Jul 82010

Misinformation and Disinformation

I wish I had a dollar for every hour we've spent trying to disabuse clients of wrong-headed notions they've picked up from conference speakers, blog posts, consultants, and sometimes the engine reps themselves. We don't blame our clients for this. The sheer volume of misinformation and downright disinformation spewed about paid search and online marketing makes it incredibly difficult for folks who don't live and breath the channel to separate fact from fiction.

In the days of yore, getting published involved having some level of expertise, and what was written had to stand-up to some level of editorial fact checking. Not so these days. Anyone with a blog can say anything they want, and publishing a "white paper" takes nothing more than someone skilled in design.

It is particularly distressing that some conferences allow charlatans to present as "experts" giving their imprimatur to the ensuing misinformation.

Here is a short list of myths that keep being repeated no matter how often they've been debunked:

  1. Bid management is all about finding the right position on the page. I heard one "expert" at a search conference say he was having great success with position 7! Salespeople on Google's vertical teams have told our clients that conversion rates in position 1 are better than elsewhere on the page, despite the research that we've done, that Hal Varian, chief economist of Google has done, and that other's in the industry have done proving that conversion rates are largely invariant with respect to position. One "expert" got so mad at me it was hilarious. She argued that doing such a study would require way too much data across hundreds of accounts and that the truth was different for every client...to which I responded "So, you acknowledge that determining the truth requires mountains of data across hundreds of accounts, but you think you can figure it out for each client looking just at their data?!?" She got kinda quiet after that :-)
  2. Account structure is key to Quality Score. False. Account structure has virtually no bearing on QS. The relationship between the Keyword and the copy matters. Once that history is established shuffling the account structure around has ZERO impact on QS. The folks on the Google agency team acknowledge this, the folks at Marin presented a good study at SMX Advanced proving this, but consultants and the Google vertical teams haven't seemed to catch on to that fact. It's such an easy thing to criticize that folks who don't know how to actually assess a paid search program immediately gravitate towards structure.
  3. Last touch attribution steals credit from general terms and gives too much credit to more targeted terms. We have debunked this buying cycle argument so many times its resiliency is surprising. Others have joined our chorus including EF and Marin, and frankly anybody who has the data and knows anything about statistics can see it for themselves. It's a great story, it just doesn't happen enough to talk about. Nevertheless, despite the fact that Google knows the truth, their vertical sales teams still give presentations and tout the importance of assists, failing to mention that the vast majority of those assists are clicks on the same ad!I've heard other people who know better say at conferences that they've seen "a 300% difference" between first touch and last touch conversions..."uh, by any chance was that a change from 1 order to 3 on a single keyword?"--(it was, I saw the data. Oh, and 1 => 3 is a 200% lift, not 300%)--" And, is your bidding system really so flighty that that kind of statistical noise impacts bidding greatly?"
  4. I could write a book on other malarkey spread by people who really ought to know better, like:
    • Multi-channel buyers have much greater lifetime value than single-channel buyers. It may be true, but I've yet to see anyone do the study correctly. The problem is they don't control for the fact that by definition you're studying people who've bought from you more than once. The way to do the study is to compare the lifetime value of folks after their second single channel purchase to folks who've ordered once on line and once in store. Anybody interested in the real number, or do we just want to keep spreading misinformation?
    • Opinion surveys pretending to be studies. Anybody see the hilarious "study" presented in multiple venues claiming that 33% of people who see a display ad respond by buying the product??? A 33% conversion rate off a display impression! Wow. How do these people sleep at night?
    • For every online order 7, 8, do I hear 10? orders are placed offline by people who did their research online. What did they research? A store-locator? How did they research? Did they navigate by brand name or domain? Did they research and buy from the same company? Did online marketing play any role in getting the person to the website?Why isn't anyone asking these questions? Aren't folks like Forrester interested in gathering meaningful, actionable information?

Granted, we're in a complex space, and a certain amount of misinformation is inevitable when we're dealing with "black boxes." In some situations there may be legitimate differences of opinion about the right approach, and that's cool to. Consultants who built their reputations in one field and now find themselves forced to opine on channels they know little about, have no alternative but to read the blog-o-sphere...which is ripe with misinformation.

I also understand that in competitive marketplaces, those competing for clients and those competing for advertising dollars are likely to spin the truth to get the business.

However, it's annoying that there isn't a fact-checking service on the marketing blog-o-sphere that stamps the misinformation with a big red x. It's annoying that there isn't a "rebuttal" given for every lousy conference talk, so the impostors can be called on the carpet at the moment.

