• by Gavin Finn , Op-Ed Contributor, July 27, 2010

Don't be discouraged by a failure. It can be a positive experience. Failure is, in a sense, the highway to success, inasmuch as every discovery of what is false leads us to seek earnestly after what is true, and every fresh experience points out some form of error, which we shall afterwards carefully avoid.--John Keats (1795 - 1821)

In almost every marketing campaign or event, managers always want to know what the conversion ratio was. Whether the objective was a sale, a registration, a white paper download, or any other marketing goal, the primary metric is typically a measure of the number of people who did what we wanted them to do.

What is easy for marketers to overlook are some of the key metrics along the way that help to understand why our prospective customers did what they did, regardless of whether they "converted." Important measures of customer behavior can lead to an understanding of what to expect when we present customers with choices, programs, selections, or offers. Ultimately, we need to be able to predict behavior, and the very best way to become better at prediction over time is to measure every customer action (and non-action) and then to analyze all of these measurements carefully.

Precise measurement can be useful, of course, but only when there's an understanding as to why you're analyzing the data, and which actions can result from this analysis. Rather than looking at whether enough data exist, ask "What's the point of collecting it?" After all, it's the why that will help you to understand what actions to take. Collecting all the data in the world does not make for success. There's much more to it than just the data. In fact, in any business setting, to make use of measurements, several dimensions need to be considered:

• Data - the raw numeric or alphanumeric values associated with specific measured events;

• Information - data collected into tables/organized areas so that they can be used in a meaningful way;

• Analytics - information that has been sorted through, using a range of algorithms and programs, so that aggregated trends and/or results are obtained and made visible;

• Insights - key leanings from the analytics are identified in terms of meaningful business conclusions that can be drawn;

• Actions - based on insights, business actions are taken to correct or exploit the results of all of the work done in the data/information/analytics/insights value chain.

While there is a story in the data somewhere, without a well-understood process of uncovering the relationships between the data and the context in which they were collected, the story will most likely remain hidden. The very first step is to identify what the intended use of the information may be. For example, in retail settings one might look to better respond to consumer demand; improve operational efficiency; or fight competitive pressures such as price deflation and eroding gross margins. Only once the objective has been identified can you look at what's possible to measure, and in turn, this can lead to actions.

So what data could be captured (in both an online and offline environment) that would lead to information, facilitate analytics, and drive insights around pre-purchase customer behavior? Information such as:

• Which products a customer looked at before purchasing the selected model;

• Which product features were important to them in their selection (for example, do customers who care more about ease-of-use choose product "A" over product "B"?);

• How many products a customer looked at before purchasing their selection;

• How long they spent making their decision;

• Whether they had researched in a different channel prior to purchasing in this channel.

Analysis on these data might show trends (for example, customers who purchase high-end digital camcorders tend to focus more on user features and product capabilities, whereas customers who purchased lower-end models focused more on size/weight).

Insights and actions that would follow could include merchandising products based on price category; demonstrating key feature differentiation at the high end; and focusing on in-store marketing that highlights size/weight benefits/differences at the lower end. In some cases, merchants might decide to reduce or increase the product variety at one or another end of the price spectrum to be able to drive purchase behavior based on insights obtained in this process.

While measuring this kind of raw pre-purchase data is easy online, with today's cutting-edge technology it can also be accomplished in an offline environment. For example, innovations in interactive touch-appliances create compelling customer experiences, and importantly capture key customer pre-purchase behavior data.

If decisions are to be made on the basis of insights, ensure that the data collected, and the analyses performed, lead to the conclusions that will allow for meaningful business insights.

Find ways to accurately measure pre-purchase customer behavior in a multi-channel environment, and then apply a process to transform the data into actions that enhance the customer experience and improve your bottom line.