Reach out with questions, bugs, or suggestions. We'll respond ASAP. Your message is sent! We'll respond via email shortly.

Heap for Product Managers

As a product manager, your daily responsibilities vary. You may be in charge of maximizing product adoption, increasing engagement, or prioritizing product development. Product managers need to make important technical and business decisions, define and analyze success metrics, and integrate analysis into product requirements to enhance user satisfaction.

When the time to make these crucial decisions arises, Heap provides the tools to enable you to answer your own questions autonomously. The feedback loop from question to answer no longer exists; you don't have to wait for engineers to implement tracking code, for data to trickle in, or for a BI team to run queries on your behalf. It allows you to see correlations between actions, understand personas, and understand engagement based on user segments. It also gives you the flexibility to investigate user interaction with different parts of your product as questions come up - no tracking plan needed.

In this guide we provide detailed instructions covering how to determine actions that are important to your product, understand how different personas use your site or app, as well as a few other useful tips. After reading this you should be able to:

  1. Decide what engagement metrics are important
  2. Determine the correlation between an interaction and conversion or retention rates
  3. Understand the flow of users through your app or site
  4. Define and understand user personas and cohorts

With this information you will be able to:

  1. Define your product roadmap and support your decisions with data
  2. Decide whether a feature should be maintained or removed
  3. Fix usability bottlenecks
  4. Adapt your site/app to different types of users

Using Heap to define your roadmap

Heap allows you to see how a feature is used and how it influences other actions, enabling you to decide whether to prioritize a feature or nix it.

Check if the feature is being used

Heap's Event Visualizer allows you to see what percent of users are using a feature on a weekly basis with a simple click. No need to include all features on a tracking plan - with Heap's retroactivity you can decide to define an event on the spot and immediately get access to complete historical data for that event. To ensure you are correctly targeting an element check out our guide for defining events.

This gives you an overview on what actions are important to users within the product. If you see no one is using an element, you can safely make the decision to get rid of it and reduce the code base you have to maintain.

Who is using the feature?

A good way to look at an element's use is to graph it. You can view event counts and unique users over time by adjusting the date range, granularity, and toggling the unique check box. You can adjust your analysis to focus on any properties captured automatically or any custom event properties by either filtering for or grouping by the property name. For example, if you defined a Search event, and captured the search term using snapshots, you can view the most frequently searched words by:

Step 1: Graphing Search Event

Step 2: Grouping by search_term Step 3: Adjusting your view to a table

Once you build out a cohort based on behaviors, you can also analyze an event for specific subset of users by adding a filter for In Segment X. Explore this segment of users further by including a Group by In Segment X statement to your graph. This will allow you to compare actions of a particular group of users with the actions of others.

You can also analyze who a feature is important to based on user-level properties. For example, you can pass unique user-level properties such as plan type via our Identify API. Group by plan type in the graph to see how many of your highest-paying customers are interacting with an element. If your largest customers are using a feature, it might make sense to bump the priority.

Is performing event x correlated with the use of feature y?

Our Graphing Feature allows you to segment users into groups based on how many times they have performed an action or whether they have performed an action at all. Using group by has done event X, you can see whether or not an action is performed more based on the correlation with a separate event. For example, to determine if viewing an instructional video regarding a feature is correlated with an increase in the number of times the feature is used you would:

Step 1: Graph Ordered a Pizza

Step 2: Group by has done Watched Instructional Video

In this example, we can see that Watching the Instructional Video shows a strong correlation with an increase in performing the Ordered a Pizza event.

Similarly, if you want to determine the correlation between multiple sessions and an event you can group by count of instead of has done.

This will give you a breakdown of the number of sessions a user has had, along with event counts for users in each category. You can modify this query to contain count of Sessions rather than has done to measure the relationship between the number of times an event has been performed with, in this case Ordered a Pizza.

How do events compare?

Once you know what features are being used you can compare event counts to help you prioritize using a multigraph. Simply click on the plus sign next to the event to compare different interactions. You can add filters to a multigraph to look at a certain segment of users.

Promoting a feature

In order to convince yourself, your team, or your manager that a feature should have precedence, data always lends credible support to your decision. In addition to interaction data coming from raw event counts, you can use Heap to demonstrate the correlation between interacting with your feature and other key metrics such as conversion, upgrades, and retention.

Are people using a feature that would enhance their experience?

Using the graph view, you can get a count for the number of users who are using a feature that you expect to be popular. If you expect the count to be higher than the result, you have a couple decisions to make. Will doing x greatly improve product quality or enhance the user's experience? If so, do you need to make it high priority to improve the interaction so the feature is made more usable? Do you need to make the feature visible, or improve education around it?

Does the interaction correlate with higher rates of conversion?

Step 1: First make sure you have an event or event combo defined that captures the interaction with the feature in focus.

Step 2: Create a funnel that walks through key steps in your product such as a simple conversion flow for a sign up, upgrade, or checkout flow depending on your use case.

Step 3: Group by has done your_event (in this example Watched Onboarding Video).

This will show the the breakdown in conversion rate based on whether or not users have performed your event.

You can dive deeper by grouping by count of the event you're analyzing (for example, Watched Onboarding Video. This will show you how the number of times your event is performed relates to the conversion rate. Feel free to set filters to limit the results displayed.

Does the interaction correlate with a higher retention rate?

Step 1: First make sure you have an event or event combo defined that captures the interaction with the feature in focus.