Most annoying of all though is when people knowingly spread misinformation to our clients. We don't appreciate the waste of time, and we aren't afraid to call out the lies in plain language.


30 Responses to "Misinformation and Disinformation"
... via rkgblog online mktg Misinformation and disinformation...
Billy Wolt says:
George, great post. I can't help but think that this came about because of a client that said 'xyz ppc expert' said this is wrong; which probably sparked a problem. If i had a dollar for every blog post i was sent by clients/bosses, i'd be rich :) I am sick of all these blogs, experts, ppc authority sites, etc. For every post that says this works best, you can find a post saying the opposite. The PPC industry is getting washed down by people that put up a few keywords for their cousins business and claim to be experts. What I like about RKG, is that you guys actually provide good info, back it up with data, and actually make one think differenlty about what they are doing.
Jim Novo says:
Hmmm, no comments yet on this? No defenses? It really is stunning how many clueless ideas are pitched by "experts" at conferences. Here's what I think happened: 1. Some entrepreneurs realized that "web people" are mostly clueless about Marketing, since the majority have zero real training in Marketing and / or have worked for people just like them (with zero real training). Buying media is not the same as understanding Marketing. 2. This audience provides the opportunity to create endless, meaningless conferences where people say the same thing over and over using different words. Example: if these folks had any real training, they'd realize that a Headline is always important - on a page, in search, in email, in any communication - and there is no need to announce a "breakthrough discovery" that headlines are important in whatever the next media darling is. 3. The number of conferences is so extreme that as the conference planner, you completely run out of valid speakers who know what they are talking about. So to pull off the conference, you have to recruit people who speak who don't know what they are talking about. And you know that doesn't matter, because nobody in the audience will know these speakers are spewing complete crap. 4. The trade mags want the ad revs from these conferences, so they no longer vet any of the material, they simply republish it and promote the conferences. This results in myths being created and then justified and repeated endlessly as bloggers and daily newsletters etc. repeat the same crap over and over. 5. This conference spew becomes "Marketing Knowledge". Sound plausible?
Tim Barcz says:
George, I've long experienced the same problems in my past. As a client, the obvious question is, "How do we sift apart the people who know what they're doing from those who don't?" You can't just trust someone's word for it and as you've pointed out you can't simply look to conference speakers or blog (or to some extent published articles). Certainly you can verify with a third party, but who is to say the 3rd party is reputable. Perhaps a fourth...but then...you see where this is going. So what do you say to the lay person - how to they sift the wheat from the chaff?
There you go again George. Always going on about 'data', 'facts' and 'evidence'. You're ruining this business for a lot people with this insistence on reason and logic. Keep up the good work ;-)
Thank god... I'm not alone!
I seem to have struck a nerve! :-) Billy, thank you so much for the kind words. As former multichannel marketers, we intended RKG to be a different kind of agency...not an agency at all, but rather a marketing service provider. We do not view our role as trying to sell as much advertising to our clients as they will buy, and that seems to differentiate us from most other firms...sadly. The blog is an extension of that philosophy.
Jim, thanks for chiming in. Your scenario sounds too dang familiar to be plausible, it must be dead on. At our Client Summit in May we invited 4 speakers who I knew would be terrific, and that little mini-conference -- with lots of networking/brainstorming opportunities -- went over very well. I wonder if putting together smaller, top quality shows might not be a business opportunity for someone. In fact, Jim, we'd love to have YOU speak next spring if you'd take the time. Charlottesville is beautiful in the spring!
Tim, you raise the key point. I wish I knew the answer. There are some very smart, straight shooters out there, and figuring out how to identify those diamonds in the rough takes equal measures of skepticism, knowledge and a keen nose for marketing BS. We've been kicking around the idea of creating an Alliance of marketing specialists/experts who really know their stuff and are straight with their clients and prospects alike. Inclusion in the alliance would require an annual "thumbs up" vote from the other members, objective client satisfaction surveys to ensure their reputation is still well-earned, etc. I've talked with the heads of several firms we hold in high regard, and they seem to see the same need in the industry. There is this feeling that there are far more charlatans in the business than quality service providers, and that if the quality providers banded together...well, anyhow, we're working on it :-) George
Beest, thanks for joining the discussion. If we had any sense we'd probably be a much bigger company, raking in the dough by over-promising and under-delivering. Those darned facts keep tripping us up where others seem more adept at side-stepping them :-)
