Analytics: An All-Too-Often Flawed Science

Chris Bretschger

June 23, 2015

Analytics, predictive or otherwise, seems to be a practice that is being abused in our industry.  Having focused on digital marketing for the last 10 years, I’ve lived my career in numbers, access tables/databases, and way too complex spreadsheets.  30-50 page (yes, some the size of ‘the yellow pages’) decks for monthly site reports filled with pretty charts, and impressive figures talking about 1,000% increases and astronomical ROI figures. It all seems so good until the day happens when those increases aren’t so high, and that ROI doesn’t look so great.

There are a few bones to pick with the industry:

  1. Definition – All reports need a glossary for data definitions.  How are the numbers being pulled, how do they relate to what the true bottom line, and what do they TRULY mean?
  2. Transparency – Where do the fees come in to play, where does the media end and the agency begin (a line continually becoming blurred)
  3. Analysis – Numbers are great, but what our clients need is analysis.  What do they mean, what got us to this point – why are things going up? Why are they going down? And what are we going to do about it?



Data and resulting analytics is a dirty business – perfect implementation is difficult, and not many folks have taken the time to clean up their analytics platform to clean the data up from the get go. Self-referrals, spam referral sources, funnel tracking, data hierarchy – if these terms don’t sound familiar, it’s definitely time to dig in – they’re common issues, and if you’re not trying to fix or implement one of these right now, you should pick up the phone and call your web folks.

Regardless, data needs to be cleaned before it’s being reported, and clean data needs to leave breadcrumbs to its source so that it can be replicated in the future.  Nothing is more bothersome than not being able to re-create data from a report.  Definitions should be readily available at all times.


In order to fully optimize your marketing budget it’s important to know what’s going towards media, what’s going towards hard costs, and what’s going toward fees.  Hidden fees are constantly becoming more and more prevalent in the industry, and this makes it difficult, if not impossible for a client truly to see how their funds are being allocated.


I can’t begin to explain how it felt when I used to be given giant decks full of numbers but not a word of explanation as to what they mean, or what we can derive/learn from what’s being presented.  Analysis is far and away the area lacking in most analytic reports today.  Always provide optimization scenarios, testing parameters, or rationale/educated reasoning as to why something is happening, as opposed to purely stating that it happened.

The analysis is the tricky part of any analytics report – it requires the agency to put in stone (writing) their thoughts and perspectives as to what is happening in the market on an on-going basis, it takes time – a lot of time, and it takes a full understanding of the different variables that impact the key metrics being reported

Wrapping it all Up

As a whole, I feel like our community goes through and speaks to a lot of numbers (we have to, right? It’s the digital age – numbers are everywhere), but we’re sorely lacking in the analysis of what those numbers truly mean, and how they can effect the business as a whole.

As agencies – this is a service that undoubtedly all clients need — a better window into the performance and future optimization of their dollars.  As clients, it’s a great system of checks & balances to keep pulse on what’s happening in the market, and working to keep ahead of the trends to come. When it comes down to your reporting, make sure to see more than just numbers and pretty charts – get down, get dirty, and make sure that there’s context, recommendations, and analysis to go along with all that tracking out there.