What gets measured, gets managed.
This is a well-known quote from the management expert Peter Drucker. This is quite practical as it applies to virtually any business activity and signifies that if you start measuring what you are doing, you’ll get a grasp on the task. Then, you should get a good understanding of the current state of your action and where you should focus to reach the target.
This also applies to online communities, so the metrics associated with building community needs to be clearly laid out. That way there is clear accountability on the team working on the community and the effort is tied to the ultimate business goals.
Now, this is especially crucial as communities have become a common strategy with at least 66% of companies building different types of communities. In the US alone more than 56% of adults are accustomed to using online forums/communities for self-service according to Forrester.
In this article, we will focus on the customer-facing communities since the current trend has moved towards external communities. In 2019, the ratio of the external and internal community has reversed in comparison to 2018. According to Community Roundtable’s survey, there were 29% external communities (e.g., customer communities) and 53% internal communities (e.g., employee communities) in 2018.
In the subsequent sections, we’ll explore the way community managers can quantify the value of the member relations, measure the impact of user-generated content, show the overall ROI from the community and communicate the same to the leadership. This way the community management team can keep everyone in the loop, justify the investment and above all get internal buy-in from different teams. This can be very helpful in getting different teams involved and bolster the community strategy with the cross-functional contribution.
In general, all of the metrics for a customer community can be categorized into the following three sections:
- Effectiveness of the user-generated content
- The health of the community
- Impact of the community on the organization’s goals
Let’s now cover these categories individually.
Effectiveness of the user-generated content
Every community, big or small, generates content. This is known as user-generated content and it is important for the community managers to measure the performance of this content. Once you understand this, it would be easier for you to drive the content generation in the correct direction.
Here are some questions to understand the type of insights you can uncover:
- What type of posts (e.g., questions, discussions, ideas) are successful?
- What type of blogs/articles written by the members drive the best engagement (comments/visits/upvotes)?
- What types of polls are able to get the most number of responses?
- What types of events get the most number of turnouts?
If you observe these questions, you would figure out that they are all attached to an end goal (success, best engagement, number of responses, etc.). So, how do you figure out which post is successful and which one is not?
This entirely depends on the community type and only you can create the success criteria for the content generated in your community. So, the first step is to build a solid scoring system for your content.
As an example, a typical online community with customers as the main audience can have the following parameters:
- Shareability (Number of time the content was shared by members)
- Uniqueness (How unique is the content from the ones already present in your community?)
- Comprehensiveness (How comprehensive is the post?)
- Instant gratification (How quickly can the reader implement the takeaways?)
- Evergreen factor (Will the content be in demand 10 years down the line?)
You can edit these parameters and create new ones depending on the type of content you produce in your community.
Finally, you can give a score to the content based on a likert-scale (1-5 points):
|Very low (1)||Low (2)||Average (3)||High (4)||Very High (5)|
In this example, this particular content 15 as the total score. You can also follow advanced methods for scoring by using the weighted average.
Essentially, you would assign a weight to different parameters and multiply that with the point to calculate the average. For example, if in your case the instant gratification factor has more importance than comprehensiveness, then the former would get higher weight.
A community platform such as Tribe can help you export complete data set around your content and members with the click of a button. This screenshot given below shows how data associated with questions can be downloaded.
Note: The downloaded data set contains many additional data fields which are not shown on the tabular view in the web page.
So you need to download the data from the community platform you are using and rate the content on an ongoing basis.
Here is how it can look like:
|Question||A question on the recent government funding for athletes.||A controversial topic that generated lots of answers||10|
|Idea||An idea of a new workout plan for football players||Did not cause much spark||9|
|Discussion||Discussion on new coaching system in schools||Members are quite concerned about the new generation||12|
|Blog||A blog on 10 most extraordinary places to hike||This can a great evergreen piece for the adventure and fitness lovers||14|
|Question||A question on safety guidelines for cyclists||This one got a decent number of responses||11|
Once this is done, you should really look for patterns in user-generated content that consistently reach the target you have set. This will immensely help you repeat the success by steering content generation.
