May 28, 2007

How Much Do I Spend On Online/Catalog Advertising?

Lands' End was a fun place to work in the early 1990s. There were a lot of interesting minds, tossing around interesting ideas.

One of our debates was about the optimal level of advertising spend. One camp, led by our Circulation Director, believed that you circulate to an incremental 7% pre-tax level (prior to subtracting fixed costs). The theory was that the return on investment had to be sufficient to cover fixed costs ... that if you actually subtracted fixed costs from the equation, you were circulating to about break-even.

Another camp believed that you circulated to -5% pre-tax levels, because this way, you were capturing long-term profit that you were losing in the short term. At the end of five or ten years, your business was much bigger, because you acquired/reactivated a lot more customers than in the situation where you maximized short-term profit.

At Eddie Bauer, we circulated to break-even (prior to subtracting fixed costs), then shifted our strategy to invest to below break-even, in order to maximize the long term health of the business.

At Nordstrom, we tried our hardest to convince folks to invest in online marketing activities that maximized the long term health of the total business. We probably under-invested in the online channel, though we had the data to tell us what the 'right' thing was to do. The process of assigning a marketing budget did not provide us the flexibility to maximize the online channel (and ultimately, to grow store sales). This is a good lesson --- it doesn't matter what data you have, there are internal processes and existing cultures that simply cannot be changed.

In the past, we didn't have the right tools to understand the long-term impact of short-term advertising decisions. With Multichannel Forensics readily available these days, we can simulate different strategies, and identify the best long-term strategy.

I crafted an online/catalog business simulation, and ran three scenarios.
  • Scenario #1 = Maximize profit each year.
  • Scenario #2 = Maximize total profit over the course of five years.
  • Scenario #3 = Maximize profit five years from now --- make that year as profitable as possible.
The table below show the results of the three simulations. All numbers are listed in millions:

Maximize Short-Term Profit

Demand Ad Spend Profit
Year 1 $44.6 $5.6 $2.1
Year 2 $42.0 $5.2 $1.7
Year 3 $40.9 $5.1 $1.4
Year 4 $40.4 $5.0 $1.2
Year 5 $40.1 $4.8 $1.1
Totals $208.0 $25.8 $7.4

Maximize Long-Term Profit

Demand Ad Spend Profit
Year 1 $59.2 $9.9 $1.5
Year 2 $66.6 $11.0 $2.0
Year 3 $70.6 $11.6 $2.3
Year 4 $72.8 $12.0 $2.4
Year 5 $74.0 $12.2 $2.4
Totals $343.2 $56.7 $10.6

Maximize Only 5th Year Profit

Demand Ad Spend Profit
Year 1 $66.4 $12.5 $0.6
Year 2 $80.3 $14.9 $1.6
Year 3 $88.6 $16.3 $2.2
Year 4 $93.4 $17.1 $2.5
Year 5 $96.3 $17.6 $2.6
Totals $425.0 $78.4 $9.5

Let's review each simulation.

In the first run, profit is maximized by year. Therefore, profit in the first year is $2.1 million. However, a much smaller business exists going into year two, with too few customers to generate large volumes of profit. Still, the management team tries to maximize profit in year two, then year three, year four, and year five. As a result, this business actually contracts. If we followed the rules of Wall St. (maximize short term profit), we may not protect the long term health of our business.

In the second case, online/catalog advertising spend is more than twice as much as in the first simulation. This means the business is more profitable in the long-term, and grows at a much faster rate.

In the third case, online/catalog advertising is fifty percent more than in the second case. This yields a marginally profitable business in year one, but in year five, the business is much larger, and more profitable.

For every online/catalog business, these scenarios can be easily created. The multichannel analyst provides management with three or more scenarios (as outlined above), and lets management determine the future trajectory of the business.

This is an important point --- abstract and geeky topics like lifetime value have little or no meaning to executives. Picking from one of three possible strategies is easy to do if you're an executive, and accomplishes the exact same thing as a geeky, technical lifetime value analysis.

Multichannel CEOs and CMOs: Simulations indicate that it is important to invest in unprofitable customer activities in the short term, in order to protect the long term health of your business. It is important not to focus on "this year". Where possible, invest in the short term, to protect the long term health of your business.

Profit per New Customer

It's common for folks to measure cost per new customer. Total Marketing Cost = $10,000. Total New Customers = 130. Cost per New C...