitable for the size of the community?
  • Is the program clearly documented – i.e. is it clear how to join, what members will receive etc?
  • Are members rewarded with additional access, status, and unique benefits?
  • Do the superusers reflect the member personas?
  • Is there a designated person responsible for managing it?

By answering these questions you should get a clear sense of the level of success of the superuser program.

Positioning and Rewards

Positioning is one of the critical things to get right for a community to succeed.

Positioning is how your members think about your community. The positioning of your community will largely determine whether people visit the community or not.

Ultimately, there are five key questions we try to answer here.

  • Does the community have a unique positioning?
  • Is the positioning aligned with urgent member needs?
  • Is the community the best place to solve those needs?
  • Does the community deliver on its promise to solve those needs?
  • Does the community consistently communicate its unique value?

The typical mistake here is to position the community to compete with something better.

For example, why would people ask questions in a support community instead of filing a ticket? It takes about as much time to do both.

Or you might position your community as a place for people to hang out and discuss the topic – only to later realize they already have social media channels which are working just fine for that.

Your community must offer unique value which people can only get from your community. This doesn’t mean your community can’t offer support if you already have a support channel. But it does mean your community should offer value people can’t get from a support channel and communicate that value.

For example, is the community quicker to get responses? More convenient? More reliable? Etc?

Review the research from earlier and determine how members think about community value at the moment against other channels.

You can see an example here.

Your ideal is to change the axis to find a unique benefit only your community can satisfy. This benefit must be urgent to members and your community should be the best place for it.

Resource: Positioning Graph Templates

Resource: Fixing your positioning problems.

Aside, this is usually where what the organization wants clashes painfully with what members desire. Make sure you reference your personas and your research from earlier when deciding if the community benefit is aligned with member desires.

Compiling The Full Evaluation

Once you have this data, you should be able to put together a detailed analysis of the current community experience.

It’s worth remembering that this isn’t a judgment. Every experience is constrained by each organization’s unique characteristics.

Resource: The typical process of benchmarking a community here.

Resource: Our community benchmarks template.

Notice in the table above how easy it is for an organisation to see what is and isn’t working.

Anyone can glance at this and see they should prioritise some areas at the expense of others.

5. Community Results

Does The Community Have A Valid North Star?

This is where you look at how the community is being measured. 

Is the methodology valid?

Is it aligned with the priorities of the organisation and the needs of stakeholders?

We need to investigate and find the answers to these questions.


What Are The Desired Results Of The Community?

The most widely discussed metric for measuring success is community ROI. 

But the truth is very few organizations measure the ROI in purist terms (i.e. the % return for every dollar spent). Most of what passes for ROI in the community industry isn’t the ROI – it’s a proxy metric used to represent ROI. 

Call deflection, for example, doesn’t measure call deflection. It measures theoretical calls deflected.

Instead, most organisations are searching for a valid measure representing impact.

That impact can be a direct financial gain (like cost savings or customers buying more) or it can be something more subtle (like a change in member attitudes).

A Process To Determine ROI

There are a variety of things you can measure which represent success.

A simple means of finding the right north star for your community is to work with the organisation through the process below.

Even this isn’t definitive. It doesn’t include ideas generated, feedback, testing products and services in the community etc…

The start of any effort to measure the results of a community is to speak to stakeholders and figure out what they wanted the community to achieve (and see if they have any metrics in mind to determine if they’ve achieved it).

Try to push past the generic ‘lots of engagement’ answers to the outcomes of that engagement. If they do get lots of engagement, what will change?

You should already have this information as part of the stakeholder research process we covered earlier. If the organization isn’t sure of the community goals you might need to host a workshop to help define them (although this becomes part of the strategy process).

By the end of this process, you should be able to get a prioritized list of results the organization wants from the community. For example

  • Increased retention.
  • Improve customer satisfaction.
  • Gather useful feedback.

This sounds simple enough, but be warned – things are about to get messy!

Try to push past 'engagement' as the goal for the community. What are the outcomes of that engagement? What changes if the community is highly engaged?

Which Metrics Reflect Community Results?

Finding the right metric to represent any of the results above is typically far more complicated than you might imagine.

Let’s illustrate this with an example. Imagine you want to know if the community is getting customers to spend more money with your organization.

On the surface, this doesn’t sound too difficult, but the reality is it’s incredibly complex. For starters, you need to figure out the precise metric.


Identify the precise metric

What does ‘increased spending’ mean?

Does it mean spending before and after they joined the community?

What if they were a customer for 3 months before they joined a community and have now been a member for 9 months, how will you make the comparison?

What if they joined the community the day they bought the product? You’ll have no spending to compare it to!

This is why a simple term like ‘increased spending’ usually drills down to something far more precise like ‘a 3-month avg. spending of comparable paired member/non-member tiers before and after joining the community’.

