In our last post, we discussed the concept of using key metrics to make sure your business does not overspend on marketing.

As an example, we calculated the profit per customer for a hypothetical kitchen remodeling business at $2,000 and decided we could spend some amount less than $2,000 to acquire that customer. But how much less? The answer to that depends on your tolerance for risk.

Let’s play with a spreadsheet and a hypothetical magazine ad for our business.

A local magazine targets homeowners and they charge $1,850 for a magazine ad that reaches 117,000 people. Some tiny percentage of those people might become leads by calling us or visiting our website — and then some small percentage of those leads might take action and become a customer.

After we run the ad and carefully track leads and customers, we fill out a spreadsheet like the one above. We see that 1/4 of a percent of the magazine’s readers (293 people) expressed interest in our kitchen remodeling service by calling, visiting our website, or stopping by the showroom. We can then calculate our cost/lead (the cost of the ad divided by the number of leads).

Let’s further assume that out of those 293 leads, we got three new customers. In essence, for our $1,850 investment in the magazine ad, we got three new customers (at a cost of $632.48 per customer). Note that the numbers above are both hypothetical and rounded.

On the surface, that seems like a decent deal. We pay $632 for a customer that’s worth $2,000. That leaves us a profit of $1368.

But…

There is a certain amount of risk that arises from this magazine — in the form of variable factors that could cut our results significantly from month to month. Maybe the magazine has a bad cover one month and its retail sales tank. Those 117K readers turn into 85K. Maybe our ad placement in the magazine isn’t that great. Or the article next to our ad depresses people. Maybe the stock market drops and fewer people can afford a kitchen remodel.

Is $632 too much? It depends on how risky you think your particular marketing program is. If you’ve advertised in this particular magazine for 6 months and your cost/customer never varies all that much, it’s probably not very risky to continue.

Personally, I prefer cost/customer amounts in the range of 10%-20% of profit. This provides more of a safety net just in case the program doesn’t perform as expected. So in our kitchen remodeling example, I’d try to find marketing programs that worked out to a $200-$400 cost/customer or less.

Just for fun, let’s try to run a Google AdWords (pay-per-click) search marketing campaign. We’ll make it simple and say that we buy the search term “seattle kitchen remodel” (since our hypothetical company does business in Seattle). As I write this, Google estimates that the cost per click (CPC) for that search term is $2.28. Let’s keep the same ad spend ($1,850) which translates to 811 clicks to our website (leads).

Now, everyone who clicks through to our website is a targeted lead. After all, they were searching for “seattle kitchen remodel.” But there’s always a certain amount of competition on Google, so let’s say our conversion rate from lead to customer is 3/4 of a percent (0.75%). That works out to six customers at a cost/customer of $304 — which is more in my comfort zone.

Professional marketers spend much of their time testing marketing programs — and now you see why. The goal is always to reduce cost/customer (customer acquisition cost), but maintain or grow the volume of new customers — and do this all in a reliable, predictable way day after day. This isn’t easy for a number of reasons (including imperfect tracking of customer source, for instance), but it is the only way you can avoid overspending on marketing.

In Part 3 of this series, we’re going to discuss start-up marketing costs. Stay tuned.