Successful business decisions are rarely the ones with the most dramatic results. While the steep growth curves can be fun while they last, most of the day-to-day trends that can make or break a company come from smaller trends and decisions. No place is this more true than with advertising, where even a seemingly marginal difference in response rates can have huge results over the life of a campaign.

With so many options — from old-media ad buys in TV, radio and print to the various online and email campaigns of new-media — you could easily find yourself with a range of channels with conversion rates separated by a mere fraction of a percent. And even when you find the right channel, how do you know if you’re getting the best response? How can you truly know what kind of advertising works best for your company?

The answer is simple: You test everything. And as you test, you gather all the data you can. The more data you have about each campaign, platform and channel, the better informed your decisions become. With enough data, you can see exactly how much it costs you to acquire a new customer on any platform.

For truly deep results, however, you need to take your testing to the next level. One of the most effective ways of doing this is by A/B testing your advertising. To A/B test an ad, you simply need to run parallel campaigns on the same channel. Both ads are identical except for a single element, such as placement of the call-to-action. By testing the responses to each ad, you can easily observe the differences in the response.

I’ve seen A/B testing yield great results for my clients, particularly those who are in the early stages of professionalizing their sales and other business practices. It’s an approach to advertising that I’ve really grown to appreciate over the years. A/B testing allows for an efficient and scalable foundation to build an ad campaign on top of. By testing multiple calls-to-action, ad copy, images and other elements, it allows the marketing team to find the approach with the highest conversion rates.

While A/B testing tends to be most popular for online advertising, you can actually use it in almost any situation where you can present slightly different experiences to potential customers. You can even adapt the idea to a retail setting, testing out everything from display placements to customer greetings.

As you develop your advertising campaigns, tracking your results as you go, A/B testing gives you yet another dimension to gauge your future success by. After substantial A/B testing, for instance, you might discover that you get the biggest bang for your buck from a Google AdWords campaign, and that conversions from TV ads simply can’t compete.

Of course, all of this testing needs to be in service of a specific action you would like your prospective customer to take. Reaching a customer is important, but you don’t just want them to hear your message, you want them to act on it. Next time, I’ll be talking about one of the most important pieces of the advertising puzzle: The call-to-action.