This means you would find customers with comparable profiles at a certain date, look to see which did or didn’t join a community, and then compare the aftermath of that.

You’re probably starting to realize what a headache it is to get that data.


Every Metric Rests Upon Assumptions

Even if you could easily get that data, you have a bunch of tricky decisions to make.

For example, what does ‘joining the community mean’?

Would you include the people who joined but didn’t participate? Would you include those who participated once and left? What if people joined multiple times from different accounts after forgetting their passwords?

The real challenge at this stage isn’t just to show that behaviour has changed but to be able to state with confidence that the community caused the change (and not vice-versa).

One way to work through this process is to build a model of how the community achieves the desired outcome. 

Commumnity strategy I online Commumnity

Five Types Of Community Models Which Succeed

Building the Community Model


An important part of this process is also to understand how the community impacts behaviour.

For example, yes, people participate in a community and might stick around as a customer for longer. But why is that the case? What is happening that makes that happen?

Is it they’re getting better results from the product? Have a higher satisfaction score? Feeling a sense of community?

This is where having the attitude data helps. Through your interviews, surveys, and other activities you should be able to build a causal diagram about how community participation might impact behaviour.

At this stage, it’s very much just an estimate. You can validate it in the analysis step. But it helps to begin with what you think is happening because this will let you set very specific targets going forward.

For example, let’s imagine our model is as shown here.

how do communities build ROI

If we know what drives the results we want (advocacy, spending, and retention), and what are the antecedents of those mediating variables, we can set specific targets for this.

In our community OKRs, we might set targets for

  • No. articles created.
  • No. community MVPs.
  • No. interactions with staff members.
  • No. attendees of events.
  • No responses to questions in the community.
  • No. posts views.
  • No. questions asked within the first 90 days.
  • % of customers who join within 30 days of buying the product.

We might also adjust these for the moderating variables on the left (i.e. not every objective above will be relevant to every category of members we established previously).

Building The Community Dataset

Now we can set about acquiring the data we need to measure the results of the community.

A quick word of warning here. Traditional community platforms usually offer some version of community analytics. But these analytics are typically so rudimentary as to be entirely redundant. You will almost certainly have to build your dataset.

This may involve completing some data processing agreements and working internally to build the relationships we need to collect the data we want. Ideally, we can have direct access to the data to pull the information we need directly. Sometimes we can use the API, other times we will need to explore to get what you want.

Alas, even once you know the metric you want to measure, you might not be able to measure it.

What if you don’t have access to the customer records? Or what if you do but they’re not complete or poorly organized? Or what if they’re complete but the people who purchase your products (organizations) aren’t the ones who participate in the community (users)?

What if you don’t have a record of every purchase somebody made? How would you know how many purchases they made in a given period? What if you don’t know when they joined the community (or when they first participated etc…)?

These aren’t unusual problems, they crop up in the majority of community measurement projects we undertake. Based on the data we have available, we often have to go back and revise our initial metric for measuring success.

For a great dataset, we typically want to combine data from the following sources:

a) The community platform. We want the member-level engagement records. This should ideally show the following:

    • Date members joined.
    • When they made their first post.
    • How many discussions they’ve created.
    • How many discussions they’ve replied to.
    • Whether they’re an MVP.
    • How many events they’ve attended.
    • No. articles created.
    • Rank or level within the community.

In an ideal situation, you can run a query to pull every discussion a member has posted within a community. This lets you see the level of participation within given time periods – which is incredibly useful.

You may also use community intelligence tools like CommonRoom and Orbit to augment this data with their social media data.

b) The customer relationship management software (CRM). You almost certainly need access to the CRM software to be able to see how member behaviours have changed. Typically here we want data such as

    • Unique identifier (name, email etc..).
    • Organisation
    • Sector
    • Job Title
    • Country
    • Language
    • Source (of customer)
    • Date of first purchase
    • Customer tenure (days)
    • Produces purchased / spending
    • No. uses of products (logins, utilisations etc..)
    • Courses completed

This isn’t a definitive list, but it’s a good idea of the kind of data which lets you run really interesting analyses.

c) Attitudinal data. The final useful data point to have is attitudinal data. The reason for this is any impact you feel the community makes upon member behavior should show up in changes in member attitudes first. Typical attitude data might include:

    • CSAT/NPS score (upon arrival and a year later).
    • Brand perception survey (upon arrival and a year later).
    • Brand identification survey (upon arrival and a year later).
    • Sense of community index (upon arrival and a year later)

You will notice you might need to set up the measurement project now to gather results months, even years from now.


Cleaning and transforming the data

This is where you pull all of the data into a single dataset (ideally a csv file) and begin cleaning and transforming it for analysis.

