The Magic of Test-and-Learn Marketing
Which is more important: What customers say? Or what they actually do? The answer, of course, is actual behavior. Actual behavior is the marketing trend that is changing how millions of business owners think. It’s called A/B testing, and at its core the concept is rather simple: Test something several different ways in the real world and see what works best. Then go with that. What could be more basic?
By Daniel Kehrer
When business owners create products or services – and the marketing plans to sell them – they often base their decisions on what customers tell them. But which is more important: What customers say? Or what they actually do?
The answer, of course, is actual behavior. And therein lies the power of a marketing trend that’s changing how millions of business owners think. It’s called A/B testing, and at its core the concept is rather simple: Test something several different ways in the real world and see what works best. Then go with that. What could be more basic?
Yet businesses both big and small trip on this every day. They survey customer preferences and base decisions on what they say. But stated preferences can’t hold a candle to actual behavior when it comes to identifying what works – and what doesn’t.
The Internet world has long used A/B testing to try out different website designs. But the concept is so straightforward and effective that it can apply to almost any circumstance where a choice must be made. And for small businesses, the test-and-learn approach makes perfect sense.
A/B testing is exactly what it sounds like. You test approach “A” and approach “B” and see which does better. The key is that you do it in the real word with real customers and real-time results. And you keep doing it with different customer segments until you have the answers you need.
In buzzword terms, this is a “data driven” approach. The “data” are the results of the battle between A and B. You need to do it for different customer segments because while Option A might win generally, some segments might favor Option B.
One reason A/B testing works is that customers often don’t know themselves what they really want. They might say one thing, and do another. Sound familiar? A/B testing cuts through the fog.
Entrepreneurs are told over and over to conduct surveys and other traditional market research. But observing actual behavior is almost always a better predictor of future behavior. The hugely popular drink Red Bull is a good example. When first surveyed about it, customers called it disgusting. Yet annual sales are now in the billions of dollars.
The beauty of A/B testing is that it has no problem delivering solid answers about highly subjective preferences for such things as taste, color, shapes, images, layouts and the like. If you simply ask customers which color they prefer, for example, they’ll express a preference. But when you actually observe and measure their response to different colored products, packaging or even web pages with A/B testing, the results might be radically different.
In the real world of small and local businesses, what this means is that while many things influence consumer purchase behavior, the consumers themselves are often unaware of them. For example, the package something comes in has a big influence on purchasing, even when it shouldn’t matter. Music influences what people buy and how long they spend in a shop, restaurant or pub. Lighting influences how people notice products and buy them. Other factors include smells, excitement, the number of choices (fewer can be better) and how prices are displayed.
Philip Graves, a consumer behavior consultant, says that while these things, and many others, influence what we buy, consumers usually don’t have the faintest idea their thought process has been altered.
The trick here is not to ignore what your customers say, but rather to pay far more attention to what they do by offering different choices and measuring and observing (first hand if possible) the results. Don’t simply ask customers to “imagine” something. You have to create it and actually let them experience it. This live testing gets the best results by far. It works even better if people don’t know they’re part of a test. You want people to behave as normally as possible.
Just remember. No matter what business you’re in, it’s vital to continuously test, learn and adapt. Asking people what they think is no substitute for experimenting in real time. You don’t have to fully understand why Option A performs better. If it does, go with it.
Copyright © 2000-2012 BizBest® Media Corp. All Rights Reserved.