Squeezing Every Last Drop From Google Analytics
How to get legit, actionable, money-saving insights from your regular old Google Analytics setup (no coding required).
a.k.a. How Google Analytics will get you to Aruba
So we’ve all heard it. You need to be tracking everything. Data is gold… or oil… or something like that… yada yada yada.
But guess what? Most people don’t have the bandwidth or the skills to do a fully custom event-tracking setup on their site.
Tag Manager? Most haven’t heard of it.
Setting up MixPanel? Good luck doing that without knowing how to code.
It’s okay, though.
A boring old run-of-the-mill Google Analytics setup is enough to start making WAY better decisions about your web presence.
So I’m going to give you a simple-yet-powerful case study. I’m hoping it’s enough that you can replicate something similar with your own data.
Case Study Time
I wrote this blurb for Quora to explain how I use GA to save me $$$ on development. Get ready, get set, let’s go!
Any time I’m deciding on adding or removing a new feature I look to Google Analytics to tell me if it will affect my bottom line. Could it possibly grow my user base? Will it alienate some of my existing customers? The great crystal ball that is Google Analytics has these answers.
So here’s how it works
I’m working with Joe Anonymous on his very well established site. It has hundreds of thousands of hits per month, great engagement, and lots of potential.
But, there’s a problem. His site isn’t mobile friendly (like, not even mobile functional). Essentially you’re not using his site on your phone, even if you want to.
Joe is deciding if he wants to drop $10,000 to make his whole site mobile friendly. Joe doesn’t want to spend ten grand on dev. He’d rather spend it on a vacation to Aruba.
Well, thank goodness Google Analytics is here to help.
So first things first, we ask GA how many people are attempting to visit the site on mobile.
The answer? Surprisingly few. Like less than 5%. This is unusual, but not surprising. Joe’s content is a variety that folks tend to use a desktop to view.
But we want to be extra sure we’re not missing something. So we can use the ever-insightful Google Analytics to learn about the types of users trying to visit the site on mobile.
We know almost all of Joe’s users that are spending $$$ are older (35+), we know where they are located (USA and Canada) and that they almost come in from one of 3 referrers (that’s 3 websites sending almost all the traffic that leads to a sale).
As it turns out, of the 5% of users hitting the site from mobile, very few fit the target profile. They tend to be younger, only 30% of them are in the US/Canada and almost none of them are coming in from those 3 referrer websites…
By our best estimate we’re looking at something like this:
Total Users (200K) x Percent from Mobile (5%) x Those Matching Our Target Profile (12%) x The Percentage of Our Target Profile That Buy (3%) = Total New Users that Buy per Month ( ≈36)
So if you didn’t follow that, we can roughly estimate ≈36 more sales per month if we make the site mobile friendly.
This is really oversimplified, as mobile-friendly sites get better search rankings, some sales come from other users, and many other factors come into play. But our best estimate( and given that search rankings aren’t a big factor for Joe’s site), we’re going to stick with ≈36 sales.
Those sales are worth about $5 each. So that’s $180 a month.
It’ll take a long time for us to make back our $10,000 in development costs at that rate.
So guess what? Joes going to Aruba, baby!
Essentially the formula is simple. Consider who is being affected by a change to your site. Use Google Analytics to filter for that segment. Learn everything you can about them. Then calculate your best estimate for how it affects the bottom line: is this gonna kick you in the KPIs or not?
Are we talking about statistically validated analysis here? Absolutely not. We’re talking about doing the best we can with the data we have. That’s a lot better than leaving that data sitting untouched in a database.
Need some support combing through your own Google Analytics setup? Click here and we can get you all fixed up.