Often this involves some measure of figuring out what to do with missing entries, typos, using consistent rules for describing countries, places, languages etc. Transforming the data also includes using existing variables to create new variables.

For example, you might want to know if someone joined the community before purchasing a product. This is very much possible with the right data. But you need to create a new variable from existing variables.

If you haven’t done this before, get help (either from ourselves or someone else).

Resource: The Dream Community Data Set

Developing The Community Dashboard and Setting Targets

The final step is to develop the community dashboard.

This should visualise the current results of the community and help you make forecasts for the future.

An example of a dashboard is shown here:

FeverBee community dashboard template

A good dashboard should show:

  1. Whether the community is achieving its goals (on a time scale).
  2. What the forecasts are for the future (note the dotted lines in the image above).
  3. Whether the key objectives are being achieved (and the forecasts for each).
  4. Whether the community is doing better or worse than in a previous time period.

Developing the full dashboard is a big topic, you can find more information in the resource below.

By this point, you should be able to set forecasts for the future, highlight what good and bad outcomes look like, and identify if the community is heading in the right direction.

Resource: Building A Community Dashboard

Make Sure You Have A North Star

Every community needs an agreed-upon north star to aim for. 

It’s not always easy to find this. You often have to work with stakeholders, undertake detailed data analytics, and build a model which you validate with data to find the right metric. 

However, the difficulty is also an indication of how valuable it is. By this point, you can specifically show whether the community is achieving its goals or not.

This is incredibly important for developing your strategy. A strategy for a community which is achieving its goals is very different from one that isn’t.

Summarizing The Discovery Process

Now we can summarise the entire discovery process is simply as possible.

There’s no shortage of ways to summarise the information you’ve collected, perhaps one of the most common is to use a traditional SWOT Analysis

SWOT stands for strengths, weaknesses, opportunities and threats. This is where you comb through all the information you’ve gathered so far and list them in a table like the one below.

In this table, strengths and weaknesses should represent things internal to the community (i.e the community analytics and results) which are present today. The opportunities and threats typically represent the macro-forces and organisational dynamics which will affect the community in the future. 

Strengths Weaknesses

List the major strengths of the community detailed in the community analysis.

List the major organizational dynamics which are already being utilised as a strength (i.e. strong support, good budget, high-risk tolerance etc).

List the notable positive results of the community (or measurement of results)

List the major problems identified in the community analysis.

List the structural weaknesses in the organisation which are already holding the community back.

List any issues identified in the community results (or how they are measured).


List the key macro-trends in your favour which might open opportunities.

List any changing audience demographics or needs which create opportunities.

List the organizational dynamics which could be changed to improve the community.

List the macro-trends in your favour which might make things more difficult.

List any changing audience trends which might cause problems in the future (i.e. changing habits)

List the changing organizational dynamics which might cause major problems.

Don’t try to cram everything in here. That’s not helpful to anyone.

But try to list the critical things in each area of the research you’ve undertaken.

We often try to summarise this stage in a simple paragraph problem statement. For example

“The [organisation] community has grown rapidly thanks to being a first-mover in [industry] and having a rapidly growing customer base.

The community is vibrant, highly active, and well-liked by members.

However, it faces rising competition and the customer base is levelling off.

The economic contraction resulted in a downsize of the community team. This in turn has resulted in the departure of key community MVPs.

Discussions are now taking longer to get responses (and they’re getting fewer of them). As a result, engagement is declining and members are frequently turning to other channels for support.

To prevail in the future, the community need to better engage MVPs with less resources, determine its unique positioning against the rising competition, and offer additional value to members to increase retention as growth slows.

Once you know the specific issues you’re trying to solve, you’ll find it’s a lot easier to build a strategy to solve them.


Present The Data To Colleagues

At this point, it’s a good idea to present the data to colleagues before continuing. There are plenty of good reasons why you do this. The most important is to get their support on the ‘problems’ you’re trying to solve before developing the strategy. It’s critical all the key stakeholders feel they have been consulted.

You might not always be able to find a common agreement on every issue, but every person should feel like they’ve had a fair chance to have their say.

Try to get something representing agreement from key stakeholders on the issues to solve before beginning work on the strategy.

Quick aside – this discovery process can take a longer time to complete. However, if you like you can identify issues as you progress through them and make changes as you go. We often find we can make quick-win optimizations as we progress through the process. You don’t need to wait for everything to be complete before improving the community.

6. Selecting The Right Strategy

The Right Strategy Is Rarely Obvious

You might be hoping that once you’ve collected all the right data, the right strategy will magically present itself.

So, let’s get something out the way quickly here. You can’t create strategy by following a decision tree. There isn’t a set list of questions you can ask and data you can use to be 100% certain you land upon the right strategy. There’s simply too much randomness and variables you can’t control for that.

