You’re sitting there, chugging through email like a boss, and you suddenly think of a fascinating question about your business that you’d love to answer. Maybe it’s something like one of these:
- Which features get used most often by different customer types?
- How do cancellations compare among different traffic sources?
- What behaviors separate your power users from your normal users?
The thought of answering the question makes you giggle with delight.
If you knew the answer, you could make your product much better. Your customers would get just as giddy as you are now.
But before you can toss rainbows of cupcakes to and fro, despair sets in. Why? You’ve just remembered how much WORK it’ll take to get an answer like that.
You see, most companies keep all their customer information in a database of some kind. This means that you’ll need to build all sorts of fancy SQL queries to pull that data. Yup, you’ll need to build a bunch of code just to pull the information you need to answer your question.
And since answers tend to lead to more questions, you’ll need to repeat this process OVER and OVER again.
So before you know it, you’ve spent the entire day mucking around with database queries instead of building a better product.
And if you don’t know how to manhandle a MySQL database, this might not even be an option. You’ll have to wade through all sorts of bureaucracy to get enough engineering time just to run a single query. If you have a small business, you’ll need to sacrifice time your engineers could spend improving your product and fixing bugs.
Now your rainbows of cupcakes have turned into pits of despair. You bail on your idea and go back to the endless flood of email.
In which case, all your customer data sits abandoned in a corner. Not exactly the best way to build a business that your customers will love.
You know, there is a better way…
It doesn’t involve code either. You can pull as much data as you want, whenever you want. As long as you’ve been collecting the data, you can easily see what’s going on.
Say WHAT? Impossible!
Ladies and gentlemen, allow me to introduce:
The Kissmetrics Power Report
I could start using all sorts of superlatives (amazing! stupendous! unbelievable!) to describe how awesome the Power Report is…
…or I could show you with an example.
Let’s say you have a website that makes money from a subscription of some kind, and you bill your customers every month.
You’ve been running an A/B test on your signup flow for the last couple of weeks. You know that different signup processes can encourage different types of people to become your customers. Some groups are more profitable than others so you want to make sure that you’re grabbing the right kinds of people. On top of all this, you’ve been running quite a few ad campaigns and want to see the impact the A/B test has on your campaigns.
The Power Report is the perfect tool for going all Gordon Ramsay (celebrity chef) on your data, slicing and dicing to your heart’s content (but try not to throw a chicken breast at your monitor).
Picking Metrics (and Columns)
How would we set up the Power Report and get our data? Take a look at this lovely screenshot:
This is the first part of the Power Report that we need to configure (there’s one more that we’ll get to in a moment). These are the metrics which go into the columns of the report.
For this example, we’ve built two columns to tell us how much revenue has come from two different types of people:
“did Ad campaign hit at least 1x”: This includes a list of everyone that’s gone through one of our campaigns at least once (ex: everyone that has clicked on a campaign link with a utm parameter…basically any clicks from Adwords or display ads).
“did Ad campaign hit never”: And this includes the people that have NOT gone through one of our campaigns (organic traffic).
We included the “did account canceled never” in both columns to focus on our current customers.
By comparing these two metrics, we’ll know what kind of impact the A/B test will make on our ad buys.
You can add more columns, pick any metric you want, change the dates you want data for, and get crazy detailed with what groups of people you want to include.
Picking Segments (and Rows)
Now we need to start slicing and dicing by picking our segments (the rows):
In the Power Report, each segment gets nestled into the segment above it. This allows us to drill down into our data and see what’s REALLY going on.
In this case, we’re going to look at our A/B tests (segment number one).
For segment number two, we’ve picked referrers. This will show us which referrers are driving traffic and revenue through each A/B test.
To get even MORE details, the customer ID is segment number three. If we see a referrer that has a surprising amount of revenue, we can drill down and see which customers are laying down the cash.
What Does the Power Report Tell Us?
Here’s the data we get from the Power Report:
When looking at the Power Report, remember that dark blue is the top segment (A/B tests in this case); the light blue shows which referrers are contributing revenue within each A/B test; and the white shows which customers are coming from that referrer.
Here are the results from the A/B test:
- Version A: $2,255 from campaign traffic and $8,255 from non-campaign traffic.
- Version B: $874 from campaign traffic and $17,022 from non-campaign traffic.
Woah! Version B may have tanked our campaign revenue, but look at the non-campaign traffic! Our organic traffic must love the new signup flow. Time to make that version permanent!
But before we jump to conclusions, let’s dive a little deeper…
In the screenshot above, we’ve broken down the “Signup Flow Experiment B” to get a closer look at the referrers. We’re trying to get a better idea of where that $17,022 is coming from.
Right away, we notice that $12, 886 came from Forbes.com. Hmmmmm… That’s an awful lot of revenue from a single traffic source. Especially when we compare it to our other traffic sources.
At this point, we’d really like to know if Forbes.com drives this kind of revenue consistently or if this is an anomaly. To figure this out, we just need to take a quick peek at our customers and get a sense of how the revenue is distributed among them. If we’re getting a number of high-value customers, we’ll know that we need to focus plenty of energy on driving traffic from Forbes.com.
So what have we got?
It looks like Mark@Facebook.com threw our metrics all out of whack when he became a $12,000 customer (he’s a high-roller). Not only did he contribute 93% of the revenue from Forbes.com, he also drove 70% of the revenue from Version B of the signup flow. If we remove this one outlier, Version B no longer looks as attractive.
It’s awesome that this business generated so much revenue from a single person, but the likelihood that they’ll continue to acquire customers at the same value from the same traffic source and with the same signup flow is highly unlikely.
With other analytics tools, we never would have been able to get so much detail about where our revenue was coming from. Nor would we have been able to deeply understand how each A/B test was performing.
What if this business was itching to drop its campaigns and focus on organic traffic? While Version B didn’t do quite as well as Version A, it appeared to do a much better job with non-campaign traffic. If we hadn’t found this outlier, the decision makers might have picked the wrong A/B test, which would have drastically slowed down revenue in the following months.
Ditch the SQL queries and having to beg your engineers to pull data for you. With the Kissmetrics Power Report, you can jump right into your data, pull whatever you want, and answer the most pressing questions about your business.
Currently, the Power Report is only available to our enterprise customers and isn’t available by default in the Kissmetrics 14-day free trial.
The Power Report is included with every Kissmetrics account tracking 5 million events or more per month (that’s about 500,000 visitors/month). And yes, we’re more than happy to upgrade you to one of these plans if you’d like to get access to the Power Report.
To get all the information on the Kissmetrics enterprise plans, click the button below:
About the Author: Lars Lofgren is the Kissmetrics Marketing Analyst and has his Google Analytics Individual Qualification (he’s certified). Learn how to grow your business at his marketing blog or follow him on Twitter @larslofgren.