We’ve heard of guerrilla warfare — whereby a small, irregular band of fighters can overcome a much stronger, more organized enemy. We’ve even adapted its tactics to marketing. But is there such a thing as guerrilla market analysis?
In today’s fast-paced world, traditional market analysis warfare — involving lengthy survey processes and hulking hard copy reports — may not be nimble enough to do the job. It’s time to “go guerrilla” with your market analysis.
Sources of data are everywhere, if you just look around, and electronic survey capability gives analysts a rich source of data to use in their ongoing efforts to understand (and affect) customer behavior. To quote Dr. Seuss: “Oh, the Thinks You Can Think.”
The goal of traditional market research is to be statistically significant, be unbiased, and provide an accurate assessment of conditions at a particular time. Unfortunately, that’s often not enough for the analysts and data collection groupies who get their hands on the results, especially those accustomed to the online world.
Fifteen years ago, the annual customer satisfaction survey was delivered to us in hard copy, with a report around 10 pages long and perhaps 30 to 50 pages of verbatim customer comments. I remember the day I asked the research firm to add the customer’s name to each comment and sort the comments by major category. The researcher looked at me like I was insane, and the price made me reach for my chair.
So we decided to do it ourselves, in Lotus 123, and we added pertinent demographic information while we were at it. We discovered a wealth of information in the raw data, even though some of this data was in the form of handwritten notes. At the time, I thought Lotus had revolutionized the way I could use survey data, allowing us to squeeze more return out of an expensive research investment.
Fast-forward to the early days of electronic surveys. The survey software was cheap and allowed us to send surveys to any customer with an email address. And the arguments started. The market research folks were suspicious. Yes, they could send the survey to a random, statistically significant group of customers and tally the results. But would using email change the percentages or makeup of the respondent group in a way that affected accuracy? For example, would a lower percentage of older customers respond since the online market was much younger?
Meanwhile, the analysts wanted to send the survey to every customer. Their argument: The data is so valuable, and how much more statistically significant can you get?
We ended up compromising, because each group had different goals. The folks in the market research group sent their survey to the random subset and prepared a traditional report. They were thrilled to see how much easier it was to incorporate demographic and sales data into the responses, taking their cross-tabulations to new levels of complexity. Additionally, the ease and low cost allowed them to do quarterly surveys rather than the once-a-year approach.
After the report was complete, the analysts were allowed to send the same survey to a larger group of customers. They created a survey database to store responses that could be used (and updated) all year long and incorporated into their normal analysis.
For example, the “Top 20 Customers” report could include not only sales history and fluctuations but also overall satisfaction and highest- and lowest-rated attributes by customer. As well, we could include many of their verbatim comments from the last satisfaction survey. The monthly “lost customer” report could include the same and provided important clues to help anticipate (and prevent) future lost customers.
In the end, the market research folks were able to protect the integrity of their research efforts while still providing the analysts with a new, rich source of data. The compromise left both groups thrilled with the leap forward.
In the end, electronic surveys gave our market research and analysis programs new life. Obviously, the researchers could do more research less expensively, with much quicker results. But the analysts usually found a way to get a few extra questions thrown in and saved the responses (even the verbatim comments) in the marketing database.
We found that in the electronic survey world, our customers didn’t care so much about remaining anonymous, and we could prepopulate the survey form with their account information. Over time, we came to know our top customers better than we ever thought possible in an offline world, and our standard monthly reporting became more customer-focused, rather than just a regurgitation of vital statistics.
This may be an old story to some of you. It came to me while reminiscing, as happens so often during this season. Still, not every e-marketer has broken free from the bounds of traditional market analysis warfare. Not everyone is a grizzled veteran who’s “been there, done that.” Those of you who started e-surveys long ago quickly realized what a gold mine the data could be. But new people are joining our ranks every day, and they’re fighting the same battles we fought to convince their superiors that technology is not so scary. I hope this column provides some ammunition to those of you who are fighting to justify an online research program.
Emily Ma, product director of Tencent’s advertising platform products department, was a keynote speaker at ClickZ Live Shanghai where she discussed the ... read more
In today's multichannel world how can marketers use data to ensure the experience a customer receives is relevant to them?
The terms that customers type into your site search function can help you to gain an understanding of user behaviour and can be used to optimise ... read more