ATLAR: Ask – Test – Learn – Allocate – Repeat

Digital marketing, in a nutshell, folks, is simple, not easy. While it’s not something most marketers or advertisers will cop to, educated guessing is the name of the game, especially early on when growing a business’s marketing efforts. Later on, when markets have been identified clearly, and enough data has been gathered to make assumptions, this changes a bit. However, until thoughtful questions have been asked about the business’s goals or objectives for each channel or platform where money is being spent then success, and the audience(s) we’re trying to seduce, may ambiguous. Continuous testing, rigorous analysis, and judicious allocation of resources on a continuing basis will ultimately bear fruit (if you’re working with Hawke anyway). The question is really over what time horizon are you and your business comfortable stomaching the spend?


Asking questions, of yourself or agency, is critical in formulating any advertising approach before shelling out any dough. How much are you willing to spend? What are you trying to learn? What suspicions or theories are you looking to bolster or disprove? From a general perspective, these questions should come fairly quickly and intuitively. Do I get better click-through rates from display retargeting efforts aimed at men or women? Is that in alignment with site engagement metrics or other advertising platforms? Why or why not? Where are those ads being served? Etc.

Once there is a general idea of what to look for, move forward and test it. DO NOT THINK TOO HARD. Just test it. Your gut is likely on point, but we’ve seen nearly all of our clients here at Hawke be fairly surprised by the most responsive audience(s) on previously untested channels at some point.


This can be the most difficult part of the process for those new to digital marketing or a previously unexplored platform or channel. This isn’t to say the idea of testing is tough to grasp, nor is it challenging in most cases to execute a spend. However, building out campaigns and [ad sets/ad groups/flights/segments] with what reporting functionality each platform offers can be daunting. Couple this with the analytics platform your business is using and the multitude of UTM tracking options, the process can start to feel overwhelming. This is why the first step of asking questions and identifying what to test is so critical before moving to the next phase of how to test.


“Big Data” is a term that has been beaten to death in the last 18 months and has been replaced with what we’ve all understood it to mean all along which is smart data. If advertising efforts have been set up correctly during the testing phase, now our job is to get jiggy wit’ some data (I personally love this part…and 1997 Will Smith).

What (if any) numbers stick out? What do these metrics look like compared to last week/month/year? Anything you hoped to learn but don’t have enough data to build a concrete understanding from? Try to look through at least a couple of lenses, or better still, three or four. If you are running Facebook ads for the first time, see what Google Analytics has to say. Last click vs. multichannel is another topic entirely but look at time on site, bounce rate, and user flow, not just click-through rates, costs, and conversions. What is happening with each of the thousand, ten thousand, hundred thousand, etc. new visitors you created? The better you can understand what they are doing, the better you can influence them to do something different or repeat the same behavior(s) going forward.


And reallocate. How much is enough to spend to get meaningful data? Will we learn what we’re trying to learn by spending $10,000 in one day or is it more educational to test $1,000/day over 10 days? Or $330/day over 30 days? Are there platforms or channels that we can use to test multiple hypothesis at once?

Budget allocation is a critical part of a marketing effort. For many of the growth-stage companies we work with at Hawke, cash resources are precious and over-spending arbitrarily is bad for business. Frankly, this holds true for businesses large or small, though the bigger clients can stomach the loss a bit better. I compare this part of advertising in my mind to F1 racing or any kind of racing for that matter. The best mechanics and engineers (agencies/marketing departments) can build a beautifully efficient race car (ad platforms), but if there isn’t enough fuel (money) in the tank, they won’t make it to the finish line. Conversely, if there is too much fuel, the car will be too heavy and becomes sluggish relative to the competition.

Another consideration is one’s aversion to risk and the expected time of return on ad spend. Many of the channels available to today’s digital marketers are excellent for top-of-funnel advertising or brand building but may take some time and continuous efforts/funding to really show results. Much like pumping a well, the first efforts yield nothing, while after the water starts to flow it doesn’t take much effort (if any) to get water out of the ground. Be sure to understand completely the risk/reward or various pros/cons of any new advertising effort or channel and spend knowing the expected returns and time horizons rather than hoping or assuming.


I heard a guy wandering around the streets in Boston one night muttering, “shampoo…repeat…shampoo…the most important step….wash, rinse, REPEAT!….repeat…shampoo….the most important step…” I don’t think he was all there, but he was making a lot of sense. The most important step of washing your hair is the last one—repeat. Circling back to the process after the first iteration and for each going forward, the questions should become more granular. As the business develops a deeper understanding of what it’s doing, new ideas to test and new testing methodologies can be utilized. Be deliberate in this phase. Marketers and advertisers often seem in a hurry to “not miss out on ______” instead of keeping their powder dry and making intelligent ad buys. I’m a big believer in iterative advertising and getting money spent quickly; however, in each season or after a specified test period, I try always to take enough time with the data and resulting knowledge to ensure improvement and additional learning going forward.

As a general recommendation, I tell folks to spend their money using a 70-20-10 breakdown and adjust as individual tolerance permits. If 70 percent of the budget is dedicated to fairly sustainable and productive ad efforts that have either been verified through previous tests or are conservative in nature, then 20 percent should be more speculative, and 10 percent should really reach. At some point, I heard the anecdote that Google made people work on projects other than their “job” one day/week or 10 percent of the time (something like this) and a disproportionate chunk of their innovation and most successful platforms have come as a result of this time. Fact or fiction, I can’t say, though, I can say with certainty this has been true for many advertisers at Hawke as well. Don’t be scared to try new things and have some fun in the process. There is no magic box that you put $1 in on one side and get $2 forever. However, thoughtful advertising and meticulous analysis can get you that kind of return and more. But the target is both constantly moving and amorphous, so we need to be diligent. Learn more, test more, call Hawke if you need help.

© 2018 | All rights reserved Hawke Media.