The better way of thinking about it is to imagine you’re a coach of a sports team. The more information you have the environment, the opposing team, your own team, and knowledge, the more you’re likely to make the right judgment calls.

That’s what the information you’ve collected so far essentially is. It’s a collection of signals to help you make the right decisions and significantly improve the odds of success!

Get The Right Strategic Approach

Years ago I created a full breakdown of a strategic plan.

At the time, I thought this was a comprehensive and definitive resource. But I was wrong!

A template is only as good as the information you put into it. It’s far too common for people to become so excited about having the right way to organize their plans that they don’t evaluate if they have the right ideas in the first place.

People often wind up hurtling down the wrong path – despite having all the right information at their fingertips – because they don’t have the right framework to make the key judgements about their strategy. This is the problem I’m going to try and solve here.


The Environment vs. Results

It’s at this stage where you should really engage others with the data you’ve collected and ask the most critical question; what kind of strategy do we need here?

These are the two tools we use to determine what type of strategy we need.

  1. How stable is the environment? If you know there was a major change happening in your industry or your organization you would want to make sure your strategy was aligned to it wouldn’t you? There’s a big difference between a strategy designed for the current environment in a rapidly changing one. By environment, we’re talking about anything external to the community. This includes the relevant macro trends, stakeholders, and organizational dynamics which come into play. Changes can be good or bad – what we’re really looking for is the stability of the environment.
  2. How successful is the community? If your community was achieving great results already, your strategy wouldn’t be very different from one that isn’t achieving results. The tinkering you do with the latter is likely much greater than what you do with the former. By success, we mean is the community broadly achieving its goals or not? Are stakeholders generally happy with the community or not? If the community isn’t achieving its goals, you need to repair the problems. If it is, you can think about how you harness the most value from the community without changing too much.

The above questions lead to four broad strategic approaches (adapted from the terrific book, Your Strategy Needs A Strategy).

Stable Environment Changing Environment

Achieving Goals



Double down on competitive advantages.

Become 'Best in Class'
Expand goals


Change goals.

Change Platform

Change target audience

Not Achieving Goals



Promotion, onboarding, and positioning.

Change team.


Close or revamp the community.

Each of these is a category of strategy rather than a strategy itself.

Once you know the right category, you should be able to take the right approach to achieving your goals.

7. Optimising A Community

The Optimisation Framework

If the community is achieving good results in a relatively stable environment (i.e. without any major problematic trends or obvious opportunities to adapt to), then you usually want to focus on harnessing the maximum value from that community by optimizing what you’re doing.

By optimising, we mean investing your time and resources in the areas with the maximum possible impact.

We can use a similar framework here to the one we used above but instead of evaluating stability, we evaluate the importance to stakeholders (members, colleagues etc…)

Growing / High Engagement Declining / Low Engagement

Critical to Stakeholders



Not Critical To Stakeholders



What Initiatives Are Critical To Stakeholders?

Let’s dive a little deeper into each of these terms.

What initiatives are critical to stakeholders?

In your discussions with stakeholders (colleagues and members), you should have determined which aspects of the community are important to them and which aren’t.

Look at your survey results, interviews, and the outcome of your workshop (if you hosted them).

Make a list of things which are important to the audience and which aren’t.

For example, a common list of stakeholder priorities might include:

  • Members getting good, quick, answers to each other’s questions.
  • Having a strong MVP program.
  • Having a better community than competitors.
  • Generating sales from the community.
  • Easily finding information in the community.

We’re using the term initiative rather loosely. An initiative can mean anything that’s important to stakeholders about the community.

Use your judgment (informed by research) on what is and isn’t important.

Which Initiatives Are Rising Or Falling In Activity?

This is where we look at activity and success metrics to determine what is and isn’t working well.

Some typical things to look at here include:

  • Which content/categories/tags are rising or falling in popularity (use % from normalised baseline)?
  • What areas, features, or sections of the community are rising or falling in popularity?
  • Is there any major shift in locations people are coming from?
  • Is there a significant change in the device people use when they visit a community?
  • Are the type of users changing (more devs, admins etc..)?
  • Any specific changes between types of visitors?
  • Is the activity level of some superusers increasing while others decrease?

If the metric is stable (i.e. not significantly rising or falling over time), instead refer to whether it’s high activity or low activity instead.

Use your subjective judgment or benchmarking here if you have to. It’s not always possible to quantify the impact of a program (gamification for example) but use your best judgment on how frequently members engage with whatever the program is.

Based on these results, we might create a framework similar to the one below.

