Front-end and back-end analytics are clearly different, and together their true power is a testament to how neither is necessarily better than the other. One is the peanut butter to the other’s chocolate – delicious apart, but better together. We define front-end analytics as: analysis performed on all available business data, combined with relevant third-party information sources, to gain insights into business opportunities before brand strategies, media strategies and creative strategies are executed. (That’s as opposed to back-end analytics which measure marketing performance.)
Specific to front-end analytics, you can’t go in assuming you know the answer, looking for data to support your presumed hypothesis. Rather, you want to go in and look for a solution to a business problem, and what that opportunity looks like as a dataset. You start with the business.
Callahan is hired as an agency not to buy media impressions, or to make ads — we’re not a transactional office of order-takers. Instead, we’re hired to solve for business opportunities: to sell more pizza, sell more heartworm medication, or construct email campaign plans that move the needle on targeted demographic groups. In short, we’re hired to make money for our clients.
Using front-end analytics to uncover a better future for your business
To that end, front-end analytics is really an exercise in examining the business, to understand the variables in the business, identify opportunities and how we can best help by making the biggest impact on a business. Front-end analytics shines a spotlight on what work we should be doing — it’s the horse pulling the cart. It helps to illuminate the path of where we should be going.
Sticking with that analogy, you also have the cart — back-end analytics, which are different in more than name alone. Back-end is postmortem. Back-end analytics look at the efficacy of what we ended up doing. So back-end analytics is what I think a lot of agencies call analytics. It’s reporting: how many impressions did we run? How many clicks did we get? What was our engagement? What was sentiment? This is all important stuff, but its value has a limit without the context of the front-end data analysis.
When you marry these two data sets together, you can start drawing correlations — you can link those back-end metrics to the business impact because you’ve already established the framework — and proper KPIs —through which you’re trying to ultimately evaluate success of any marketing campaign. The true KPI should not be to achieve a certain click-through rate, or a certain cost per click, or a certain cost per conversion. It should be to sell more of a product or service. Front-end analytics sets things up for success from the beginning.
For example, consider a hypothetical client who works with us and says they want to sell more widgets. Well, we can look at that back-end marketing KPIs and measure what media produced the most impressions, but that doesn’t help understand what tactics best impacted actual widget sales.
Now, consider what those back-end outputs would look like if we could correlate them with front-end, business-focused data analytics. How many varieties of widgets are being sold at each store, when and where where they sold, average cost per sale, quantity sold per customer, inventory and foot traffic, average overall sales data for each store, seasonal information — and then, what marketing activities drove all that. That adds additional context to otherwise flat back-end KPI measurement. That’s when it gets truly interesting and enlightening in a way that can drive business outcomes.
A lot of times back-end analytics are the basis for how you optimize media plans or marketing campaigns, so not to diminish that — it’s really important in that regard. But in isolation of having anything on the front end, it’s a pretty myopic, narrow view, and it can leave you scratching your head without a connection with front-end data. Without that marriage of the two, what really was the impact of the media buy, the marketing plan?
If this interplay between front-end and back-end analytics piques your interest, I invite you to read Zack Pike’s in-depth white paper on the Callahan Intelligence Platform.