The agency billing model is brokeJuly 7, 2010
Agencies typically bill their clients by the hours they spend, not by the results they deliver. What’s wrong with this model?
We’re ultimately in the business of selling stuff. Billing hours for making things doesn’t necessarily mean we’re selling more stuff. So why are clients content to pay their agencies by the number of hours they work? And why are agencies content to bill by the hour if they’re confident they can sell more stuff more efficiently then their competitors?
Our industry is notoriously bad at connecting the dots between a marketing communications activity and a sale. John Wannamaker was famously quoted as saying “Half the money I spend on advertising is wasted; the trouble is I don’t know which half.” After 100 years we’re only slightly more equipped to answer that question.
Agencies and brand marketers work on the premise that marketing communications can create awareness, brand favorability, and purchase intent. Each of these can be measured to some degree through consumer surveys, but drawing a direct line to a sale is difficult.
Most marketers also have many marketing activities running simultaneously, so quantifying exactly what parts of their marcom is working can be a challenge. Marketers are also faced with deciphering what factors are driving the effectiveness of any given activity: The creative? The media? The tactics?
Over the past few decades we’ve made some progress in cracking the code of marcom ROI, but many of these techniques are extremely expensive or based on anecdotal evidence. Marketing ROI has yet to become a quantifiable science.
Should agencies throw up their hands and continue to bill by the hour? Not necessarily.
In the digital marketing world we can readily measure engagement. Here are few examples of rudimentary engagement metrics:
Comments, posts, reviews, retweets, likes/follows, email opt-ins, downloads, uploads, subscribing (email, RSS, podcasts, video series, etc.), links to, embeds/installs, user-initiated plays/replays, favorites/ratings, social bookmarks, page views, clicks, game plays, time spent, poll participation, entries (contests, sweepstakes, etc.), coupon prints, and product demos.
Marketing researchers are getting a better handle how different types of engagement impact purchase behavior. And although we’re still years away from quantifying the exact value of a particular engagement activity, we have enough evidence to begin associating a value to an engagement.
I propose agencies involved in digital marketing efforts begin developing engagement models and attempt to associate a dollar value to each activity. If nothing else, brand marketers will have a better benchmark for measuring success, and who knows, agencies might end up with a better model for billing clients.