SEM Arbitrage: Golden or Cooked Goose?

One doesn’t often hear the word “arbitrage” used in marketing and advertising environments. It’s more common in finance. NASD defines arbitrage as:

The simultaneous purchase of a security in one market and the sale of it or a derivative product in another market to profit from price differentials between the two markets.

In search engine marketing (SEM), arbitrage is rampant. Instead of securities, it revolves around differences between the cost of search traffic and the payout a client is willing to make, either on a CPC basis or a return on investment (ROI) metric (e.g., cost per lead, cost per sale, or commission per sale).

Earlier this week, my ClickZ colleague Fredrick Marckini zeroed in on one facet of search arbitrage, the affiliate channel. Affiliate-based SEM arbitrage is by definition self-funding. The affiliate funds a marketing campaign in the hope its payout will be large enough to cover campaign costs and leave a tidy profit (the arbitrage spread) for his trouble.

Fredrick recounts how his views on affiliate arbitrage changed from negative to positive. He makes the point search engines appear to turn a blind eye to paid search results that contain both a merchant and its affiliates.

I wouldn’t count on that continuing indefinitely, particularly if results start to look like a blockade (three or more links to the same advertiser, with several from affiliates). Marketers with generous affiliate payment plans may see a situation in which affiliates have significant incentive to engage in arbitrage. Run a test or model to indicate whether the marginal profit from affiliate arbitrage is positive, taking into account increased CPCs and cannibalization from your listings (natural or paid). If it works, by all means, crank up the affiliate arbitrage machine.

Some marketers rely heavily on affiliate arbitrage. They set limits on the highest ROI keywords (e.g., their trademarked brand names). eBay recently stopped affiliates from bidding on “eBay” and several other trademarked terms. In such cases, the incremental cost of paying affiliates a handsome arbitrage (due to high brand ROI) isn’t justified.

I’m not sure I’d give affiliate arbitrage the thumbs-up in every case, particularly if the merchant doesn’t engage in SEM. Let’s look at reasons arbitrage may not be a good fit for you.

Risk-Averse Marketers

Say a marketer is risk-averse and doesn’t want to put his own capital in a paid search program. He doesn’t feel competent managing a complex search campaign. Affiliates, then, may not be willing to take a risk on an arbitrage opportunity.

In SEM, arbitrage is very risky for the affiliate because conversion data are scarce. How much capital would most affiliates devote to arbitrage? Marketers know a huge amount about their own businesses. If the marketer’s unwilling to risk going into search, why would an affiliate take the risk? At least a marketer running search himself has some organic search traffic conversion data to go by; the affiliate has none.

As smart affiliates want a reasonable return on SEM arbitrage, they’ll likely be ultraconservative or pass up opportunities altogether unless they seem like sure winners (low CPC). A risk-averse marketer leaves the market open to a competitor who understands search is a zero-sum game.

Affiliate Program Execution

Most affiliate programs pay a percentage of revenue to affiliates. This percentage is set to reflect the average a merchant’s willing to pay to acquire a sale. Merchants don’t know how much of that commission affiliates will use as a target return on advertising spending (ROAS) for their activities. Marketers who rely on affiliate arbitrage don’t have control. They don’t know what keywords are used or which work.

An added problem is any uniform ROAS or ratio objective for search fails to take into account differences in margin or lifetime value. There’s also no guarantee affiliates will choose the “right” keywords. For an electronics store, an affiliate may elect SEM arbitrage on high-priced computers. The merchant may have a low profit margin on the product, resulting in disproportionately low profit (possibly even negative return) on that affiliate’s activity.

Efficiency Wins in the Long Term

If you encourage your affiliates to do search arbitrage, at least conduct your own search marketing. If your affiliates out-bid you, chances are you’re leaving profit opportunities on the table. You have more margin to play with than they do, and you can make money on every sale. If search engines continue to permit SEM affiliate arbitrage, evaluate incremental profit against any loss in profit when setting policies.

Arbitrage isn’t just an affiliate marketing phenomenon. Many search agencies, interactive agencies, and lead-generation firms have a “don’t worry about the details” attitude with their clients when it comes to paid search marketing. The challenges in these cost-per-order/-action or ROAS arbitrage relationships are similar to the affiliate model. When any third party manages an arbitrage campaign, it sets its own profit levels and determines how aggressive to be in the marketplace.

When a scarce asset is in an auction market, the long-term winners are efficient marketers who can afford to play. They have data to support every facet of their search investment. The more competitive search marketing gets, the less arbitrage. Efficient marketers will lead, grabbing high-volume, profitable traffic and controlling their search marketing destinies. Where will you be?