Aug 272009

PPC Agencies Promising/Forecasting Results

Oftentimes when speaking with a potential client we're asked to predict how much improvement we can make in their program. Our answer is always the same: "We don't know."

We thoroughly understand why prospective clients want to know: we all want to know in advance what the impact of a change will be on ROI. What I don't understand is why they'd believe what we tell them.

While we can usually guarantee a prospect that we can hit their competitive search efficiency targets (unless the volume of search is too small), we can't project what the impact will be on their top line.

If we have good sample data from them, we can often speak in detail about what their current problems are -- thin keyword lists, poor bidding, bad use of match-types, poor landing page choices, whatever the case may be -- and what we'd do to address those problems. We don't know what the impact of those fixes will be. Indeed, we can't know.

Three factors determine the size of the opportunity:

  1. The Volume of Search. PPC responds to demand, it cannot create demand. The company providing bleeding edge service that no one knows you can outsource is going to have a hard time with search.
  2. Relative Conversion Rates. If competitors are more effective at converting visits to sales due to selection, pricing/offers, name recognition, or site design it will be difficult to generate traffic cost effectively.
  3. The Efficiency Targets. Even if a company competes well, if its competitors are willing to spend a larger percentage of their revenue on marketing than our prospect can, we will have a hard time competing for prominence.

We can't gauge any of these factors, let alone all of them. Even if we could know all these factors historically it wouldn't necessarily allow us to predict the future in each area. After all, how many companies out there are hitting their monthly forecasts for 2009?

Many agencies are more than happy to make bold predictions -- some without even seeing any data! -- but what mystifies me is that advertisers actually listen. Some agencies even tout their advanced, highly scientific, very very complex algorithm for predicting the future. Hokum. They can't make these projections any more accurately than we can.

We've lost out on many RFPs because we wouldn't predict the lift our services would create while the winning agency could predict a 73.4% increase in the top line and 22.7% increase in the bottom line, or whatever.

More than a few of those advertisers have come back to us after a year, because the other firm couldn't deliver on their promises.

I don't really blame the agencies for making the projections. We've been strong-armed into writing down a number from time to time. Because it's so often demanded, we even kicked around the idea of building a fancy looking crystal ball calculating machine that would: take inputs, chug impressively and belch out an answer. We didn't do it; too disingenuous for our tastes.

I do blame the advertisers for a) asking for something no one can give them; and b) actually paying attention to the responses. The axiomatic old joke:

Q: How can you tell when a salesperson is lying?
A: Her/his lips are moving.

is not too far from the truth. We try to be different, and I think we are, but why would anyone believe that?

Selecting a vendor should be based on a good match between the agency's tools and knowledge and the advertiser's needs. Their performance should be judged based on their reputation and careful questioning of their references. The reference calls can be very revealing. Every agency will give you references from clients who love them, the difference is whether the reference knows their stuff. Good agencies make the sharp clients happy, bad agencies make only the uninformed happy.

Choosing a vendor based on who makes the biggest promise is a recipe for disappointment.


