If there’s one thing that I obsess over more than trend metrics, it’s superheroes. It’s no doubt that Hollywood has been riding the superhero train to oodles of money the past decade. However, if you’re a comic geek like myself, you’ll notice studios have been playing it safe with the more popular characters (Spider-Man, Superman, Batman, X-Men) and not until recently have they tried their luck with collaboration movies (The Avengers, Batman v. Superman, Justice League) which have been ginormous blockbusters. Historically, they have given low budgets to movies not widely known to people, and usually ends up becoming box office flop (Blade, Catwoman, Hellboy) and rarely, the high budget fan-favorite might even tank (fuck you Mr. Reynolds, for that frat-boy Hal Jordan). But we still hold out for hope that an unknown, underdog might surprise us (Mr. Pratt and the Guardians of the Galaxy did just that).
It wasn’t until the last year that there was a use of analytics in testing crowd response towards a movie pre-production. If you’re a fellow Redditor like myself, you know I’m referring to the “leaked” footage of the Deadpool test.
The controversy is this superhero/anti-hero is not relatively well known to the public eye, and really is just a champion in the eyes of geeks and freaks. The Marvel character’s comics are incredibly crude, extremely violent, and sometimes sexual. So the risk would be a big, steamy R-rated film written for a nerd following and a strong convincing of the general population to go pay to watch Ryan Reynolds in a red jumpsuit, right after he tanked a multi-million dollar film in a green suit. Disaster. Yet, the internet’s voice somehow made it’s way to Fox Studios to reconsider. So, the predictable happened. Someone grew some balls and dropped the footage designed for Fox reps (notice the last line in the footage “Hi Tom!” –> Tom Rothman, Co-Chair/CEO of Fox Film Studios). The internet exploded, and it wasn’t long until the “merc with a mouth” was green-lit for production and is currently underway.
So what do you care? What does this have to do with digital marketing? Well, they took an idea, and decided to A/B test it with the public to see what the response would be and measure the data. They tracked the responses on all platforms; the number of reposts, retweets, and observed stream hits, which more than likely led to an implementation plan to ultimately create a summarized report to hand to the big guy upstairs (no, not God, the president of the studio) to get approval. And guess what? This is exactly the basic step-by-step measurement plan used by Google Analytics, and probably something future projects might do.
So what’s in it for you to learn Google Analytics? Well, other than if you love numbers and love quantitative & qualitative data measuring, it’s really fascinating to see marketing plans built entirely off mountains of data. It’s like Neo-vision in the Matrix. All you have to do is analyze and interpret what you see to manipulate it to your advantage. If you want a career in digital marketing, you probably want to take the Google Analytics Certification.
Again. Why should you care? Well, LOTS of companies use Google Analytics to understand what their customers find and interact with when it comes to monitoring their channels., Analytics can be used to measure everything from the device, to the operating system, the browser, right down to the time and date of the kiosk to complete the sale. Creepy? Kind of. Effective? Unbeatable.
So what does the G.A. certification cover? Well, that will be something I have to get back to you about when I take it next week. But the basic structure (from what I’ve been told) is the understanding that customers can start their purchase journey at any point along the decision path. Marketers must tap into this and anticipate where they will be. The analysis of micro and macro conversions and other basic fundamentals.
From Unit 2 in the training, the principles of what a digital marketer would do is essentially What do you need to measure? Define the measurement plan: identifies business’ objectives → Understand the technical infrastructure: what are our server technologies? Are we on mobile? are we using responsive design? → create implementation plan: specific to analytics tool you’re using. (Google analytics: define code snippets in product features) → have web team implement the tracking recommendations made. Measurement plan needs to be maintained and refined.
How, you ask, is this done?
Approach digital measurement by starting with business objectives because measurements understand if you are making good or bad decisions and what actions to take to move forward
- document objectives
- identify strategies and tactics
- choose KPIs
- choose segments
- choose targets
Afterward, you would create implementation plan:
- standard dimensions → basic page tag & metrics (gets bulk of data)
- track KPI: business outcomes → goals and ecommerce
- clean data → filter/setting (normalizes data to be useful)
- marketing channels → campaign tracking and AdWords linking
- simplified reporting → custom reports and dashboards
But the most useful thing you could take away is the 4 step process of Google Analytics inhow data is collected and processed
- data collection
- collect user interaction data from
- websites, mobile apps or digitally interactive environment
- track a website using javascript code
- pulls from website itself, the browser, the OS, the device
- pulls from referring source (did you get here from Facebook?)
- interactions are called “hits”
- Mobile app tracking doesn’t use javascript; it uses methods specific to OS on the device (not the same as website tracking)
- collects data after each activity
- mobile devices are not always on the internet, so its not real-time. Google analytics just stores the data and will send it when it can
- you have to define your “hits”
- configuration
- “transformation” step from raw data to something usefull
- filters can include or exclude certain data from reports
- processing
- once data is submitted to a database, it cannot be changed
- reporting
- collect user interaction data from
If I went over your head, let’s explain this like you’re 5.
You have your website (whoopee). You decide to sell your Kool-ade on your website. After a week, you only have one sale (mommy). So you log on to your Google Analytics account and see you actually have 100 hits (visits) on your page a day. So why aren’t they buying your drink? Bounce rate is 60%, and no leads have been generated. Why? You track the behavior of the qualitative data tracked because you were smart to put the javascript code on every page for metric tracking. You see people are looking from their phones and have a session time of under 30 seconds and are being referred from tastydrinks.net but have a strong suspicion people want Gatorade, not Kool-ade because you have very similar keywords and are listed very closely on the listing page. Because of this new data, you decide it is in your best interest to change your keywords and maybe reallocate your ad spending on a site that is more likely to navigate users looking for juicy drinks and not sports drinks. Boom. Analytics.
brianhaug 