As a startup founder, you have a million of things to do and getting your analytics setup can be seen as a “we’ll do it later” type of task. This is especially true for non-technical founders who may not have the technical expertise to figure out how to tackle an analytics project.
You may even have investors or even your own team members who are asking for more data on how your product is doing. Getting an analytics foundation set up in your company is crucial to your success.
Ben Portfield, co-founder at Looker had this to say about why founders and startups put off getting their analytics handled
“Analytics is something that’s easy to put off. When you’re actively building a company and trying to figure out your value proposition, collecting and splicing data can seem non-critical or premature. But then, all of a sudden, you hit a point where things get complex, you need to understand your customers much better, and you have lots of unusable data because you captured it the wrong way — or didn’t capture it at all.” –Source
Through my work with venture-backed startups, I have learned what things founders should care about when it comes to analytics and what things they can ignore or delegate to someone else in their team.
Let’s explore the 5 things that you need to understand when it comes to setting up and using analytics data.
These Are the 5 Things You Should Know About Analytics If You’re a Startup Founder
Number 1: Your Key Metrics and How They Are Calculated
You need to know what your key metrics are. You can track a million of things at any given time so it will take some work to distill your business into a handful of core metrics.
For example, your key metrics might end up being CAC, LTV and CAC/LTV as you work to figure out the best customer segment for your business. It’s fine if these key metrics evolve as your product or company grows.
If you’re not sure what your key metrics should be, start by reading the resources listed below. They cover the most common metrics that venture capitalists tend to want to see and talk about.
- 16 Startup Metrics by Andreessen Horowitz
- SaaS Metrics 2.0 by David Skok (Matrix Partners)
- AARRR Startup Metrics by Dave Mclure (500 Startups)
The second thing to keep in mind is understanding how your key metrics are calculated. I constantly get questions like this from clients and readers.
“Can you please confirm precisely how Mixpanel is currently calculating the conversion rate? i.e. what exactly is (and is not) in the numerator and denominator? FYI, this is partly because we need to understand what is happening with our business. But also, it is the type of question I am starting to get from key stakeholders who want to understand the specific dynamics of our business.
Any time you look at a metric, you need to understand the calculation behind it. You will have to explain this calculation to your investors, team members and anyone who will be looking at your data.
I see companies use incredibly loose definitions of core metrics like retention to the point that it doesn’t mean anything. Don’t assume that “active user”, “CAC, “ARPU” and other common metrics mean what you think they do.
Number 2: Who Owns Your Data (Externally and Internally)?
Most of these tools let you export your data but that is tedious and painful. This is where vendor lock-in comes in.
Related: If you’re interested in properly setting up Mixpanel, then I recommend you check out a free video course that I created that will teach you everything you need to know about Mixpanel.
Ideally, you want all of your data to be in one place. This could mean a database (Amazon Redshift, Postgres, etc) that your company owns and maintains.
Tools like Segment.com make it easy to set this up so you can own your data while still sending it to tools like Mixpanel or Amplitude. I wrote an article on the pros and cons of using Segment.com as part of your analytics stack.
Related: If you’re not familiar with Segment.com, I recorded a 25 minute video that will give you a crash course in this tool. You will learn what Segment is and why your company should use this too alongside your analytics tools.
That was the external component. Internally, you also want someone on your team to own the data. This person is ideally someone in your development who takes responsibility for the integrity and accuracy of your analytics tracking.
Analytics tracking can easily break with new product releases which would make your analysis useless. This person is responsible for saying “we can’t release feature X because the analytics tracking isn’t ready”.
Number 3: What Data Does Your Team Need and In What Format?
Next, you need to know what data your team needs. This will change from team to team but you should have a general understanding of what data your product managers, UX, marketers, etc need.
You then need to figure out the best way to give this data to your team. It is pretty popular these days to set up all of your analytics using SQL tools like Mode Analytics, Looker or Chartio. This setup is simple and can scale beautifully.
The downside of this approach is that most non-technical people don’t know SQL. They wouldn’t be able to create their own reports and would have to rely on a developer to do it for them.
In the early stages of a company, this is a no go. You want your team to be actively exploring the analytics data which means you need to set up a more friendly too like Mixpanel or Amplitude.
Number 4: Is Your Data Accurate?
The fourth thing you should know is whether your data is accurate or not. This is important because if your data isn’t accurate, then you can’t trust it.
If you can’t trust it, then it is likely that your team won’t trust it. Even the perception of inaccuracy is a big problem and it can prevent your team from relying on your analytics data.
Your development team should be able to confirm that the data is accurate on a regular basis. You can also run your own analysis on basic numbers like signups to see if the numbers match what you see in the backend/admin portal of your product.
Number 5: Are You Making Progress on Your Key Metrics?
Finally, the last thing you should know is if you’re making progress against your key metrics. Once you set up a good analytics foundation, added a couple of core tools and gotten your team trained on how to use these tools, you can then look at the progress that is being made.
Let’s Also Look at What You Can Safely Ignore or Delegate When it Comes to Analytics
There are also a few things that you can safely ignore if you’re a startup founder.
Number 1: Implementation of Analytics Tools
However, you should know how long it takes to implement and maintain tools. This comes back to knowing who owns your data internally (number 2 above). Most founders underestimate how long it takes to properly implement a tool like Mixpanel.
Number 2: Learning SQL
SQL is a tricky one but for most startup founders, it is not worth the time to learn it. You can work with others to get reports generated in SQL and also implement tools that let you explore data without it.
Most founders tell me they want something like this:
“We want to see dashboards summarizing our important metrics: retention, total active users, user and onboarding funnel, etc. We also want to see our progress on these metrics over the last 7 days, last 30 days, last 60 days, etc”
These reports or dashboards can be set up by your development or marketing team. You don’t need to go in and write the queries yourself.
Number 3: All of Your Analytics Tools
Your company will most likely implement multiple tools to get access to the right data. You don’t know need to know each tool and instead, let your development bring them together into reports or dashboards.
However, you should know how to work at least one tool that lets you explore core product data. This is usually Mixpanel, Amplitude, etc.
You keep hearing these same tools over and over again and that’s because they can be incredibly powerful for getting a sense of what is happening inside your product. You can also learn how to use them relatively quickly which makes them a great fit for founders who don’t know SQL but still want to interact with their data directly.
When I work with founders, I’m focused on figuring out what they should focus on and what they can ignore. I’m hoping this article gave you a quick sense of how you should approach analytics inside your company.
If you’re a technical founder, you could probably care about more things, especially when it comes to implementation and using SQL. However, keep in mind that where your time is more valuable. Don’t get caught in implementation details just because you want to “stay up to date”.
Finally, don’t get into the habit of building too much. One of the hardest things for early stage startups is to actually use their analytics data. Obsess about getting your company to use your analytics data to make better decisions and then worry about tracking more..
P.S. If you’re interested in implementing Mixpanel, then you should check out a short series of 5 videos that I created to help you understand core Mixpanel concepts.
These are the concepts that if you get them wrong, you will have a hard time making use of your Mixpanel data. Worse of all, you will waste time implementing an analytics tool that doesn’t actually help you make better decisions
You can access these free videos by clicking the image below.