Why Test Doesn't Have to Be a Four-letter Word

Last week I wrote to you about my personal discontent with the way the word "test" is being thrown around these days. A couple of people on the SPINboard dubbed it perfectly. They called it an "experiment." While there is no perfect word, we need to perfect the concept.

I wholeheartedly agree that tests can be a bad thing. They can also be a safe way for you and your client to dip your toe in the water of digital media. Tests can be a comfortable way for clients to ease into a medium they haven't experienced. They can be low in cost. Flight times can be short. The list goes on.

Before you go into or pitch a "test" to your client or internal management team, consider the following:

It may be under your nose. Before you look outside, look inside. You may have access to log files with usage patterns, click streams, and the like. Review any online activity be it email, SEO, paid placement, online ads, etc. Look at best of breed.

Dot your I's and cross your T's. Make sure your goals are clearly defined. Don't leave any stone unturned. Put as much time (if not more) into the planning of a test as you would with a large budget campaign.



Get it down on paper (or email). Thoroughly research sites, vendors, and venues to insure proper impression levels and placements. Study the competition. Where are they? Should you swim where the fish are or are you soon to be floating face up in the same water? Send out RFPs and touch base by phone with your sales reps. Time can kill you. It can't hurt to get all credit apps and ad specs straight away. As you are waiting for reps to respond to your RFP, you can get the ball rolling on the paperwork end of it. If you wait, this can slow down the process dramatically.

Be on the same team. Sales reps and vendors need to understand the importance of your efforts. Share what you already know about your/your clients' business, traffic, buying habits, etc. Make sure your reps stay on top of their proposal. Don't have them kick it to an internal staff person and walk away. Pick their brains. They are a wealth of information. Ask if they've worked with any of your competition. Ask to see as many stats to back up the recommendation: traffic, unique users, tracking methodology, indices, and comps. This should be a collaborative process.

Backward-forward. Implement an aggressive back-end tracking program. Your team and your reps should be watching this closely. Fairly early on you'll start to see patterns of activity and perhaps trends among your target audience. Don't waste time: optimize when you can. Let's face it, if the campaign works, you will most likely be given more money to implement a full campaign.

Idea generation. Constantly look for a better idea. I find the best ways to do this is to tap into the industry. Ask around. You most likely have sales and planner/buyer friends. Ask if they've had success with certain sites, programs, ad units, tracking elements, and the like. Don't just look at successes; look at failures as well. Read. There are several good sites, blogs, and publications out there. Search for case studies. When it comes to online achievements, people talk. This can certainly help you.

This article is by no means written to contradict anything any of the readers, SPIN posters, or I had to say last week. We al need to realize tests can be either a disaster or something to cheer about. I want to hear them both. Feel free to post or email me directly at

Next story loading loading..