For example, if you see that questions posted around controversial news are getting a good result, you can boost this type of content by adding incentivization to content creation, i.e., by offering points to members to post and participate in this type of question.
Here we focus on the state of the community itself, i.e., exactly how well the community is performing. This can reveal valuable insights by answering questions such as the following:
- What’s the growth rate of the community?
- How sticky is the community?
- What’s the churn rate of the members?
- Which member cohort is the most active in the community?
- How is the active member cohort changing with the growth of the community?
- What’s the retention rate of active users over time?
Ideally, the community platform that you are using should have in-built reporting system along with integration with other popular analytics tools such as Google Analytics and MixPanel.
Google Analytics allows you to get an understanding of the community visitors and the bird’s-eye view of the actions performed in your community by the members. And, tools such as Amplitude allow you to measure granular metrics on member behavior by receiving member properties and their actions sent from the community solution.
Tribe Community Platform’s app store contains these types of deep integration with a wide range of third-party tools along with in-built apps to improve the value of the core community solution. These integrations take a few minutes to set up and completely automate data flow from the community to other popular tools.
Based on the data gathered from the community, given below are the metrics that you should be tracking:
Visitors per month
An external community focused on customers and prospects should expect steady growth in the number of visitors per month. You can easily track this metric with web analytics tools like Google Analytics.
Number of posts per month
Post creation in the community is an activity performed by the members. If your community platform sends the member activities to your preferred analytics tool, you can measure the number of posts created in your community.
In most of the customer communities, the community builder expects to grow the number of members. Hence, it is a good practice to measure member growth and ensure that your acquisition channels are working. Here is how the growth chart looks like:
The member growth rate would be calculated based on the following data points:
Number of members at the end of the selected time period = A
Members in the previous time period) = B
Members in the previous time period = C
So, the formula for calculation would be (A-B)/C.
Message origination channel
This metric helps you identify the source of the content that gets generated in your community. For example, your members might have the option to post on your community by directly logging in, replying to an email, Slack and Facebook Messenger integration.
Now you should measure the channel from which most of your members are interacting and posting content. Depending on the channel, you should optimize the important channels so that the engagement increases.
Simply put, text mining employs statistical and machine learning techniques to derive high-quality information from the text. Since communities generate text content in the form of questions, answers, comments, discussions, blogs, etc. there is a lot of insights to be uncovered. Given below are two use cases:
Sentiment analysis uses natural language processing (NLP) techniques to measure emotions (e.g., anger, happiness, excitement, etc.) and sentiment polarity (positive and negative) from data present in different formats like text, visual and audio.
Since your community generates a lot of text, it can be a good idea to run sentiment analysis on the content posted by customers. Especially, if you are running a insights community which helps you collect customer feedback, you would ideally want to increase positive sentiment score and decrease negative sentiment score.
Here is how it would look:
You can apply statistical techniques to find out frequently used terms in a specific topic, content type or group. You can also use sentiment analysis to identify clusters of negative terms and positive terms. Then, you can have more context by identifying the terms associated with the frequently used words.
For example, here is a word cloud based on the term frequency of words present in the questions posted in a community.
Now, we can further analyze to find out the other terms associated with ‘User’ which is the most frequently used word.
This will help in understanding what exactly is generating conversation or creating interest in your community. Laced with this information you can closely observe if there is an issue and any action needs to be taken from your end for a specific topic.
This metric is used to measure the number of members who stop using your community. Now it’s up to you to define churn. For example, churned users could be users who delete their account or those who stopped logging in (became inactive) in the last 3 months.
Based on the criteria you would need the following data points:
Members becoming inactive or deleting the account in the selected date range= A
Total members at the beginning of the selected date range = B
Churn rate would be A/B.
As a community builder, you want to increase the engagement rate. But, how do you measure the engagement? How can you find out the number of members who are engaged in your community?
