When analyzing the effectiveness of various marketing channels, the first platform that most marketers turn to is Google Analytics. Understanding page views, sessions, and distribution channels – among many other key indicators – is made easy, and the entire process of website optimization has come to rely on it.
…And don’t get me wrong; we love Google Analytics as much as the next product marketer, but where we have found its real power is in using it in conjunction with product lifecycle analytics.
Today, I’m going to run you through the series of steps we took to analyze Shift, an electron email client we launched on Product Hunt back in December, and show you some of the ways that we were able to easily attribute purchases and installs all the way back to our marketing and distribution channels.
First things first: Google Analytics
This was my first stop to see which paid and organic channels were most effective in bringing customers to our new site. By this measure, the answer was clear within 60-seconds: Facebook.
Google Analytics also gave me some great insights into our main call-to-action. We wanted prospective customers to submit their email so we could send them a secure download link for Shift. It was working.
As tempting as it was to stop there, and simply increase our Facebook efforts – both paid and organic – with expanded ad sets and organic content, we wanted to dig a little bit deeper and understand how our Facebook customers were actually engaging with Shift, within the application.
Second step: Deskmetrics Analytics for Install Data
While it was clear that Facebook was the medium that drove the most traffic to our website, further analysis within Deskmetrics revealed that the quality of leads generated by Facebook was nowhere near that of other channels.
This revelation meant that increasing engagement would be a lot more complex than simply tinkering with ad sets; we had our work cut out for us.
How many website visitors, after submitting their email, actually installed Shift?
This is where Deskmetrics came in. We use our own platform to dig in to this type of data because it provides end-to-end customer tracking, and custom engagement reports in real-time.
By pulling a simple engagement report, I could see a very clear picture of page view to install to purchase, for each channel, within seconds.
As a gut-check, I further sorted these results, and even exported them to Excel to compare them directly with Google Analytics. Though Facebook was the most effective medium for generating traffic and call-to-action completions on our website, there were several channels that were more effective in actually prompting customers to install (namely, Lifehacker, DesignerNews, and Feedly). As earned (aka, non-paid) media channels, this was quite exciting for our marketing team to see.
From there, we could dig in much further, to our most (and least!) engaged customers, and really paint a picture of each source.
For example, we wanted to know:
How many email accounts did the average customer add to Shift?
As it turns out, the average customer adds 1.97 accounts, but we have one user that has 14!
With over 10,000+ new users in the first 2 weeks, we had a lot of data to mine. Here are a few fun facts Deskmetrics revealed to us, in a few simple reports:
- The average account switches per day, per user: 25+ (and one customer switched 415 times!)
- 10% of customers used our Referral Program
- We have an incredibly international user-base: only 37% US, followed by CA, FR, UK, IR, ES, AU, IN, PL
- 94.5% accounts added to Shift are Google accounts, 4.5% are Microsoft accounts
As a final step, I needed to learn more about install to purchase behavior:
Which channel was the most effective in converting customers to paid, after their 30-day Pro trial expired?
I ran the exact same report in Deskmetrics that I had for the install data, but this time I simply switched out Install for Purchase in the Deskmetrics report query.
This Custom Report revealed that Product Hunt was our most effective medium, exceeding the next closest channel by more than 288 per cent!
While our entire team could *see* the engagement and enthusiasm via comments, social media, and press coverage throughout the launch period, one simple Deskmetrics report confirmed it.
So, what does all of this mean?
Google Analytics is a fantastic analytics platform. Full stop.
It provides insights that are crucial for everyone involved in product marketing, and development. That said, it is very tricky and time-consuming to actually gain a full understanding of product lifecycle, from page view all the way through to in-app conversions using solely Google Analytics.
In our case, Google Analytics helped us to understand which customers submitted their email, and from which channel. Things got a little hairy when we tried to attribute this to downloads, and ultimately, purchases.
Had we only looked at Google Analytics, we may have misidentified some prospective customers as being ‘ready to purchase,’ and made the assumption that increasing our Facebook spend would correlate directly with a bump in purchases.
Deskmetrics quickly revealed the fact that Facebook users were submitting their email (and therefore, completing the Google Analytics goal) but not actually following through with the download in as great of numbers.
This meant that we would need to change our angle, and send a simple reminder to Facebook users to actually download Shift, rather than wasting time, and potentially confusing them, with other call-to-action requests. Luckily, we were able to respond to this, and refine our email campaigns accordingly.
Google Analytics is a fantastic resource for understanding your product website. For understanding the entire product lifecycle, and the customer’s full experience with your application, however, another layer of analytics is critical.
In a marketing and/or product role, it is crucial to understand and respond to the buyer’s journey, as it happens. Making sense of how customers experience your website, from page view right through to download, purchase, and beyond, is really the only way to help your development team iterate effectively. With the right software, this is possible, and believe it or not, it doesn’t have to be painful!
We built Deskmetrics to help marketers, and product managers, just like you, do exactly this.