You may have noticed by now that we believe that the ability to gather and analyze data is a vital part of building a strong marketing function. You have to be careful how you manage this, because it doesn't take much to overwhelm a sales force so that they're spending more time collecting data than selling!

Every company, however, needs a customer database. In fact, every company HAS a customer database, even if it's just their account receivable records. (You'd be amazed at what you can learn about a business just by analyzing customers' transactions over time.) But regardless of whether you sell fast food or fighter planes, it behooves you to understand everything you can about your customers and prospects and that means gathering and analyzing data about them.

Any company with many customers and/or many transactions should keep a customer database. Current and former customers are often your most likely source of additional business. In addition, whomever has chosen to buy from you over time has "self-selected" as a target customer and you can use this information to determine additional prospective customers: Your new customers will most likely look a lot like your old customers.

Other details:

• Almost every company will benefit simply by taking its current customer list, "appending" data from an outside list like Dun & Bradstreet or infoUSA and using this information to "segment" their customers. You can append all kinds of data, including company size, SIC code, distance from them to you, # of employees, # of PC's they own, gender of the owner, and much, much more. When you are trying to figure out who to target as prospective companies, this is highly useful information.

• You are usually better off having a third-party help you build your customer database. There are many of these agencies around and they do a great job helping you "de-dupe" and "household" your data. "De-duping" is when you identify duplicate accounts and combine them. "Householding" is when you determine how accounts are related to each other. For example, you might determine that "ACME Manufacturing" is the parent company of "ABC Rivets" and "Cogswell Gadgets." This will also help you determine that "ABC" and Cogswell are "siblings" (two companies owned by the same parent.

• There is always a trade-off in householding and de-duping: If you want greater "matches," then the algorithms get very aggressive; JB & Sons, for example, will be combined with John Baker & Sons if they share a similar street address. However, this will lead to making some assumptions that later turn out to be inaccurate. On the other hand, if you set the criteria much higher so that the accounts only "match" if the phone number or email are identical, then you will have fewer duplication errors but more accounts you market to as separate entities that are actually a single customer. The same logic applies to de-duping. An experienced database management agency can help determine the right "tightness" of the matching.

• Your customer database shouldn't be housed in your operational data warehouse and it shouldn't be managed by your IT group. It doesn't NEED to be in your operational data warehouse because you only update it once or twice a month for marketing purposes. The IT group will tend to be focused on whether the database is functional and being used to prepare reports - not on the vital data completeness and accuracy you need to make great marketing decisions and measure programs precisely.

• Many companies think they are too small to benefit from using a customer database wisely. That's simply not the case. Even small businesses like salons and restaurants can benefit from gathering names, logging them into an Excel file and doing some basic analysis. It's amazing what you can accomplish with a simple geo-mapping software package like Microsoft's Mappoint, some data analysis you can do for free on www.census.gov and a couple of high school kids leaving door hangers at houses in neighborhoods that have demographics similar to your current customers'.

• Every business starts out data-dumb. Your long-term competitive depends on you managing your marketing data more effectively over time.

• The old saying, "There is no substitute for experience," is only partially correct. Combining experience with data always yields a more sophisticated and useful understanding of your business' realities. Sales people, for example, are too-often ignored by senior managers during strategy exercises. Field input should certainly be a vital input into key decisions. However, every individual only sees the marketplace from his or her own perspective and develops paradigms accordingly. Combining disparate views and opinions with robust data will always yield the most accurate view of the marketplace.

• CEO's are often terrible sources of market input. Not only is everything they see carefully "scrubbed" sterile by people with their own agendas, even their customer interactions tend to be artificial and not representative. If your CEO wants to visit customers, be sure she visits a lot of them - many CEO's meet one customer who thinks the company should launch a line of thingamajigs and next thing you know the whole company is investing in them even though there are countless more important things to do.

• Question data. Don't skip through tables and charts in reports and just read the text. Stop at presentations of numbers and analyze them carefully — pull out your calculator (or start up Excel) and re-do the calculations. Often, the numbers either have errors in them or there is some "spinning" going on that makes them very biased.

• Make sure anyone who is going to do market research or analysis for you explains carefully the methodology they are going to use. Then when you are looking at the results, make sure you're seeing the raw data - the actual questions asked, the exact answers, etc. If you don't know how good research is done, either hire someone who does or get trained on it yourself. This is the information you are using to steer the ship.

• Like most things in business, proper research and analysis is a game of inches. You rarely have some huge "AH HA!" insight that illuminates some treasure trove of opportunity. By implementing rigorous research and analysis, reviewing it carefully and constantly adding to your learning, you will continuously make slightly better decisions than your competitors. Over time, this can lead to a big lead in profits or market share.

• Categories of Research
• Market Analysis and Measurement
• Observations on Research