If your measurement time is simply setting up and looking through various Google Analytics reports then this post is for you!
Nothing wrong with Google Analytics, it’s a great tool but it’s merely one tool that looks at a very specific and limited data set i.e website behavior and channel performance.
Out the box it tells you nothing about:
- Life Time Value
- Revenue vs target
- Whether someone liked an article or not
- How a website visitors relates to a customer in a CRM
- How website visitors relate to an email contact
- Customer satisfaction
- Where visitors clicked on a webpage
- A funnel that spans multiple platforms
Google Analytics doesn’t even tell us how revenue or leads compare to target.
There’s a lot more we could look at but you get the picture, there’s a lot more to measurement and insight than just the data in Google Analytics.
What gets measured gets improved
If you work in Google Analytics to any degree your main job is probably to extract insight from the data and present it in a way that drives meaningful business change i.e. insights form recommendations, that when implemented, lead to the company’s KPIs improving.
If there are more data sources within a business then there’s more potential for insight!
This really is a case of leave no stone unturned; do a quick audit and list all of the data sources your client has available, some may not be obvious. For example, an independent review platform about your client’s company can be scraped and made into a new data source.
If there’s data missing that would be useful make a note of that – as well as making use of existing data we need to plan how and where additional data will be stored. Before implementing technology to gather more data make sure there is a clear rationale for the data; don’t become one of these analysts that gathers data just for the sake of it.
So what kind of insight could we gather from this additional data?
The biggest complaint from customers is delivery time. What would be the expected uplift in recurring revenue if we fix that?
Analysing all of the clients Trustpilot reviews we identified three new personas; let’s run a split test that weights the copy on the homepage to the identified personas in terms of priority.
This set of products generates revenue but not very much profit – turn down the amount we spend on those adgroups.
The lifetime value of customers over 12 months, coming from the paid channel, is $2,356. Adjust our cost per acquisition limit to align with this new data.
Channel X is 13% behind target in terms of driving profit; review the marketing activities on this channel.
Overall customer satisfaction has dropped by 8% compared to the previous period. Identify recurring issues and implement a strategy to rectify this.
Content feedback for the last 6 months articles, captured by a smiley face widget, has dropped 43% compared to the overall Avg. Organic traffic has also plateaued whereas previously it was increasing at 20% month on month. Work on the editorial quality as a matter of priority and develop a KPI dashboard for this team.
This list could go on and on.
Naturally to work fluently across many platforms requires a bigger skillset that just opening Google Analytics and looking through a few reports.
SQL and Python are languages for ensuring you can: