As a marketer, you have to learn to predict ROI, not measure it after the fact, and that takes some major fuzzy logic. To start with, refuse to accept the common lies of the industry, and you'll be able to make solid decisions.
The genius of Dan Bricklin‘s invention, the spreadsheet, is its simplicity. That is also its greatest trap.
Measuring return on investment (ROI) is relatively easy. Though you may miss hidden costs or value created, in general you just figure out your gross margin, subtract the cost of advertising and marketing, and see if you end up in the black or in the red.
Most of the time, however, you can’t afford to wait to measure ROI. You have to make decisions before you make a buy. You’ll never have enough money to test every possible channel, and that means making some tough decisions before you get a single ounce of hard data.
I’ve spoken with a lot of ad salespeople in my time, and they’ll talk your ears off about their huge response rates. Only once has the response rate ever been close. The fact is, you can’t take anyone’s word at face value. Nor is reputation necessarily a good measure. Sometimes, even a company with the best reputation in the industry (say internet.com, the parent of ClickZ) can’t deliver the same ROI as a goofy little list that you find on Ezine Ad Auction.
As a marketer, you have to learn to predict ROI before you measure it, and that takes some major fuzzy logic. In other words, you have to take all the different factors into account and make an educated guess.
But just because it’s difficult to build an accurate-to-six-significant-figures spreadsheet, doesn’t mean you should give up. Refuse to accept the common lies of the industry. If you remember to take these fudge factors into account, you’ll be able to make solid decisions — even if you’re off by a few decimal points.
1. “Of course we’ll hit the numbers on the IO.”
Just because your insertion order (IO) specifies 100,000 clicks, don’t assume that’s actually what you’re getting. The online advertising industry often applies what I can only describe as non-Euclidian mathematics to its metrics. I can remember being told by several leading ad-serving companies that a loss of 20 percent of theoretically available impressions was standard for the industry. Can you imagine if the healthcare industry saw 20 percent as an acceptable loss rate for patients?
Of course, imaginary numbers can work in your favor as well. Technological inflexibility sometimes forces publishers to overdeliver, just to be sure of fulfilling your IO. As a wise marketer, you can take advantage of that fact. Simply inform the salesperson that you’ll raise unholy hell if you don’t get full delivery of your IO, and she’ll probably harangue her techs into overdelivering by 20 percent just to avoid running afoul of your lawyers! (Besides, those impressions represent funny money anyway — more on that later.)
2. “You’ll be reaching millions of inboxes.”
If a tree falls in the forest and no one is there to hear it, does it still make a sound? I don’t know, but I do know that if an email is deleted without being read, it often still counts as an impression!
The dirty little secret of the opt-in email industry is that many of the emails that actually reach legitimate inboxes end up being deleted without being read. Think about it — when was the last time you got a LifeMinders email that you actually read?
Before you buy, demand statistics on what proportion of a publisher’s emails are actually read. If you don’t, you may end up paying for dead electrons.
3. “Our click-through rates are through the roof!”
Many publishers take the Chicago election committee approach to measuring click-throughs. Click early, click often, and don’t look too closely if some of the clickers happen to live in the cemetery.
Just like the quality of a customer, the quality of a click is all-important. The fact is, unless the click comes from a qualified customer, it’s worthless. This means that you have to discount the value of the clicks you purchase depending on their provenance. If you’re buying clicks from a niche publisher that targets your ideal target audience, you can apply a low discount rate. If you’re buying clicks from a network that claims you can target your desired audience, discount them 50-75 percent. If you’re buying clicks from a network that allows incentives for click-through, you might as well flush your media buying budget down the nearest lavatory.
Of course, even if you take these common fibs into account and do all your homework, you may still run into publishers that take you for a ride. That’s why you need to follow the disciplined marketer’s first principle of ROI:
Never pay a publisher up front!
If you get terms, you can always respond to wretched performance by simply refusing to pay. I’m not advocating simply stiffing your creditors (that only works if you bought advertising from the World Bank), but delaying payment until you satisfy yourself that the publisher has acted in good faith. After all, you may have had unrealistically high expectations, in which case you should pay your bill and consider it your tuition to the school of hard knocks. On the other hand, if a publisher uses the three common lies and is trying to simply take your money, you have a right not to hand it over. Just don’t tell them that you got the advice from me!
