How to Tell Compelling Stories with Google Analytics Data
Sarah Danks | January 8, 2016
Originally published July 17, 2015
Jeff Sauer’s 2015 MnSearch Summit Presentation
Jeff, founder of Jeffalytics, international speaker and digital marketing consultant, says he created his first website as an eighth grader and hasn’t looked back since. Well, I disagree with the last part of that statement — I’d say he looks back a lot. At the analytics history of EVERYTHING, that is.
If there’s data to be aggregated, analyzed or agonized over, Jeff’s going to figure out how to do it — and what it all means.
We were pleased as punch to have Jeff come and speak at the 2nd annual MnSearch Summit, held in St. Paul, Minnesota. He joined 16 other top-notch speakers and over 300 attendees that all joined together for a day chock-full of online marketing goodness at the lovely Rivercentre venue.
Here’s a recap of Jeff’s MnSearch session:
How We Can Tell a Compelling Story With Our Google Analytics Data
First of all he introduced himself and said he LITERALLY drove to Minnesota from San Francisco on his motorcycle. What a prince. (PS: he says blouses and motorcycles aren’t a good mix.)
(PPS: he was lying. Not about the blouse/motorcycle thing; about him driving across the country on a Harley.)
But why would he tell such a whopper of a story? Because it got our attention.
He then smacked us right dab in our egos by telling us:
“Most of the stories we tell as marketers are true…
…but they’re boring as shit.”
~ Jeff Sauer
So let’s see another story, then.
Who’s ever received or sent a report like this?
Hang onto your hats for the moment of truth: this was written by a computer.
Did you know there’s a FREE tool that does this? It’s called QuillEngage. (And that’s a true story.)
In March, the New York Times had a quiz. They asked, “did a human or a computer write this?” They then supplied readers with a text snippet — readers were supposed to determine whether or not said content was written by a computer or an actual person.
Jeff, being the analytical thinker he is, scored 50% on the quiz.
So, time is not on our side…we’ve already lost this battle. Wait a second. Can a computer REALLY replace you at work?
But, but, we can’t let the computers take our JERBS (ermagerd)!
So, how can we do MORE? You know, to ensure computers won’t take over the world — I mean our careers. Let’s borrow some storytelling skills from Prince Shakespeare…
5 Stories We Can Tell With Google Analytics Data
Google releases 70+ updates per year, so how the hell do we keep up?
With science, of course.
(periodic table of Google Analytics available at Jeffalytics)
#2: GOOD vs. EVIL
Bm: Benchmarking the Competition
Here are some stories:
- earned media rockstars
- email laggards (is email performing poorly or just not tracked?)
- falling behind with paid media (we’re not spending anything on advertising — but should we be?)
- we’re big in Japan!
- we missed the mark on mobile
How do we benchmark?
Easy. These reports require just one setting:
Benchmarks are set by industry vertical —
You can even choose location and website size.
Supplement your story with additional data:
Use SEMRush or SpyFu for competitive intelligence; SimilarWeb to compare traffic; and/or Moat for Display Intelligence.
Now you can project your marketing budget. Provide several well-researched options:
- Low Watermark (what WE are currently doing)
- Medium Watermark (what our COMPETITORS are doing)
- High Watermark (our market potential)
Develop a low/medium/high watermark:
Project revenues for each budget:
Put together a profit model to share:
Story: Give us a bigger budget and we’ll generate $21 million in profit!
#3: COMING OF AGE
Dg: Demographics Reports
Demographics reports will bring in data. For example, this data is for approximately 50% of users:
Affinity & in-market segment reports:
In-market segments are a goldmine! You can directly target these people with display ads and remarketing. Then there’s the demographic segmentation opportunity. You can review users by in-market segment.
How to Get Demographics?
One setting and some code…
…and you can enable demographics reports setting:
This requires one line of (Universal) code:
But it’s even easier using Google Tag Manager:
More on insights from demographics.
More coming-of-age stories you can tell:
- our content marketing efforts are improving
- we have grown our presence YOY because (x)
- new advertising opportunities to test and learn
#4: DARKNESS vs LIGHT
Np: Organic search and (not provided) keywords
10 alternatives to (not provided):
- Measure overall organic traffic over time
- Segment organic search traffic by landing page
- Use landing pages as a secondary dimension
- Use filters to make (not provided) more meaningful
- Use multi-channel funnels to prove value
- Hook up with Google Search Console
- Segment organic search traffic by demographics
- Use dashboards to surface the most important metrics
- Paid & organic search reports in AdWords
- Use Google Trends
(you can see Jeff’s full article on his ten alternatives to (not provided) over at Moz)
Cg: Custom Channel Grouping
Default channels are too broad
So now what?
Define New Channels; Tell Better Stories
Story: Brand vs. Non-Brand
Story: Non-Brand organic search rules
Story: We still don’t know a lot!
Story: we’re earning it
Story: paid media drives our sales
Story: we have 45% unaccountable traffic (where’s it coming from?)
How to Get Custom Channels?
Define your own settings. Tell Google Analytics your channels.
Then customize each channel:
- non-branded organic search
- mobile + organic
- social ads vs PPC
- guest posts vs earned links
- mobile + paid media
- webmail + email
- earned social vs paid social
Co = Content Grouping
Analyze content performance by theme:
Then group content by year of publication.
How does word count affect organic?
How to Get Content Groupings: Rules, Extraction or Code
Some groupings require code
It helps to know the art of content grouping by waterfall:
Get the Wordpress content grouping code:
(more on content groupings)
#5: APPEARANCE & REALITY
Lots of us know this story: would you just look at all this great traffic! Tell everyone!!! (Let’s get t-shirts made — they shall say “Keep Calm Cause We Killing It!“)
But, that story’s about as real as Ice Cream Land. Or a world where Prince is a sex symbol.
Dq = Data Quality Issue
Rs = Referrer Spam Issue
Jeff first noticed this in late 2014 and consequently wrote about 8 steps to eliminate bad data in Google Analytics.
Thing is, these referrals aren’t real. In fact, they’re CRAP!
And all from one person in Russia (probably drinking vodka to keep varm in vinter).
(I’m entirely obsessed with Outlander right now, so och, this next slide is verra gud.)
There was no response from Google for six months…
…which is awkward.
So Jeff nudged a little harder in April of 2015 when he asked is Google Analytics’ newest data quality issue the most challenging?
Here’s an excerpt from his article:
“There is a massive data quality issue happening right now in Google Analytics, and not enough people are talking about it. This is the most challenging data quality issue that I have seen in the 10 years I have used the product.
I worry that if a permanent and swift solution is not found for this issue, it will cause long term damage to the credibility of the product. That damage will eat away at the market share that Google has fought so hard to gain and drive visitors to look for other solutions. I am concerned.”
And we keep seeing shit like this:
And pageview spam. And event spam.
Is THIS what it sounds like when doves cry???
This has become a massive issue, with only one solution:
We need a fresh start.
Jeff’s Key Actions To Take Today:
- Tell a story from your data — don’t let computers win
- Develop a low/middle/high budget projection
- Use demographics to inform marketing segments
- Set up content & channel groupings for better insights
- Install filters to block some referral spam
- Demand a Google Analytics data annulment
To see his MnSearch deck in its entirety, you can see Jeff’s presentation here.
***I didn’t recap a portion of the slides as Google is currently working on a solution to the bad referral data issue.***