Auto-tracking vs. Precision Tracking: What’s Right for You?
One of the main questions when companies decide to implement product analytics tracking on their website for the first time is, “Should we track everything?”
Throughout this post, when I use the terms track everything or track every user action, I’m primarily referring to auto-track tools like Heap and Freshpaint. These tools can track most user interactions such as page views and clicks without developer help. However, even when using an auto-track tool certain events always require developer assistance such as identify events.
The appeal of tracking every user action on an app or website makes sense. It provides you the opportunity to be thorough and uncover unexpected user pathways and journeys.
However, analyzing all the data associated with tracking every action once it’s in your product analytics reports can be overwhelming. What should you query first? How do you know that you aren’t missing any insights?
Being inundated with data can kill your curiosity to explore it. I’ve seen clients who have all the data in the world but don’t want to dig into it.
Before I jump to a definitive conclusion, I’ll help you weigh the pros and cons of tracking everything because the reality of what works for one company, can be very different for another. There are also hybrid approaches to tracking, which allow you to blend the benefits of auto-tracking with the robustness of precision tracking.
Pros of Auto-Tracking (or Broad Tracking)
- Automatically capture many helpful events and user properties.
- Gain access to retroactive data from the time of your tracking code’s installation.
- Allow your product, marketing, and data teams to define events instead of relying on developers.
- Uncover common user pathways that you weren’t expecting.
Cons of Auto-Tracking (or Broad Tracking)
- Relying on CSS selectors may break events if there are frequent code changes to your website.
- Interpreting event data is messy if no data governance strategy is implemented.
- Knowing how the events are triggered or where they occur on your website is difficult without proper documentation due to the number of events tracked.
- Needing development work to track additional events and user properties that aren’t captured automatically.
Pros of Strategic Tracking (or Precision Tracking)
- Hardcode events instead of relying on CSS selectors to provide more robust tracking.
- Streamline events with intentional tracking of user actions and properties to simplify analysis.
- Define your KPIs up front since a measurement framework is typically completed before implementing precision tracking.
Cons of Strategic Tracking (or Precision Tracking)
- Receiving event data only for events that are explicitly tracked.
- Losing out on event data that’s not available retroactively.
- Requiring development support to implement events, which can take weeks or months depending on available resources.
As you can see from the pros and cons, there are times when auto-tracking (or broad tracking) can be useful and other times when strategic (or precision) tracking makes more sense. What makes the most sense for your team ultimately depends on your available resources, what properties and events you want to track, the stage of your product, and the platform it was built on.
If your product is on the newer side, your company might not be sure what actions are most important to track. Your organization also may have no data or limited data history to use as a reference in what to track. Using auto-tracking or a broad tracking strategy in this case can be helpful because it means you can discover what features and pathways are important to your users.
If your team expects to make frequent changes to the website based on insights from your data, you may also want to consider auto-tracking. It provides a way to quickly update your tracking without the investment in development support.
Event Tracking Goals
If your team hopes to track events like login events, then you’ll need development help to implement them. In addition to tracking the login action, you’ll also want to have your developer implement an identify call. The purpose of an identify call is to associate known identifiers — like an email address or customer ID — with a user. This allows for a more robust analysis of your registered users.
However, if your website doesn’t have a login component, then auto-tracking is likely a good fit. Auto-tracking solutions assign a distinct user ID to every visitor who visits your website or app regardless of whether they log in. Also, they track certain user properties by default which can be used to segment users and perform analysis. With this data, you can get a strong understanding of individual user behavior across multiple sessions.
If your team doesn’t have development support, then auto-tracking is a helpful option. You can implement tracking using a tool like Heap, which has a drag-and-drop interface that allows you to set up events. If you do have development assistance, then precision tracking is on the table because you will need it to implement the tracking calls.
For certain websites, using auto-tracking might be more complicated like with a single page application. Since auto-tracking often relies on state changes, there are instances where it may not capture all changes on a page due to how a single-page application can render. In this situation, you will inevitably need development assistance to implement the push state changes so an auto-tracking tool can detect what’s occurring on the page.
Consider Precision Tracking (Yellow) vs. Auto-Tracking (Green)
Use the following table to help determine whether precision tracking or auto-tracking is better for your organization. Look at the cells that apply to you and count the number of each color.
|Event Tracking Goals
|No login events
|Developer support is available
|No developer support is available
|Single page application
|Non-single page application
Yellow = Consider precision tracking
Green = Consider auto-tracking
Did you get mostly greens? Consider auto-tracking.
Did you get a mixture of yellow and green? Consider blended tracking.
Did you get mostly yellows? Consider precision tracking.
What's a Blended Approach?
A blended approach involves using a combination of auto-tracking and precision tracking. For example, let’s say your company selected the tool Heap. Heap will automatically track page views, clicks, form submissions, and field changes. For these events, they will automatically track properties such as page title, page URL, and previous page, etc. If your team needs additional information about a user’s form submissions that is specific to your business (e.g., type of user) then you can leverage Heap’s add event properties API. Without the addition of this property, your team would have known the action occurred, but the contextual business details would have been missing.
Ultimately, it’s hard to definitively say what tracking solution (auto-tracking, blended tracking, or precision tracking) is the best overall since it’s highly dependent on your company’s needs. The factors I’ve outlined can help your organization discover what the optimal solution is for your goals. Also considering these factors can help your company avoid surprises and pain points during the implementation process. Velir recommends completing a proof of concept with a product analytics vendor as another way you can explore what tracking solution is the best for your company.
Want to get started with a proof of concept? We can help! Reach out and we can help your organization determine what the right tracking approach will suit your needs. Then we can implement it for you and help you make the most of the data it provides you.