Finding The Ultimate Cause
A lot of our work diagnosing communities is finding the ultimate cause of problems.
You can’t really improve a community until you know the ultimate cause behind a problem.
To find the ultimate cause, you need to dive deeper into your data.
Let’s imagine a typical problem, your posts are declining.
To diagnose this you have to go through a series of specific, binary, questions.
Question 1: Why Are Posts Declining?
For example, imagine the number of posts in your community has dropped by 25% in the past 3 months.
There are two possible causes behind this you can investigate.
- Members are posting less when they visit (measure # posts/# visits has dropped).
- Fewer members are visiting the community (# visits has declined).
By looking at the number of visiting members and posts per visit, you can soon identify the answer.
Let’s assume it’s the second answer. This means it doesn’t really matter what you do within the community as fewer people will see it anyway. This guides you into your next question.
Question 2: Why are fewer members visiting the community?
For example, let’s assume fewer members are visiting. This too leads you into two unique options.
- Members are visiting less frequently (visits/active members).
- The community isn’t attracting as many newcomers as it used to (visits from newcomers).
By looking at first-time visitors to the community and visits per month you will soon get to the answer.
Let’s assume it’s the latter (few newcomers). This means you’ve still got the same level of visits from regular members, but fewer newcomers are arriving to replace the churn.
Any activity you take to resolve this problem should be focused on newcomers.
Question 3: Why is the Community attracting fewer newcomers?
Again, you can see this as a binary option based upon whether fewer newcomers are reaching the site in the first place or whether fewer are completing the registration process.
This would tell you whether it’s an acquisition problem or a registration problem.
- Fewer members are arriving via search (search traffic).
- Fewer members are completing the registration process (registrations per visit).
If we assume the answer is the former, we now know too fix the problem we need to focus on search traffic.
Question 4: Why is search traffic declining?
Search traffic is fiddly but you’re working within two obvious reasons why search traffic has dropped. Either you’re not ranking as highly or fewer people are clicking on the terms you do rank highly for.
- Community ranks lower in search terms.
- Fewer people are clicking the relevant search terms.
Notice at this stage you can’t start to think of really specific resolutions to these problems. If we assume it’s the latter again.
We now know the answer lies in something which has changed about the people rather than about our website or Google algorithms.
We need to figure out what.
Question 5: Why are fewer people clicking on our search results?
Now we can get to the final question. Has the terminology our audience used changed (relatively rare) or has the popularity and interest in the sector/topic begun to decline?
The community is optimised for the wrong search terms.
The popularity/interest of the sector is in overall decline.
We might research metrics that show interest in the sector to identify if this is the solution and accept or eliminate based upon the results.
If we again, finally, assume it’s the latter we can make a really specific problem diagnosis.
Our 25% decline in post quantity over the past 3 months has been the result of a declining interest in the topic which in turn caused a decline of search traffic to our website, fewer registrations, and fewer newcomers to replace the natural churn of members.
Once we have a clear diagnosis, it’s obvious that the typical approaches we might’ve tried (more on-site engagement activities) would only have resulted in a short-term boost at best. The real solution, as you can see in the decision tree below, must be to expand the community to cover a broader array of topics which are of interest to the same audience.
If you want to see the full decision tree, click here:
We can interview and research our members to identify additional needs and expand the community to accommodate those needs.
Trust me, never try to improve a metric without being crystal clear about what has caused it to go up or down.