Bryan, glad to hear the call of another soul wandering in a desert of misinformation looking for a way out! Keep the faith!
Robert Brady says:
The problem with "experts" is that so many are self-proclaimed. Expertise is gained by experience and backed up by valid data. PPC is very simple; choose relevant keywords, write relevant & compelling ad copy, send the search user to a relevant landing page with a clear CTA and benefits.
Jim Novo says:
George, I'd be honored to speak at an RKG event! There have been several attempts at creating high quality "knowledge rings" of various kinds over the years. Unfortunately, most fail because the business models usually rely on Advertising. This is a huge mistake because now you're set up for a "quantity versus quality" optimization and, well, you're back to the beginning. Just get content; it becomes just like the conferences. So, the solution is probably subscription-based, like many of the academic journals, e.g. Marketing Science. Also similar to these journals, peer-vetted. They simply do not publish anything that the "Board of smart people" doesn't agree is top notch and fully verified. Now, those academic journals can also be boring as snot, and most people will have to skip the math, reading the Intro and Conclusion to glean the nuggets (I often do just that). Perhaps there is a middle ground somewhere...frankly, I am surprised the trades have not picked up on this, but they simply have devolved into publishing what's "popular" and don't seem top vet the material at all; it's like they are just grabbing stuff from blog posts!
We're taking you up on that, Jim!
Jim Jansen says:
George, Always love your posts! Do you think part of the problem may be "speaker saying one thing and listener hearing another"? (or maybe, in the cases you bring up, the speaker being confused!) For example, conversion rates. It is a ratio of conversions to clicks (e.g., 10/100 is same as 1/10). However, it is very easy to mistake this for an absolute number (i.e., bottom line revenue based on ad rank). So, the conversion rate of an ad at rank 1 may be the same as conversion rate of ad at rank 20. But, obviously, in terms of revenue generated, the ad at rank 1 is going to generate more revenue because it is going to get more eye balls on it and more clicks. (see Hotchkiss eye tracking studies http://pages.enquiro.com/whitepaper-enquiro-eye-tracking-report-I-google.html showing the Google Golden Triangle and Brooks analysis of clicks http://www.atlassolutions.com/uploadedFiles/Atlas/Atlas_Institute/Published_Content/RankReport.pdf showing a curve-linear function of the effect of ad rank) If you want to read a good article about the relationship between ad rank and conversion rate, see this article also by Brooks http://www.atlassolutions.com/pdf/RankReportPart2.pdf. Nico reports little change in conversion rates but a 90% drop in **conversion potential** due to the decrease in number of clicks. Again, good posting! Jim
Tomas says:
Here's another one that google straight of a Google rep's account evaluation. They suggest to bid more aggressively in order to "Improve average CTR and Quality Score for these terms". I would have hoped that they knew by now that the algorithm uses CTRs adjusted for position and that you can't bid yourself to a high QS *sigh*
Jim J., There isn't much debate in the industry that you get more traffic when your adrank is higher. The economics are tied to the value and cost of the traffic, not the quantity. If the value of the traffic is pretty close to the same, but the cost goes up the higher the adrank then pretty clearly, once you start spending more per click than you can afford to spend the higher your adrank the faster you lose money.
Tomas, that's a beautiful example, thanks for sharing!
Pat East says:
"Account structure is key to Quality Score. False. Account structure has virtually no bearing on QS. The relationship between the Keyword and the copy matters." 100% agree that the relationship between the keyword and the copy matters. Isn't the best way to do that by having a good account structure though? Tightly-knit ad groups are going to produce higher CTRs, and the best way to have a tightly-knit ad group is to have a solid account structure. I recognize that the account structure -- in an of itself -- won't bear on quality score, but it certainly is the basis for being able to have solid CTRs, which will generate a good quality score. The two seem very closely intertwined. Anxious to hear your thoughts on this.
Pat, there are many good reasons to have a coherent account structure, but QS is pretty far down the list. The reason gets back to Google's mechanism for determining QS, which gives the heaviest weight to the most granular data that is statistically significant. If the ads have been in the account for a while, Google is probably relying exclusively on that ad's performance history (combination of keyword, text, landing page). The rest of its AdGroup's history and Campaign's history will have no bearing, hence the Account Structure has no bearing. However, when adding new keywords with no history, having rich adgroup history can help or hurt that keyword as it gains its own data. Writing targeted ad copy requires finely parsed AdGroups to begin with. The question is does it matter whether all the "left-handed metal widget" keywords live in one AdGroup or several and the answer is: it doesn't make much difference. We're finding that the top reason for developing user-friendly account structures is so the consultants and Google sales people have even less to talk about :-)
