Edit ModuleShow Tags
Edit ModuleShow Tags

How to mine data properly from multiple sources


Published:

Mark Siebert

llustration by Jonathan Hankin

People have been digging for gold for centuries—some speculate for more than 7,000 years. The discovery of gold in California in 1848 led some 300,000 people to leave their homes and seek their fortunes despite the risks. And while there were certainly those who made their fortune when they lucked into a find, far more found their fortune through a systematic approach.

Like “the 49ers” of old, when we find a franchisee in a particular place, our first inclination may be a headlong rush down the same path, with the hope of replicating the results.  

But as much as we would like to believe otherwise, there is no silver bullet. Instead, franchisors need to know how to better mine (and understand) the data they receive from their lead generation activities to determine where they might find their next strike.

Is it played out?

Like gold, the number of people who are willing to quit their jobs and invest their life’s savings in a franchise represent a very limited resource. No matter how fruitful a franchise marketing channel has been, you don’t want to be the last prospector to show up in California.

Consider this hypothetical: You spend $1,000 per month to advertise in XYZ Magazine. A year later, you determine that your cost-per-lead from XYZ Magazine is running $108—a figure you deem to be quite strong. Should you continue your commitment for another year? While your first reaction might be to stay the course, you need to do a deeper dive.

Some kinds of subscription-driven advertising, as an example, will show diminishing returns over time. This is because the incremental impact of increased ad viewership (often called frequency in the advertising world) diminishes over time, while the majority of the subscriber base remains fairly constant (often called reach).   

The theory of marginal analysis invites us to take a closer look at that data to see where things stand in the present. Perhaps when the ad was fresh, you received an average of 10 leads per month—but the ad’s effectiveness may have diminished as it was continuously pumped to the same audience over and over again. What if, for example, your monthly lead yield looked like this: 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 4, 2.

Looking at your leads more closely, you can readily observe that your lead count for the last several months had fallen dramatically. So, while your overall cost per lead looked great at $108, your marginal cost per lead in the most recent month has skyrocketed to $500—providing you with a very different view of media effectiveness.  

Fool’s gold

Another problem when attempting to determine marketing effectiveness is substituting activity for effectiveness. I once analyzed the efforts of a franchise system that spent over $15,000 on a web portal. The sales representative had promised them a flurry of leads—and, in fact, the site delivered plenty of leads. The problem was that only a tiny fraction ever responded to the first follow-up call.    

Early successes can exacerbate this problem. From a statistical standpoint, assume that you were trying to draw a black marble out of a hat that contained only one black marble and a thousand white marbles. Your odds of selecting the black marble are the same on your first and last attempt. But, if you managed to draw the black marble after only a handful of tries, you might be inclined to throw money at that tactic forever. Conversely, if the same black hat held 30 black marbles (the equivalent of a 3 percent close rate), but your first a50 draws came up white, you might be inclined to close your books on that lead source forever.

This is where the Law of Small Numbers comes into play. Essentially, the Law of Small Numbers involves the judgmental bias that can creep into decision-making based on a limited number of observations. The problem is that in order to avoid this bias, a franchisor would need to gain enough data from each media outlet to provide it with statistically valid conclusions.

Sink another shaft

In order to combat this, some veteran franchise systems may look to mine the data from their current franchisee base to determine if there is any correlation between specific media and their most successful franchisees. The problem here is a fallacy of composition. This type of analysis will always lead you back to what has been successful in the past, and it will never lead you to newer, more productive alternatives.  

Newer franchise systems have even more difficulty. They don’t have the benefit of years of data to tap into. Moreover, salespeople besiege them and salt the mines with success stories illustrating why their particular media will prove the most effective.

In order to avoid these issues, franchisors need to get more creative when collecting data and need to add more rigor to their data analysis.

Franchise systems, from the emerging to the established, would do well in looking for data in places they may have never looked before. Consulting firms and advertising agencies may compile data from dozens or even hundreds of franchise clients. Alternatively, franchisors can share data with non-competing franchise systems through the use of forums. 

Franchisors can also use the research tricks that consultants and agencies use by conducting primary research, perhaps interviewing the franchisees of competing franchise systems, talking with sales representatives at trade shows, and reading industry-specific research on what is working and what isn’t.

Ultimately, there is no single solution to franchise lead generation.  But if you systematically mine and analyze data—and use that data to continually refine your lead generation strategy—you are much more likely to hit the mother lode.

Mark Siebert is CEO of consulting firm iFranchise Group. Reach him at 708.957.2300 or info@ifranchisegroup.com. His new book is “Franchise Your Business: The Guide to Employing the Greatest Growth Strategy Ever.”

Edit ModuleShow Tags
Edit ModuleShow Tags