Getting Started with Analytics — A quick intro for Startups
Congrats, you have started up, you have this awesome new product which folks are adopting like hot pancakes. You have so much on your mind, Term Sheets, Marketing, Growth, Branding, Hiring, Burn Rate etc. Your team & you are working 100 hour weeks to get everything done. Depending on your past experience, most of the tasks may be new to you, you have no idea what CPC or PPC means, you probably hated Ads and branding was something to farm animals who had no choice. You might struggle to set priorities. Let me simplify your life a little, put Analytics First. If you don’t you won’t be able to measure if your efforts are helping you grow or just causing a growth placebo in your head.
You might say getting your sales team up to speed is critical for increasing your revenue, or getting to 1000 followers on twitter is crucial for marketing, or getting your next version of your product out the door is important to get your first customers. Let’s assume we put analytics second, how will you go about measuring sales, marketing or product’s effort? Will the next prospect, marketing campaign or product feature be not as important as the first one? Will you then, still place analytics first? Most startups fall in the cycle of ship quickly and usually miss out on putting in Analytics. After 6 months of effort, you are probably left with a bunch of questions and no way to get to answers.
Now that we have that sorted, let’s be clear, Analytics need not mean tracking everything in your business on day 1. You need to take it a step at a time and grow. In early days, you would need to see a higher level view of your business/product, focus on a few metrics and then start diving deeper. As you grow what your measure and how intricately your measure will evolve.
Here is a quick checklist of what you will need, or not need.
Step one: Google Analytics
The reason I state GA (Google Analytics) is because it is free & good for a start. That being said, just copying over GA code is not enough. This is going to get you minimal insights and most folks will never get back to the dashboard. Here is what you need to do:
- If your office has a static IP, make sure you filter out your office from the data.
- Set up daily reports, if you don’t know where to start, just set up a report for source/medium data. GA was plagued with spam traffic till recent history, which it now automatically removes, but this will still help separate good traffic from the bad. Will also give you an idea of which channel is working better
- Create a dashboard, again if you are not sure which reports to startup from, select the browser & country data, this will help you make future decisions, for e.g., if your early adopters are not using firefox ask your QA & dev teams to de-prioritise firefox. The geography information will help your sales & marketing team.
- Create Segments to filter data, it could be used to compare traffic from different countries or split the behaviour of people who converted with those who did not
- Create Goals to have your first set of KPIs
Step Two: Tracking Events
Now while step 1 gets your GA to a useful level, specially for websites, you need to start planning what else you need to track. For a start, know that tracking page views or screens is not fruitful, they are high-level indicators at best. Neither is time spent on a page or screen. They are vanity metrics which will give you more questions than answers. For a start, read up on the Pirate Metrics for Startups and build your user life cycle, this will give you a very high level plan. For e.g., if you have a beautiful sound recording app like Skyro you need to make sure you are tracking the following:
- App Opens — This will give you how many times folks are opening the app. This would be similar to counting sessions, depending on how you to define a sessions
- Record Start event — This is both an activation event and a core event, you need to make sure that people are doing this basic function
- Record Stop Event, you can add time recorded as meta data . This will let you segment the frequent users from the testers & drive by folks. This will also point to issue if the number of start events is not equal to stop events
- Keep a count of Total recordings — A quick way to segment your users
- Record an event of an attempt to purchase IAPs (In-app Purchase)— This is an equivalent to starting the purchase cycle
- Complete Purchase event. This is the total number of sales. Combined with the previous event it will help you narrow down the users who started a purchase but did not complete it. In addition you could discount purchases for your uses who are using the app for a long time but are not able to make the jump to an IAP as the price might be high. Most games use this mechanism to get in what ever revenue possible
Step Three: Engagement & Retention
Once you have some basic analytics you should plan for marketing & engagement which to start off can be emails & push notifications, but it depends on what kind of app you have and if you need any of those channels. Push would be the most critical, but don’t abuse the channel. Also send segmented and targeted messages rather than blasting the same message to all your users.
Step Four: Finding Insights
While one part of your analytics effort is to drive engagement & retention, you also need to derive insights to improve your product. See where users are dropping off, make sure your users are activated, search for fiction points in the app and fix all those issues.
This is not an in-depth guide, expect more articles soon with more details. If you have any questions email me at firstname.lastname@example.org, post a comment here or ping me on twitter (@ravivyas84)
If you liked the article do share the love via the 💟 button.