George, love the post. The points you speak to are actually significantly subtler than many of those that routinely get mangled out there on the circuit, however. It's whack-a-mole just to stop people from making the most obvious mistakes, let alone the stuff about calculations of lifetime value. As such, I confess that after years of trying, I can't quite figure out what Jim Novo wants from us. I have a vague sense that we're all wrong about everything, and that none of us are qualified to speak on anything, but beyond that, I get a bit lost. I'm sorry, Jim, but the "zero real training in marketing" thing sounds like gatekeeping. The vast majority of the entrepreneurs and tech companies I admire have zero real training in marketing. Whether that's Zappos, Google, RIM, Yelp, the Ruby on Rails Pub Night, or YouTube, disruptive companies and/or movements are supposed to disrupt. If we had time to go back and get some re-education in marketing, of course it would be valuable. But a lot of great people have "zero real training" in one thing, while having a lot of formal and on-the-job training in other areas. That's by necessity. Things move very quickly. A lot of time is wasted by "real training". The people I see coming out of "real training" programs often miss most of what they need to be/do to succeed in our world and are looking for guaranteed salaries to sheepwalk by the book. Anyway, a much simpler way of putting Jim's point is that there is an ever-fluctuating degree of vendor bias on panels, full stop. The worst type of bias is actually not the direct kind where someone's clearly got an interest in selling you an account structure tool, so they blather endlessly about account structure (obvious bias, right?). It's all of those speakers who just bull their way onto panels somehow, speak on a specialized topic without much expertise on that topic, just as a means of getting up there for the exposure, the link, a show pass, and an opportunity to schmooze and collect business cards. There's also an imperative to avoid giving "stale" speakers too much airtime... they're not fresh, right? But the "stale" speakers are experts, and they know stuff. Novelty amuses show organizers but I'm not so sure it helps attendee companies profit. Anyway, on point 2: George, I'll go along with your point, but hasn't account structure always been a nice conduit towards achieving the appropriate degree of granularity, forcing people to better understand relevancy, allowing them to think and manage their accounts better, etc.? Arguably yes that's harkening back to a time when the sense of a fixed ontology in an account was a real aid (when PPC was new) as opposed to just being a nice bonus what with the many new features, third-party tools, and analysis methods available to us. But just because annoying folks are over-selling the account structure thing, doesn't mean it hasn't been a useful aid. And it still can be. I can think of clear examples (for example, involving geography) where a restructured account improved conversion rates, CTR's, etc. (I don't care that much about QS, it tends to fall into place). Granted, that means rewriting ads as well... but I'm sure you can see my point.
Terry Whalen says:
Darn, the cat is out of the bag (on position #7)!!!
Terry, we were bummed that that guy let out the magic of position 7, too :-) Andrew, we're honored to have you drop by our humble blog! As I mentioned to Pat, there are very good reasons to have clean coherent account structures, it's just that QS isn't high among them. With respect to Jim's point, no question that the game changers often do so by going it alone and ignoring conventional wisdom. Certainly, when you built your business and we built RKG we kind of had to figure out the rules of the game, and realized that we'd learn more by doing and studying the data than by reading what others thought. That said, great scientists and artists go places others have never gone by learning the fundamentals first. No need to re-invent calculus. I think folks benefit tremendously from learning the fundamentals of direct marketing and applying those fundamentals in new ways. Great discussion topic! Thanks again. George
"Aren’t folks like Forrester interested in gathering meaningful, actionable information? " The folks there yes... The company itself, maybe not. I've yet to be really impressed by any research they've done. Their main utility seems to be assisting in lead gen, afaik.


Check out what others are saying...
[...] This post was mentioned on Twitter by Alan Rimm-Kaufman, ana peña and others. ana peña said: Misinformation and Disinformation: George rants about myths and the people who propagate them. http://bit.ly/aXHTzc #fb [...]
[...] sensible column about “fake” PPC experts over at Search Engine Land last week. (George Michie also nicely pointed out some useful advice he’s really written lately.) Surprising that it was so controversial. We [...]
[...] Which is all, naturally, assuming you’re not mislead by common PPC myths. [...]
[...] Which is all, naturally, assuming you’re not mislead by common PPC myths. [...]
[...] Which is all, naturally, assuming you’re not mislead by common PPC myths. [...]
[...] Michie clears the air on Misinformation and Disinformation about paid search, myths that even the paid search vendors themselves are [...]

Leave A Comment