On too many projects the conversation around analytics goes something like this:
Marketer: As we're getting close to the 1.0 launch of our new app can you put in these 370 events that we want to track in Google Analytics?
Engineer: Sure I'll put them in. (Thinking at the same time that no one, ever will look at these events and thinks its a waste of time).
Done, that's the entire conversation.
A trite example as a set-up for the rest of this article? Sure. Mostly true a lot of times? Yep.
Having our roots deep in product we realize that product drives the organization -- and there are many different parts of the organization that are interested in how the product is performing. Each of these parts has a different function and requires different types of insight and analytics that they can act upon.
When we work with companies we typically three audiences:
- Product Managers
Each has their own set of unique needs and require different tools, solutions and communication. We'll take a look at the different audiences, the type of questions they might ask and the tool solution that can be used to answer those questions.
Engineers are most interested by understanding how the product is performing (and, more importantly, when its not!). Typical questions they may ask are:
- Are there certain APIs that are taking too long?
- Are there errors happening in my client side app?
We find that using tools like New Relic, custom solutions on top of Firebase/GA and SwiftyBeaver are useful in helping to answer these questions.
Product managers are motivated by understanding how the product is being used. They may ask questions like:
- Did the copy update in the last update result in higher % of people signing-up for my service?
- What does my Customer Lifecycle Funnel funnel look like? How do I optimize conversion along the funnel?
- What tasks are people most frequently trying to accomplish in my app?
Typically using tools that have strong support for event based funnel analysis (open and closed funnels), user segmentation and real time event insight are most helpful. Some solutions we have found that work well are Localytics, Mixpanel and Firebase.
Marketers are driven by figuring out how to acquire and re-engage valuable customers. Typical questions they may be trying to answer:
- What is my CAC for each of my advertising channels, campaigns and ad groups?
- How does each acquired customer translate into LTV?
- How do I effectively re-engage my existing customer base?
The marketing/growth area is perhaps the most complex. In fact, we dedicated an entire blog post to it. Check it out for some of the tool recommendations.
Right place, right time
These different audiences give us a framework in how we approach and think about analytics. Its important to figure out the right time in a products lifecycle to apply each of these milestones. As Khaled says, its 🔑.
For example, the 1.0 launch of product will most likely be concerned with helping answer the product management questions. After product/market fit has been established its time for the growth folks to step in and help pour gasoline on the fire.
In future posts we'll take a closer look at the things that we've learned when setting up analytics for these different use cases.