Growing / High Engagement Declining / Low Engagement

Critical to Stakeholders


Superusers 11, 14, and 21 (individuals)
Product 1 questions
Visitors from mobile devices


Superusers with a high frequency of poor responses
Superuser with low frequency of good response.
Product 2 questions

Not Critical To Stakeholders


Visitors from Philippines / India


Off-topic discussions

This is a simplified example, but it highlights where we might want to invest our time to have the biggest impact.


Investments Must Match Divestments

The critical thing here is to find balance. If you’re suggesting the community should spend more time on an initiative, you need to explain where that time is going to come from. This means you need to find an initiative which isn’t performing well and recommend ceasing that activity.

It’s easy to simply recommend the community team invest more time and resources in everything. It’s far more difficult (and more important) to figure out what to cut to have the resources to invest. Prioritisation is the critical outcome of this process. Don’t shirk the difficult decisions.

Be warned, you need to be ruthless about this. You can’t satisfy everyone. When you cut an activity used by just 2% of the community, that 2% will be very vocal about it.

How To Invest More In An Initiative

How you invest the additional time and resources will vary considerably by the initiative.

However, we will share some common approaches to this.

1) Use your ‘best in class’ benchmarks. The easiest thing to do is see which organisations with more time and resources do with their initiatives and then identify which parts you can incorporate into yours.

This is where comparing your community to the best in an industry is incredibly useful. Having a deep understanding of what other organisations do is critical.

2) Reverse engineer success. Reverse engineer what’s driving success and then invest more time in those areas.

For example, if the time to first response is important, what would it take to get the average time to response down from 3 hours to just 2 hours? You can determine which metrics best correspond to this and then invest more time in those areas (more virtual agents, MVPs, better FAQ etc…).

3) Collaborate to identify improvements. Collaborate to identify improvements. Evaluate each initiative individually and see which aspects of it can be improved with more time.

This is often a collaborative process of gaining feedback from members and stakeholders on a specific initiative and letting them identify improvements.

This is especially useful in situations where you might have people arriving from a particular location or using a particular device. You can simply spend time speaking with this audience to determine which improvements to make. This is where your UX research will also be useful to identify improvements.

Again, this isn’t a comprehensive list but it serves as a useful way of turning the strategic approach into specific tactics.

Your mileage with each of these options will vary, but take time to figure out the right approach for you.

How To Experiment With An Initiative

If an initiative is not succeeding but is important to stakeholders, we may wish to experiment with different ways to do it better. The simplest way of doing this is by adopting the repair approach we will cover in the next chapter.

Always begin with why the isn’t doing what it is supposed to do before jumping in with potential solutions. This will help narrow the range of possibilities and significantly increase the odds of you getting it right.

If you’re not sure where to begin, a good place to look at is the reach, change, and longevity of the initiative. These are three distinct questions

  • Did the initiative reach enough people?
  • Did it change the behaviour of the people it did reach?
  • Did the behaviour change last for a long time?

This won’t help you uncover every issue, but it’s a great place to begin. If the answer to any of these questions is no, you can usually zero in on a problem. You can also engage with members to find out precisely why something did or didn’t work.

Industry benchmarking can also help you here.

Look at how organisations similar to you have approached the same issue and see if they have done something unique or special which you can benefit from. You might also want to look at platforms similar to you too.

How To Maintain An Initiative

Maintenance is by far the easiest of the four approaches.

This is simply a case of continuing what you’re already doing. The only addition is to keep an eye on the metrics and monitor stakeholders to see if interest changes or if growth slows.

Most of the time, initiatives in this column are things which members like engaging with but don’t drive direct outcomes for the community.

However, it’s worth maintaining them because they keep members returning to the community.

The more members visit, the more likely they wil engage in behaviours which do benefit the community.

How To Divest An Initiative

In theory, divesting is easy. You simply stop doing something. In practice, divesting can cause negative externalities which should be addressed.

A few things to consider here:

1) Who will be impacted? Make a list of internal stakeholders and members who might be impacted by the change. For example, if we’re stopping announcements in the community due to low interest or engagement, we should list the stakeholders behind the announcements.

2) How can we limit the impact? Make a determination in how they will be impacted. How important is it to them and are there alternatives which might lessen the impact. For example, if we’re eliminating off-topic discussions, we might encourage members to create their own place for these kinds of interactions.

3) Communicate the impact in advance. As even beginners in the world of PR know, it’s a good idea to communicate changes in advance and the reason behind them. Be clear that the decision has been made, why the decision has been made, and how members can still achieve their goals if they are impacted.

4) Determine tech and UX implications. Decide the technology implication of any change and how the change might impact the user experience of the site. Will the user flow change or not? Be clear about what will be required and ensure resources are aligned to make changes at the desired time.

5) Stage the changes first if possible. If feasible, stage the changes on a testing site before implementing them into production. This isn’t always possible, but it often helps identify potential issues before they arise.

