Do you think your good intuition and maybe some user testing is all you’ll need to grow your product online? Or do you believe you know how to convert users given some of the strong progress you’ve already made?

Not so fast, you’re probably missing out on really understanding what’s driving use of your product or app. And then, your chances of scaling online? Very low.

The great thing about the web, is it’s fairly simple to get valuable insight and measure every step to acquire a new user. Let’s go through some basics with how to set up your funnel.

To start, check out the free option of Google Analytics. Paid options such as KISSmetrics and MixPanel (funnel image below) can be easier to set up, and allow you to track events not just page views.


Choose your Metrics
No matter what type of product you’re selling, there are two parts of the funnel that will always need to measure: (1) acquisition which is the broadest measure at the top of your funnel and (2) an end goal.

One useful measurement of customer acquisition to begin with is daily unique visits (UVs). It’s analogous to the number of people who visit a brick-and-mortar store. This broader measure is the number of people you’re going to have a chance to sell to each day.

The next step would be establishing a goal. This could be sales if you’re distributing a product. With an online sale, you can track how many people hit the completed order confirmation page. Starting from visits, you now know the percentage of users that view your site to purchase.

For engagement focused websites, your goal will likely vary but revolve around a core use of your product. Along these lines, Fred Wilson talks about the atomic unit of unit of a product. For example, on Twitter the atomic unit would be a Tweet, and something to measure could be Tweets sent per day, or number of users followed. For your product, think about what your goal is and then determine a metric to optimize for.

Implementing Tracking
At this point, we now we know how many people convert to a transaction from visiting your site, but not where those who don’t make it drop off. Knowing where potential users are leaking out of your funnel is critical in understanding where you need to focus to increase conversion rates to your end goal.

On an eCommerce website, a critical part of the funnel is tracking when a customer adds an item to their cart. Here, the funnel should track how many people convert from the cart page, to billing page, to payment page, to the review order page, to the confirmation page.

Using Google Analytics page (as shown above) each step should be built with a unique URL. Once implemented, you’ll now be able to see where any significant drop offs occur from the cart page to the billing page, and you’ll know where to focus your resources on improving conversions.

The process is similar when focused on a non-revenue metric. Returning to our Twitter example, acquisition tracking happens first upon a unique visit on the Twitter homepage.

If the core activity and goal initially is daily use, one metric to first track and optimize for could be “number of follows.” After all, users are more likely to use the product once they understand how the product works which comes through following other relevant users.

In the sign up process, after seeing the homepage, a new user is directed to a one-page signup with preselected options such as staying signed in and tailoring Twitter based on recent website visits.

Allowing Twitter to use website history from the browser helps the app tailor its recommendation of Twitter users to follow. After first landing on a “welcome/intro” page, the below onboarding process begins with providing suggestions for users to follow. Using Conrad Wadowski from GrowHack’s web history, Loius CK, Keven Hart, and Wall Street Journal start this process off, and help train the new potential user to follow others.



After following a few users, a new user is then directed to find other users given a from a few popular categories.

Once you track this onboarding process for the new user, you can begin to see where they fall out, and optimize for conversions. When you’re satisfied with the results, you can shift efforts towards other metrics which help define an active user, such as times logged in, or number of Tweets. You can then experiment with adding other steps into this initial onboarding process (and measuring along the way!) to better ensure a user is primed to be active in the long term, while keeping an eye out to make sure your onboarding is still working. Some additions in Twitter’s case for example could include adding a step to import contacts and create a new user profile.

This should be a start to get you thinking about converting customers in your own funnel. No matter what your goal is, it’s important to make sure each of these steps are defined and measured. This will allow you to best understand where the bottlenecks are, and begin hacking growth.

If you want to reach out, say something in the discussion or connect with me over Twitter @eric3000.

(Thanks to Conrad Wadowski from GrowHack for adding to this post!)