12 Responses to "PPC Agencies Promising/Forecasting Results"
Josh Winkler says:
IMHO... Management and marketing departments ask for projections because that is what they are used to seeing from other channels. I'm solely a web marketing person, so I could be wrong, but from the little bit I've seen of traditional marketing in a few different companies, someone makes a hypothesis of what the results will be and then everyone else runs the day to day around that hypothesis and you wait weeks and months to see if they were right. With search, the data cycle is so much shorter, it makes it easier to hit the desired goals or shift your internal momentum if the data is trending to either extreme. Because of that, it's very hard to have a truly detrimental online marketing campaign. You either have to make the choice to let things go that far south to test the waters...or you just aren't paying enough attention to the data. Maybe what companies are really asking for in lieu of projections is an assurance that whatever agency they choose will be paying attention and will react when things begin to go south in whole or in part. I know I've dealt with agencies in the past that were pretty much asleep at the wheel. "Set it and forget it" types who then left the data checking to their clients or checked in once a month to make "tweaks" to the campaigns. Nowadays, I find myself beginning to take more of the set it and forget role, as I know I have an agency who's not asleep at the wheel. And if references should be the true measure of performance, keep my number handy, because you guys are great!
Josh, you're a sweetie! Thank you for the kind words. I think you're right, projections and forecasts are more important for budgeting and cash flow when the lag between investment and return on investment is long. In search, it's not just short, it's the other way around. You get the sales before you have to pay for the clicks! The consequences of missing forecasts in cataloging are large because you invested based on the predicted ROI. The fact that it matters doesn't make any of us any better at forecasting, just makes it more stressful.
Tom Hale says:
Excellent! So concise. But how often have you given this explanation to a prospect and still gotten the "quit the song and dance and give me a number" response from some hard driving type that thinks "nuance" is something foul? Making sure you understand that they are in this to make money, in case you were under some other impression of course. grrrrrr The truth can't always fit on a bumper sticker or Tweet. This is the truth, about as bare bones as you can get and still qualify as truth in my book. Well done. -T
Thanks Tom, We've gotten that response many times. Usually they end up going with someone else; either we just refuse to give a number, or we give a very conservative estimate that we're pretty sure we can beat easily -- we prefer to under-promise and over-deliver than vice versa. As a result, these types of prospects usually pick another firm. That's probably good for us in the long term. Our business model is based on retention and word-of-mouth marketing. Bringing in clients on false premises is bad for both retention and word-of-mouth. The folks who make decisions based on foolish projections probably aren't going to "get" what we do for them, so in some ways it helps us vet prospects. Expectation management is good for all parties concerned.
3. The Efficiency Targets. Even if a company competes well, if its competitors are willing to spend a larger percentage of their revenue on marketing than our prospect can, we will have a hard time competing for prominence. Me: If I comprehend correctly, this above statement is conveying a message that if they do not compete in marketing dollars, it almost definite they will not be able to compete. How do others understand? In my opinion, it is not only the amount you spend. However, the more efficient program with a higher ROI is prefered. This being a program that delivers an appreciating asset. In this case the asset being the investment and the marketing program that you put in place.
Tony, you raise an excellent point. If Acme's arch-rival Apex is willing to outspend them in marketing does that really mean Apex is "winning"? Not at all, and I didn't mean to suggest that spending more at lower ROI was what it takes to win. We're big fans of letting the deep pocket folks spend themselves into Chapter 11 while our clients grow profitably. The point of the whole piece is: we can't predict what Acme's sales are going to be through paid search because it depends on what Apex and each of their other competitors does with their marketing targets. Targets that can and do change. If a number of Acme's competitor's get more aggressive in Q4, Acme's sales will dip through no fault of their own.
Tim says:
This post is spot-on! I don't know why agencies are so afraid of setting realistic expectations from day 1 with clients. (Well, actually I do know why, and it's out of fear). Even further, I've had numerous clients that have a particularly hard time accepting the fact that SEM campaigns do NOT drive demand - all they can do is be visible for relevant searches. And that is directly responsible to what is happening in other advertising channels. I've had numerous conversations with (some very smart!) clients wondering why their search programs have leveled off, at the same time they've been cutting their TV or radio advertising. Even worse is when your agency colleagues ask the same questions. Thanks for this post! Nice to see other people are into transparency and setting reasonable expectations :-)
Thanks for your comment, Tim. We learned very early on that if the clients don't get it the relationship is going to be tough. Expectation management is key to long relationships.


Check out what others are saying...
[...] Pay Per Click (PPC) company writes about the dangers of predicting results for a PPC campaign. Link Here. The bottom line is you do not know what the results will be before a campaign starts. However, you [...]
[...] PPC Agencies Promising/Forecasting Results – view page – cached Neither we nor anyone else can predict the future, so why do people keep asking? — From the page [...]
[...] PPC Agencies Promising/Forecasting Results, Rimm Kaufman [...]
[...] PPC Agencies Promising/Forecasting Results, Rimm Kaufman [...]

Leave A Comment