Step 2: Create a retention report from either session to session or engagement action to engagement action.

Step 3: Group by has done your_event.

This will show the the breakdown in retention rate, or how often someone returns to do action X, based on whether or not users have performed your event. Similarly, you can dive deeper by grouping by count of your_event. This will show you how the number of times your_event is performed is related to the retention rate. Once again, you can limit the results displayed by adding filters.

Does the interaction reduce the amount of time it takes to perform another KPI?

Step 1: In the retention tool, select Session for first event, and your KPI for second event

Step 2: Group by has done your_event

Step 3: Select first time in the right hand corner

This will show you the days/weeks/months it takes a user to complete action 2 after they have completed action 1 depending on the granularity you select. This example shows the majority of users complete the second event within the same day as the first, but a small percentage complete the second event one to two days later.

Understand your user's 'Aha Moment'

You can also use Heap to determine that 'Aha Moment; for users - what actions indicate that a user will be a loyal adopter of your product. After determining which features have the largest effect on retention (find how to here), you will want to see how interacting with that feature influences retention. How many times does a user have to do x before your app becomes sticky? In order to assess the turning point where a user is hooked, use the retention tool.

Step 1: Use the retention tool to set event 1 and event 2 as session to session or engagement behavior to engagement behavior

Step 2: Group by count of Interaction with Key Feature

Looking at this retention table, we can predict that someone is more likely to be an avid user of the product after ordering a burrito 12 times. Once they hit 12 the retention bumps to somewhere over 50 percent retention. Once this is identified, you can focus your efforts on getting users to the point through marketing campaigns, changes in product education, and changes in UI.

Fix usability bottlenecks

Heap's flexible funnels allow you to do ad-hoc funnel analysis of expected user flow. Once the area of falloff is identified, you can iterate on your funnel to take a deeper dive into problem areas and learn what users are doing instead.

For example, say there is a 3 page onboarding process. The first step in understanding how a user migrates through this flow would be a simple three step funnel to give you a general overview.

In this example we see that there is a drastic drop off between steps 2 and 3, revealing most users aren't getting onboarded properly. To investigate, you can quickly adapt the funnel to contain the important steps getting to step 3, which may be filling in a form or viewing an instructional video.

From here, you can continue to alter your funnel on the fly, defining new events if necessary to pinpoint where users are falling off. You can also iterate through several possibilities for step 3 to determine where users are going instead of performing the expected action.

Segment users to understand personas

Using Heap's Identify API, you can pass any user level properties to Heap including company name, user type, or other identifying information. Additionally, many properties can be attached to anonymous users via the Identify API such as A/B test values, input values from unsubmitted forms, and last touch properties (check out our Identify guide). Using Segments you can create a subset of users who match a certain profile using both user level properties and behavioral metrics. You can filter users in two separate ways. It is easy to filter on property values using equals, does not equal, contains, does not contain, and wildcard matches (the wildcard is an asterisk). You can also filter behavior by adding filters that include has done, has not done, and count of actions is <, =, > x. As long as the date range is at any time, you can then analyze any graph, funnel, or retention view for this segment by including a filter for In Segment x.

Using time bounded segments to monitor active users

Heap allows you to create flexible and more accurate definitions for active users. You can combine behavioral filters to create the true image of an active user (a session doesn't necessarily mean a user is engaged)! You can also include a time range in your definition - does active mean past week, past month, all time? Once you have defined your segment you can analyze how the number of active users (or any time bounded segment) changes over time. Simply graph the segment or select analyze this in graph view.

This graph will show you the number of active users as time passes. You can see if the release of a new feature or a feature enhancement causes a spike in engagement, and investigate when you see a decrease. This will also show trends in adoption - and brings up another set of questions. Is your audience growing? Are retention rates consistent? Did something happen with another team causing a change in activity?

Exploratory analysis in the List View

The list view provides a great qualitative look into users’ activity on your site or app. It can provide you with great insights on what might be a key event, what actions users perform, and what actions users have not done. To investigate how a certain type of users interacts with your site or app you can filter the listview based on both behavioral criteria and any user level properties. If you have already created Segments you can also add a filter for users In Segment x. Overall, the list view is a great place to come up with hypotheses for further analysis. For more information on how product-focused customers have used our list view, check out what theSkimm did in our customer stories!

Other Important Docs

Snapshots: Adding more context (such as copy) to events

Heap automatically captures the Target Text of buttons and links in our Visualizer. If you want to attach metadata from the page, use Heap’s snapshots feature. Without code, you can use the Event Visualizer to capture copy, form field values, or any other information stored on the page and attach it as an event-level property. This is a great way to implement A/B testing based on copy.

Defining Events Guide

Use a guide to properly define events, and check previously set definitions. This is a great read for anyone new to the team or anyone who wants to ensure their accuracy!

Implementing Identify: Use cases

If you or your dev team is implementing the Identify API, here is a great read on best practices and use cases. This includes a section on best practices with integrating Optimizely, that applies to all A/B testing.

Heap for Marketers

This guide is a great way for the marketing to get started with Heap. It covers topics such as conversion funnels, attribution channels, segmenting users, and understanding Heap's data model in order to help make important decisions about campaigns, advertising spend, and targeting user profiles.

More?

Have any more cool use cases you’d like to share or use Heap to solve? Contact us here at support@heapanalytics.com or if you’d like to give Heap a test drive, start a free trial today or email us at sales@heapanalytics.com.