It’s Dumb To Not Track All Your Data, Here’s Why

Archana Madhavan

Instructional Designer

people reacted
4 -minute Read,

Posted on June 16, 2016

One of the most common mistakes people make regarding their analytics is that they simply don’t track enough.

According to Ben Porterfield, co-founder and VP Engineering at Looker, there’s no such thing as tracking too much data. In fact, one of the most common mistakes even “smart” people make regarding their analytics is that they simply don’t track enough, he said in his interview with First Round Review. Why should you want to track all your data in the first place?

Track more data to get more knowledge

When you’re first implementing your analytics (hopefully early on in the company’s life), you’ll have a set of goals and core success metrics that will inform what data (or events, in the case of event-based analytics) to track. But as your product grows, evolves, and acquires more users, you need the ability to dig into your data with enough granularity to actually learn something. It won’t be enough to just get high-level metrics. If you’re not tracking enough data, you won’t see how trends change over time or why some users behave one way versus another. You won’t understand user engagement. Ultimately, you just won’t have the knowledge to make smart product or marketing decisions. Porterfield puts it bluntly in his interview:

“If you don’t track enough you won’t learn enough.”

Track more data to answer future questions

When you’re at the early stages of building a product, it’s tempting to treat data analytics as a secondary concern. But if you start out tracking just the bare minimum, all of a sudden, when you get to a point where you really have to understand how your customers are engaging with your product, you just won’t have that data. Your product and marketing strategies are ongoing dialogues with your users, facilitated by analytics. Data that you consider as superfluous today could turn out to be incredibly significant in the future. That’s why it’s important to think about tracking the data you need for both your present and future needs.

“Not tracking enough stuff can end up being like a forest fire. It starts out with small tradeoffs but will grow to impact your entire business.” — Ben Porterfield, co-founder & VP Engineering, Looker (source)

Don’t Let These Excuses Stop You From Tracking What You Want

a) When tracking events, if you don’t have a sense of your overall business goals or the critical paths users can take through your app, you won’t know where to even start.

b) If you’re not mindful and thorough about naming things consistently across platforms and using language that can be understood by everyone in the company, people will become easily frustrated.

c) If you don’t have a clear, logical record of your event taxonomy in place, future analytics users will find it extremely overwhelming and difficult to get started. This record should also be easily accessible so that everyone in your organization can quickly look up what an event or property is.

d) If you don’t really know what you’re doing when you’re instrumenting your analytics, you’re setting yourself (and future analytics users) up for failure. It’s better to discuss with an analyst, growth engineer, or your platform’s customer success manager to make sure your setup is correct.

  • ****Tracking more events slows down query speed. Yes, it can, but that means your analytics solution wasn’t designed for scale and it may be time to look for a better solution. Some platforms, like Amplitude’s Nova architecture, are far better equipped to handle large volumes of data and complex behavioral queries than others.
  • ****Tracking more events is expensive. Yes, it can be, if your analytics provider is charging you by event volume alone. Most of the time, when you’re charged astronomically as your event volume grows with no added features or benefits, it means the platform simply cannot scale in a cost-effective way. You, the analytics user, are bearing the brunt of this limitation. It’s extremely critical to start out with an analytics solution that can grow with you as your user base grows and scales cost-effectively.

“People really need to get over their fear of having too much data. If you don’t use it — fine. But I guarantee you’re going to wish you had it down the line.” — Ben Porterfield, co-founder & VP Engineering, Looker (source)

An Important Caveat

This may seem counterintuitive, but we often strongly advise our new customers to beware of tracking too many events at the outset. Other things–your own time, resources, manpower–affect how effectively you can instrument your analytics, so it’s always best to start out tracking the data that will be of the most value to you. Starting out with instrumenting everything at the get-go could require energy and thought that perhaps you can’t afford to expend; not being meticulous about what you’re tracking could leave you with a mess of a data that’s impossible to decipher. When you’re just starting out, it’s important to take time to

  1. make sure you’re tracking everything that aligns with your business goals
  2. that your events and properties are organized, and, most importantly
  3. that you’ve instrumented correctly.
    A good analytics platform should give you the capacity to track whatever you feel like. You should never feel that you’re missing out on key insights because you’re unable to track something you wanted to. And at the same time, it’s worth taking time to think about how to track data in a way that lets you derive the most value from your analytics.

Archana Madhavan

Archana is an Instructional Designer on the Customer Education team at Amplitude. She develops educational content and courses to help Amplitude users better analyze their customer data to build better products.

More from Archana

Inside Amplitude