We use cookies. You have options. Cookies help us keep the site running smoothly and inform some of our advertising, but if you’d like to make adjustments, you can visit our Cookie Notice page for more information.
We’d like to use cookies on your device. Cookies help us keep the site running smoothly and inform some of our advertising, but how we use them is entirely up to you. Accept our recommended settings or customise them to your wishes.

PPC Agencies: The Roots Determine the Tree

No PPC agency is right for every advertiser. Andrew Goodman's recent SEL Post dividing the world of PPC practitioners into "Muddy Ones" and "Quants" was quite entertaining and accurately characterizes most firms in the space. We at RKG like to think of ourselves as "Muddy Quants" in Andrew's vernacular. Alan has the PhD in Stats, and...well...I'm muddy. We believe that for large, competitive, ROI driven accounts you must have both pieces. You need an advanced system to predicatively model data and execute bid changes automatically, because hand-bidding off of spreadsheets simply doesn't compete in 2009. At the same time, we know enough about numbers to recognize the limitations of the data and the need for human 'intervention' and control of the process. The predictive model has no way of knowing that there's a sale next week on Gibson Guitars or anticipating how that might impact performance; it has no way of knowing that you're out of Merrill boots sizes 9-11; or that a retailer has just gotten some co-op advertising dollars from Kohler; or that customers of certain types of products are much more likely to pick up the phone and call or visit the local store than the average spillover rate would suggest. Marketing is muddy, no matter how skilled one is at analysis. But not every advertiser is both ROI driven and competing a large, complex marketplace. How does your company think about advertising/marketing? Is the primary goal of PPC advertising to:
  1. Create positive awareness of your brand?
  2. or
  3. Generate maximum revenue within some acceptable ROI?
Many will answer "both", but in our experience only one of these dominates an advertiser's thought process. Agencies come with the same biases: either they're fundamentally direct marketers and live and die by ROI calculations; or, they're advertising agencies that focus on share of voice, brand awareness and creating positive experiences with your brand. If the principal objective is branding, you don't need the mathematicians/engineers, and likely you don't want them. When you ask "What sort of 'outside the box' ideas do you have for creating brand awareness?" the quants will scratch their heads and send you a spreadsheet pointing out why the changes you suggest would damage conversion rates, lowering revenue per click forcing bids down the page. Not what you wanted to hear, and, frankly you're asking number crunchers to paint a portrait. If, like our clients, you enjoy talking about hold-out tests, the incrementality of a marketing program, lifts necessary for offers to pay-off, and the cannibalization that occurs between different marketing programs, you will be miserable if you hire the advertising agency style of PPC firm. They will talk about metrics that don't matter to you like impression share, they will talk about buying cycles without supporting data, and most importantly the results will stink. At a gathering of agency heads I heard one say "(egads) Our clients are starting to ask us to separate brand from non-brand performance data, and (horror of horrors) some are even asking for keyword level performance data!?!" I almost fainted! (As Direct Marketers, RKG has been providing that data to clients since our inception, and indeed encouraging them to think through the consequences of overspending for branding purposes.) Agencies often claim expertise in both, but take a look at their "about us" page to see their roots. The roots will determine the type of tree. Those who aren't direct marketers by training do not build the right analytical systems, the data warehouses, or the algorithms and don't train their staff to eat, breath and sleep ROI. Advanced data modeling doesn't help every company that seeks ROI. Data modeling requires data. If an advertiser's niche is too narrow in scope or geography, the best mathematicians in the world won't be able to materially outperform the advertiser's own staff. An agency of muddy-ones will do every bit as well as the muddy-quants like RKG, and probably better than the algo-only firms. Finding the right PPC agency for your firm requires knowing yourself and whether your organization sees PPC as primarily an advertising vehicle, or primarily a direct marketing vehicle, and finding an agency that shares the same view.
Join the Discussion