6) Evaluate impact over 3 months. Anticipate a negative reaction in the first month, but measure this over the next three months to see how members respond. Has sentiment or participation shifted in a clear direction since the changes?

Optimising Is A Good Problem To Have

Compared to the other strategic directions, optimisation is usually a good challenge to solve. You get to double down on the things which are already working instead of trying to spend your time repairing a major problem with the community.

You should be able to systematically review where you and your community team spend the bulk of your time and make specific improvements to get the best results.

8. Adapting Strategies

Making A Major Change

If the community is achieving good results, but the environment is unstable, then it’s time to change. A critical part of any strategy is to ensure the community is aligned with current trends. If it isn’t, it will always feel like you’re swimming against the tide.

To determine if you’re in an unstable environment, look at the analysis you’ve undertaken so far of the trends in the environment surrounding the community. This includes what’s happening in the organisation and what’s happening outside of the organisation.

We’ve already identified which are most important and urgent to adapt to. Now we can begin the process of adapting to them.

There are plenty of ways to adapt to the environment. However, these typically fall under one of three categories.

1) Changing The Community Goals

It’s common for the organisation’s strategy to change or for the community to fall under a new department.

When this happens, the goals of the community often shift too.

It’s critical in these situations to repeat the stakeholder engagement process (and potentially the workshop) to uncover the new community goals.

Once these new goals are established, they should once again be translated into member behaviours and new targets based on these behaviours should be adopted.

This will then filter into the decisions about what activities should take place in the community and what platform you will use.

2) Change The Platform

It’s common in many sectors for the habits of the audience to shift from using one channel to another.

This is especially problematic if your community is hosted on a channel which is falling out of favour with the audience.

In this case, you have to examine where the audience is shifting to and decide if you should create an outpost on those channels or move the community itself to this kind of channel.

There are two parts to this challenge.

1) Clarify your technology strategy. Is your goal to build a hosted platform over which you exercise full control?

Is it to build a platform which integrates with the organisation’s technology stack?

Or is it to use the platform most widely adopted by your audience? Each has pros and cons.

2) What are the right technologies within this strategy? Within the answer above, you have to select the right technologies for your situation.

Here you can use comparative tools, reviews, and 3rd-party analysis to find the right technologies for your experience.

This isn’t the only reason why you might change platforms.

Sometimes a platform move is prompted by internal needs – often related to a change in the budget or a need to comply with new legislation and internal policies.

If this is the case you need to clarify the requirements and then evaluate each community option against those requirements.

Finally, you might also consider if there are any new technologies you should integrate with your community technology stack.

There are plenty of new technologies emerging – but do they solve a problem which makes their incorporation worthwhile?

Resource: Which 3rd Party Platforms Should You Use?

Resource: Community Platform Comparisons

Resource: Community Platform Wars (YouTube)

3) Change The Audience

Perhaps the least common approach is to change the target audience (or, more likely, cater to a broader audience).

A common example of this is when localization. This is where an organisation wants to support audiences in their own locations and in their own language. In this case, you might either develop a unique instance of the community for that region or figure out how to translate community materials to accommodate and support this audience.

Alternatively, you might add new topics to the community.

Many communities have customers, developers, partners, and even leadership from organisations participating at different levels. This is a way of expanding the potential reach of the community. A community might create unique areas within the community with content, discussions, and activity for each of these groups.

Be mindful that it’s not always easy to simply incorporate a new audience.

Different audiences’ preferences (especially regarding technology) can be very different from one another. Often it’s not a case of just adding your audience to an existing technology, but adding an entirely new platform.

Proceed With Caution

Adapting strategies tend to get the most attention but are also the riskiest.

A failed community migration or constant change of goals can cause a huge amount of harm to the community.

These kinds of strategies should only be attempted when the trends which require them to be undertaken are clear and undeniable.

It’s tempting when you arrive at a new organisation or take the reigns of a new community to commit to a major change.

This is often the desire to leave one’s mark on the project. Resist the temptation unless the data supports it.

9. Repairing The Community

Diagnose The Problem Before Presenting Solutions

When something isn’t working, it’s your goal to find out why.

A common mistake is to dive into solutions without properly diagnosing the problem.

You can look at almost any community and recommend what they should do to resolve the issue.

However, you’re essentially sharing an opinion if you don’t have data.

The opinion might be good or bad, but it’s only an opinion. And opinions are a reflection of personal taste and experience.

We use a root cause analysis to identify why a problem has occurred.

A root cause analysis is a systematic process where we identify what factors caused the problem, analyse data to find the primary cause, and then develop the right solutions to prevent it from happening again.

Resource: Root Cause Analysis

Build The Framework To Diagnose Root Cause

There are two broad approaches to repairing a problem.

You can build a detailed framework of potential causes or you can work backwards until you find a clear issue.

