Kevin Hillstrom: MineThatData

Exploring How Customers Interact With Advertising, Products, Brands, and Channels, using Multichannel Forensics.

June 03, 2009

Case Study: Back To My Office To Run Some Numbers

Whenever you work on a Multichannel Forensics project, it is a good idea to "take the temperature of the room".

At Gliebers Dresses, we'll have success if we can find a solution that gets people in the room to work together.

This doesn't mean that we lie about the results. Absolutely not. We will be 100% accurate in the reporting of our findings.

But we will find common themes in the data that unify the room. We will find cases where catalog marketing is absolutely essential to the success of the business. We will find instances where we can reduce advertising cost, instances that don't hurt sales much but do increase profit.

Lois Gladsone is the Chief Financial Officer. She is motivated to reduce unprofitable expenses. She seems to have come to a pre-determined conclusion, that customer acquisition is unprofitable. So my job is to validate or disprove her assumption.

Meredith Thompson is the Chief Merchandising Officer. She needs to know that customers love her product. One way to expose a customer to merchandise is to put expensive catalogs in the mail to customers who might be interested. There are many other ways to expose customers to merchandise.

Sarah Wheldon is the Chief Marketing Officer. Clearly, her teammates feel they can perform the role of the Chief Marketing Officer, and that suggests that her teammates do not have confidence in her skills --- that they may even believe that the struggles of the business are her fault. We have to give her tools that demonstrate that her thought process is on a different level than the rest of her staff. Boris Feldman's year-long test represents that opportunity.

Glenn Glieber is looking for fifty two-percent solutions. My job is to give him a series of solutions!

Everything that the management team looks at is historical in nature. I have to present to management a vision of the future. This is done by illustrating the future trajectory of the business, assuming they continue current business practices.
  • We know that 61% of telephone buyers purchase again next year. So we multiply last year's telephone buyers by 61%, then we distribute these customers across telephone and online channels. This process is repeated for the online channel.
  • New customers are added to the telephone and online channels.
  • Now we repeat the process for year two, then year three, then year four, and then again in year five. This is a classic application of a Multichannel Forensics project.
We'll end up with a five year forecast for the business. Based on forecasted new customer counts and marketing spend, we can project the next five years. Here's what the next five years look like, based on my spreadsheet model (here's a sample spreadsheet --- different data).
  • Year 1 = $62.8 million demand, $0.3 million profit.
  • Year 2 = $62.8 million demand, $0.3 million profit.
  • Year 3 = $63.0 million demand, $0.3 million profit.
  • Year 4= $63.5 million demand, $0.4 million profit.
  • Year 5 = $64.1 million demand, $0.4 million profit.
In other words, this business is stuck. Only a significant increase in merchandise productivity, or a change in customer acquisition practices will cause this business to move. Nobody will be happy with this level of performance. People lose their jobs when a business stalls like this, especially the Chief Merchandising Officer and the Chief Marketing Officer.

Ok, let's try something. Ms. Gladstone believes that customer acquisition activities are unprofitable, and are hurting the business. So I go back to my spreadsheet, and I crunch the numbers, cutting back customer acquisition activities by 50% for each of the next five years. Here's the result:
  • Year 1 = $53.8 million demand, $1.5 million profit.
  • Year 2 = $46.9 million demand, $0.4 million profit.
  • Year 3 = $43.2 million demand, -$0.2 million profit.
  • Year 4= $41.5 million demand, -$0.4 million profit.
  • Year 5 = $40.8 million demand, -$0.6 million profit.
Oh goodness. If we follow the advice of Ms. Gladstone, everybody will be pleased after one year of following her prescription. Sure, the business gets more than 10% smaller, but profit increases somewhat. You can only imagine the glee in the Executive meetings, as everybody celebrates how they "fixed" the business. But look at what happens over time. The business is a third smaller, and is unprofitable. The finance, merchant, and marketing leaders wouldn't survive this level of deterioration. In fact, the business may not survive this level of deterioration.

It turns out that Gliebers Dresses needs to acquire ample quantities of new customers to be successful in the long term. Without a healthy customer acquisition program, Gliebers Dresses becomes an irrelevant brand.

I will always run the opposite scenario. I use my spreadsheet models to show what happens if we acquire 50% more new customers than what we are currently acquiring.
  • Year 1 = $70.3 million demand, -$0.6 million profit.
  • Year 2 = $76.3 million demand, $0.3 million profit.
  • Year 3 = $79.9 million demand, $0.8 million profit.
  • Year 4= $82.3 million demand, $1.1 million profit.
  • Year 5 = $84.1 million demand, $1.3 million profit.
These simulations illustrate how sensitive a business is to customer acquisition. If Gliebers Dresses is willing to make an investment in the future, then long-term profit significantly improves. But the business isn't in a position right now to invest another three million acquiring customers that will result in a short-term loss.

None of the scenarios gets Gliebers Dresses back to printing money like in the middle of the decade, a time when Gliebers Dresses routinely generated between five and six million in earnings before taxes.

We're midway through a Multichannel Forensics project. I've been able to demonstrate to myself that Gliebers Dresses should not cut back on customer acquisition activities. A long-term test indicated that there are opportunities to reduce advertising spend among existing customers.

In the next week, my approach will include the following tactics:
  • Explain the customer acquisition issue to Management. This should be a stimulating conversation.
  • Research the long-term mail/holdout test, evaluating how e-mail, catalog, and search customers behave when advertising is withheld.
  • Identify an optimal e-mail and catalog marketing contact strategy for each customer segment.
  • Quantify the five year profit and loss impact of my proposed strategy.
  • Quantify what happens to the business when the economy improves --- can the business grow if productivity improves by, say, 10%?
Ok, your turn. What is missing here? What are the pitfalls, given the personalities of the Management team? How do you think people will respond to what has been learned about acquiring new customers?

Labels: , , ,

0 Comments:

Post a Comment

Links to this post:

Create a Link

<< Home