More Data Than Budget, Part 2

Melaney Smith

Updated · Nov 26, 2003

This column is the second in a series for data-starved, financially challenged marketers struggling to make decisions without access to critical data. We're looking at tricks for getting seemingly unobtainable data, no matter how low-tech or manual the process. In situations where you must make decisions without the necessary data to evaluate your options, do you flip a coin? I hope not! Below, more tricks from the trenches.

Unique URLs, With a Twist
First up is CipherTrust, an e-mail security company based in Alpharetta, GA. I spoke to Matt Anthony, director of marketing, and was impressed once again with marketers' ability to adapt. Three years ago, Matt was a one-man marketing department and needed every dollar he could muster for lead generation. “I needed information but didn't have the budget to purchase an analytics tool,” he said. “I flew blind for a little while, but how can you spend your marketing dollars effectively if you can't tell what's working?”

You can use Matt's solution. The first part is a low-tech, low-cost trick: a different URL for each campaign. “The pages look the same to the site visitors,” he said, “but embedded in the code for each page is a string that's unique to each campaign. I can get as granular as I need to: date, campaign vehicle, message focus, etc.”

Those of you familiar with the technique know all it requires is someone willing to keep track of the many Web pages and make sure they all reflect current information. CipherTrust added a twist.

“Visitors who were interested in our product would fill out a form to request a white paper, pricing, or other information,” Matt said. “When the person clicked the ‘submit' button, we not only sent the information requested, but I got an e-mail copied to my inbox. The e-mail contained the marketing source information… in a comma-delimited text string. At the end of the day, I'd just export my entire inbox to an Excel file.”

I like it. Yes, it's low tech. But when you open the Excel file — voil` — data is separated into meaningful columns with a row for each new lead. You can evaluate your campaigns. Matt's perseverance paid off. His method and the subsequent analysis it allowed laid a foundation for an automated SQL marketing database. “Our company has grown so much, and the Excel method obviously isn't scalable,” he said. “But when you're desperate for information, you have to get creative. It served its purpose and was worth the time and effort.”

Find an Ally
CipherTrust's method requires at least some Webmaster support. Which is a great lead-in for a tried-and-true no-to-low-budget strategy for the data impaired: Find an ally. The data exists, but you can't get to it, for whatever reason. Time to make friends in the IT department (if you haven't already). Find a potential ally who will listen to your plea. If necessary, determine if bribery is an option.

I'm not suggesting putting your job at risk by circumventing corporate procedures. But don't dismiss the power of friends in the right places. All it takes is one willing database admin to put you on your way to crunching some numbers.

Software Trials
Reader Adrian Salamunovic, president and found of Navasys Group, shared this trick for no-budget log-file analysis: “Most Web analytic providers have 30-day free trials of their software. Not worth it if you have to do any type of complicated integration, but if you're looking to crunch data and you literally have no budget, it will do the trick,” Salmunovic told me. “After a few months of doing evaluations, you'll also have a great idea of what's out there. So when you do manage to get a budget, you'll be aware of all your options.”

Now why didn't I think of that?

Home-Grown Surveys
The last low-budget trick is widely used, but still worth mentioning: do-it-yourself surveys. It's easy to build a survey, post it to a Web page or e-mail it, and collect results in a database. A recent client hoped to make improvements to his company's Web site but had no budget for collecting user input. The client was fairly certain he knew what users wanted (aren't we all?) but needed to prove his ideas and establish a method of prioritizing changes. All it took was a short list of questions and an open-ended “How can we improve this site?” question. In less than a day, a single developer was able to convert the questions to an HTML file and build a simple SQL table to store responses.

Caution: Research conducted incorrectly can be damaging. Use this strategy only if you're well versed in market research techniques.

It's not too late to share your low-budget tricks from the trenches. Click here if you have a favorite.

  • Data Management
  • News
  • Read next