We’ll tackle each of them in turn beginning with a root cause analysis.

Let’s imagine the number of posts in a community has declined by 30% over the past year. It’s easy to suggest a bunch of ideas which might improve this (host more events, gamification, or promote the community better!). But these are guesses, not solutions.

The first thing we do is structure the problem into two possible causes.

In this case, either members are posting less or fewer members are visiting the community.

Notice how already the solutions we take to solving these problems would be completely different.

Then we can break these challenges down further. If members are posting less, is this a decline in new discussions or a decline in replies?

If it’s a decline in new discussions, we can evaluate if discussions have become more difficult to create or if members have less need to create them (e.g. they’re getting their questions answered elsewhere).

Likewise, if the problem is a decline in replies, we can look at if the difficulty of questions has increased or members are simply less motivated to reply.

You can see what a full framework looks like below:

The key point step is to build the framework in advance and use data to guide you to the most precise problem possible.

If we look at the right-hand branch of this, we can see a decline in active members. But is this because fewer people are registering or current members aren’t visiting as frequently?

If it’s a decline in new registrations, is this because fewer newcomers are reaching the community or fewer of them are converting into members? And if it’s a decline in returning members, is this because newcomers are churning or visit frequency is declining?

Notice how each answer gives us a very specific problem to solve than just ‘posts has declined’.

You should also notice, using the template above, if you begin diving into solutions without first gathering the data to find the right solution, your chances of even trying to solve the right problem is about 1 in 8 (or 12.5%).

Aside, often the data won’t give you a definitive answer. Sometimes multiple metrics change at once or problems have multiple causes.

However, this simply means you need to implement two or more solutions to solve a problem – or you can prioritise the solutions which will have the biggest impact first.

Check The Preceding Behavior

If you don’t have the capacity to build a framework, you can simply examine what behaviours would have had to precede the metric you’re tracking.

It’s good to actually visit the community and track the behaviour to be sure about it.

For example, in one community the number of posts has declined.

The first question you might ask is what is the behavior which immediately precedes the one you’re tracking?

Well, they have to select whether they are publishing a reply or a new discussion.

So you might look at whether it’s the no. replies or the no. new discussions which have declined.

This alone will give you a ratio to see what’s happening.

For example, you might find that the number of replies has stayed consistent, but the number of new discussions declined.

Now you know the problem is people aren’t posting new discussions as often as they used to.

Next, we have to ask what people do before posting a new discussion.

The answer is they need to visit the community. So we might look at the ratio of new visits to new discussions to see if the decline is because fewer people are visiting or because when they do visit they are posting less frequently.

If we find that it’s the latter, we now know that the problem is ‘on the site’ so to speak.

We’re still getting the same number of visitors, but for some reason, they are not posting new discussions as often anymore.

Now we can put forward some hypotheses to test.

Are they filing support tickets instead of asking a question? For this, we can look at the relationship between the no. support tickets and the no. questions.

Are they finding the answer without having to ask the question? For this we might look at the number of views of questions against the number of new questions.

If we find the issue is the former, we can adjust the flow to make the community more prominent. If it’s the latter, we might be ok with that and instead focus on getting members to highlight if the answer they found helped.

This is a ‘quick and dirty’ approach. It’s not ideal, but it’s far better than guesswork. 

Prioritise Your Solutions

Once you zero in on potential problems, you can begin gathering qualitative data by undertaking UX interviews or interviews with members to see what’s happening and why.

This will help you identify the right solutions to the problem. Data can only take you so far, but qualitative insights from speaking with members will usually uncover the right kind of solutions to solve the problems identified in your root cause analysis.

There is unlikely to be a shortage of things you can do to repair the community. So make a prioritised list of solutions by their likely impact and the level of effort they require. You can determine the impact by how serious the problem is.

To understand the effort required, you have to have some idea of what each solution will involve. This means researching each solution in detail and estimating the time-frame, human resources, and budget required to implement it.

Aim to implement the low-hanging fruit first and then move on to the biggest things. Often you need to build up the momentum for bigger solutions.

10. Relaunching / Closing A Community

Taking The Tough Decisions

If the community isn’t successful and the environment isn’t stable, it’s time to completely rethink the nature of the community. This is one of the hardest things to do because of the sunk cost fallacy.

If an organisation has invested a significant amount of time and resources into a community, it’s difficult to accept it’s not succeeding. It’s easier to keep it going and hope things will turn around. That’s very unlikely to happen. Even if you could achieve your goals, the environment is changing and you would need to adapt.

The most common scenario for this is when a community doesn’t quite achieve the critical mass necessary to justify itself or when a formerly popular community is a shadow of its former self.

You broadly have two options here. You can close the community or you can revamp the concept and relaunch it.