In the case of a community (or any other website or app), a good way to measure engagement is to identify user engagement actions that result in engaged users. For example, your community could have the following engagement actions:
- Upvotes a post
- Asked a question
- Posted an answer
It really depends on how you define these set of activities for your community. Now, you can say that if a member performs these activities, then the member would be considered an active member.
Once this is defined, your analytics tool should allow you to measure stickiness, i.e., how often members engage with your community based on the active user criteria.
Here is what you need to calculate Stickiness:
DAU or Daily Active Users = The number of unique active users in each day of the month
Average DAU = Average DAU for the month
MAU or Monthly Active Users = Aggregate sum of daily active users over a period of one month
Stickiness would be the ratio of Average DAU and MAU.
Given below is a chart with DAU numbers and MAU along with stickiness.
Your goal is to increase the stickiness by optimizing the engagement criteria you have defined.
Retention rate is an important metrics as it tells how often your community members are returning to the community after a certain initial event. This can be used to unveil the reason why members are returning and why they are not returning.
Although there are different methods of calculating retention, N Day Retention is the most commonly used method. It literally computes the percentage of members who come back on a specific day. So, a day 7 retention measures the percentage of members who returned on the 7th day,
So, the initial or Day 0 trigger for retention measurement could be anything from the first visit to the first time your member followed someone. You can use this to play with different trigger events (joined a group, followed 5 members, completed profile, etc.) to see which activity results in higher retention and then guide new users to follow the path that results in higher retention.
Point to ponder: is there a reason why popular social networking sites ask you to follow a certain number of existing members when you create a new account?
Here is how a retention curve looks:
If there are 10,000 users on Day 0, then this means on Day 1 about 3% (300) of the users were retained and on Day 3 about 4% (400) of users were retained.
Ideally, you would want the new members of your community to get engaged and start contributing. This signifies that the members feel connected with the community and interested in adding value.
Here is how this can be calculated:
(Number of new members who became contributors/Number of new users)*100
Members at different lifecycle stage
Similar to new contributors discussed above, you need to define the stages of your community members based on their activities. Here are some of the lifecycle stages you might consider:
- Trusted member
Based on this, you should create a breakdown of the number of members present in each cohort. Your goal is to provide the best possible solution to the members in each stage so that they can move from one stage to stage. And, ultimately become the evangelist of your company.
Impact on the business goals
The traction that your content generates and the vitality of the community should ultimately impact the larger business goals. If you, as a community manager are not able to prove the positive result, then it is a very troublesome situation. However, based on the primary objective of the community there a finite set of business objectives with which a community can help.
So, what type of metrics should you choose to evaluate the community’s impact on these five business objectives?
Let’s explore each of these based on the associated metrics:
You should find out the number of new leads acquired via your customer community and how many of them ultimately became paying customers. You would want to measure the network of the existing members — how many new members are joining after being referred by the existing members and ultimately becoming your customers.
Here the important metrics are the number of support queries deflected, support cost saved, and customer satisfaction score.
Retention and revenue
You would ideally measure the average lifetime of the customers who are community members and those who are not members. If you are a marketplace company, you would be evaluating the number of transactions of members and non-members.
This can be easily done if your community is integrated with your CRM system. That way you can easily identify the contacts in CRM who are community members.
If you have a community dedicated to ideation, collecting feedback, and performing market research, you should measure the number of new features implemented in your product that originated from the community.
You can also measure the number of improvements done, issues fixed, and time to market (TTM) for the features.
If your community is used as a channel to establish thought leadership and propagate unique ideas, you should check for the number of media mentions and links to the concepts posted in the community. Another metric could be looking at the number of search queries (via Google Search Console) for your brand associated with new ideas you have shared.
We covered different types of metrics that you should be measuring — right from identifying the traction of the user-generated content, and vitality of the community to impact of the community on the organizational goals.
The metrics that you measure should be presented in a dashboard in a compelling format to invoke further ideation on the progress of the community. This should again lead to adjustments in your community strategy and result in better alignment with the business goals.
Also, the reports should be communicated across teams, so that you can easily get internal help for brainstorming as well as execution of tasks for any improvements.