1) Relaunching A Community

The entire point of relaunching a community is to not be beholden to past mistakes. Technical and social debt continues to grow if you try to keep what you’ve already built.

A fresh start lets you launch a community which is aligned with current trends from day one and truly reflects the needs and motivations of members.

You will still have members’ contact details so you can inform members when the community is relaunched, but you can now design a community experience from scratch which will blow them away.

However, if you’re going to relaunch a community, you need to go through the entire process.

Don’t try to have a big bang launch. Start with a small concept and grow rapidly from there. It’s good to be bold with your launch. You should genuinely be looking to see what’s on the cutting edge of your industry and rapidly growing in popularity.

Your new community approach might look very different from what you had previously. For example, instead of a forum-centred platform, you might have a community spread across a dozen of social media channels.

Instead of repeating the entire process here, you can find a step-by-step guide for launching a community in my Book, Build Your Community.

Closing The Community

Another common approach is to close the community.

There is no shame in this, often it’s the right decision. Like a business, no community is going to last forever. People and needs are often changing.

A great community one year can be a ghost town the next.

When you find yourself in a situation where the community is dying and trends are working against you, the best course of action is often to close the community down and recommend where people can go instead.

However, make sure you do this with honesty and integrity. Don’t cause unnecessary harm. Give people at least three months’ notice.

Be clear about how they can download their data and recommend where else they can go to engage with each other.

It’s sometimes possible to find a partner or a third-party site which would greatly welcome new members.

In rare cases, it might even be possible to find a third party willing to run the entire community without your involvement. If this is legally viable, it’s worth exploring.

Follow the same steps for removing a feature in the repair approach above.

Strong communication here can prevent some of your biggest fans from turning into your biggest critics.

At the closure date, we recommend setting the community to read only for up to a month before finally removing it.

Rarely Popular, Often Necessary

Closing a community is never a popular choice. But it’s often the right choice. Similar to how closing a business frees up capital for more productive endeavours, closing a community frees up members to engage elsewhere.

They might become prominent advocates of yours on more popular platforms or take other approaches to support and engage with you online. Naturally, encourage them as you can.

It’s common to think of closing a community as a failure. And in some cases, it is a failure. An idea didn’t work and the community didn’t reach critical mass. It’s hopefully a failure the organisation can learn from.

But this is not always the case. Remember on a long-enough timeline all communities will fail. The success of a community isn’t defined by its ability to defy fate indefinitely, but by what it achieved when it existed.

11. Prioritising Your Solutions

Prioritise and Implement The Next Steps

You might have noticed the research section of the process consumes the majority of the time we spend developing the strategy.

This reminds me of the (probably apocryphal) Einstein quote:

“If I had an hour to solve a problem I’d spend 55 minutes thinking about the problem and 5 minutes thinking about solutions.”

Our mission when developing a community strategy is to be crystal clear about the problems we’re trying to solve.

Once familiar with the problems we’re trying to solve, we can zero in on the right solutions. This sounds obvious, but it’s remarkable how rarely it happens in practice. This is why so many strategies are not successful.

It’s worth noting we’re using the term ‘problem’ rather loosely here. Seizing an opportunity or launching a new community is also a ‘problem’ to overcome.

A problem doesn’t necessarily have to mean a threat, it can simply mean any challenge.
The next step is to prioritise your solutions to overcoming the challenges presented by their severity and the level of effort they would require to resolve.

For example, if this is a ‘repair’ strategy, we might have identified and prioritised several key issues to resolve, as shown below.

Priority score
Doesn’t realise the login option is also the registration option.Easy tweak to login/register.459
The forgot password feature isn’t working.Needs a developer to review the process and resolve it.538
The community is overwhelming – people don’t know where to ask their questions.Need a revamp of the community homepage with separate instances for members depending upon profile data.426
Members can’t find the posts they’ve recently made.Need to show members the recent posts they’ve made on the homepage when they visit and in the member profile.246
Members can’t find the latest information about products.Need a separate signposted area on the homepage and on discussion pages.336
The registration process takes too much time to progress through.Need to reduce redundant pages and reduce the information we ask from members.325
Members are unable to connect with people like themselves.Need to create a member directory and solicit metadata from members.123

Note: Above we have inversed the effort score (so the lowest effort shows up as a 5, and the highest effort as a 1).

Now you can add these actions to your community roadmap or simply use them as an action plan and get started.

Build And Adapt This Structure 

You can do what many of our clients do with this process and build or add their own customisations to it.

You can use the same process with the entire customer support experience or simply integrate the results into broader platform or technology changes within the organisation.

The ultimate goal is to be completely aligned on the challenges you want to overcome, clear about the strategic approach you’re taking with your community, and then specific in the tactics you will execute to overcome these challenges.

Good luck!

Subscribe for regular insights