Commentary
Give Me Buyers, Not Searchers!
- by Steve Baldwin , May 4, 2009
Offline direct marketers have long had the ability to buy lists of many different types. The lowest-priced lists, which are called "compiled lists," have the least likelihood of resulting in conversions because they only contain very basic information about people, e.g. the fact that an individual resides in a certain DMA or is a member of a certain professional group. Compiled lists aren't exactly worthless -- in fact they may work very well for marketers selling products like shoes (everybody has feet, after all) -- but their cheapness reflects their low conversion potential for specialty marketers.
advertisement
advertisement
Up the value chain a notch, we find lists containing more behavioral data, such as lists of people who've "raised their hands" by filling out a catalog request form or otherwise expressed interest in a product, service, or manufacturer. These lists are more expensive because they're more likely to result in sales. But just because somebody wants to peruse the L.L. Bean catalog doesn't mean that they're ever going to order something through the mail.
The most valuable direct marketing lists are of people who've actually bought something. These people haven't merely raised their hands; they've reached into their wallets. Many retailers make a fortune selling their customer databases to other firms hungry for such data (provided they are not direct competitors). Response rates and ROI from such lists are orders of magnitude higher than lesser lists, as are the prices for renting them.
OK, let's get back to search. Marketers buy keywords to trigger ads, searchers click on them, essentially "raising their hands" to express interest, and the search engine gets paid regardless of whether or not an order takes place. The obvious problem is that the overwhelming majority of such paid clicks don't convert. A whole cottage industry has evolved counseling search marketers about the importance of compelling copy, landing pages, segmentation and bid strategies to raise conversion rates, and that's not a bad thing. The problem is that nobody has addressed the fundamental problem, which is that most clickers, shoppers, and other "hand raisers" have neither the will or the way to actually execute a transaction. Put more crudely, the vast majority of clicks that search marketers buy are wasteful, everybody knows they're wasteful, but we all keep paying for them just the same. In any other industry this situation would have long become intolerable and something better would have replaced it, but we just grit our teeth and accept it -- how crazy is that?
What search marketers want more than anything is what old-line direct marketers have had for a long time: lists of people with money in their pocket and a history of executing online transactions. Search engines have this data (how many sites now run Google Checkout?). But inexplicably, the engines refuse to let marketers bid for these valuable people, reducing them to fighting over keywords that only prompt waves of noncommittal "hand raising."
Why? Conspiracy theorists would argue that the engines would make less money selling search media this way. Another possibility is that the engines fear that if the masses ever learned how much intimate information the engines have gathered about all of us, including our buying behavior, privacy advocates would raise hell (a fear that's never deterred offline direct marketers from working this way). Mark my words: the first engine to come out with a way for marketers to buy qualified buyers will blow its competitors to Kingdom Come.
It's ironic that an industry that so conspicuously trumpets its "innovativeness" is stuck in such a Jurassic time warp. But there are third parties working on a solution. It seems to me that as long as we're targeting, slicing and dicing every minute aspect of a user's online activities, we shouldn't overlook the most important one, which is whether he or she has the capability of ever being a customer. None of this is rocket science: it's just common sense, and if search engines won't sell media this way, it's high time that third parties take the plunge.
In a perfect world we could also predict buyer intent. It is not that we don't want to market this information or these "potential buyers". The problem is that the consumer protection groups have drawn a privacy line in the sand that makes it difficult to use the information gathered for anything other than anonymous stats. The other question is do consumers really want ads this targeted? My guess is most will feel it a bit creepy that advertisers do know that much about their buying habits. So there is a fine line with privacy,creepiness and consumer protection to walk with behavioral and list targeting.
Yet another thing to consider in the quest for conversions is the economy. No one is buying much of anything. So why do we keep using old models to gauge success?
I also take issue with your statement that "In any other industry this situation would have long become intolerable and something better would have replaced it....". While traditional media sales are declining advertisers still put up with their inserts getting torn out of magazines and trashed and people leaving the room while their ad is on in the background because it is a numbers game and always will be.
in the online world they call it click fraud or non converting traffic....traditional media does not yet have a pithy catch word or phrase for the missed opportunities mentioned above so it doesn't get mentioned as much.
No medium is perfect nor ever will be because as long as consumer information is and should be protected there will be only so much targeting you can do.
A Magic Pill ! WOW! Bryan said it better than I could and you should.
Very powerful point, which is why we have developed auto online studies that report on the online behavior of actual buyers. We clearly see the difference in behavior between shoppers and buyers. A reveiw and sample of the results are at www.OnlineAutomotiveReview.com
Lol...you can reach users when they're search on terms such as "buy movie tickets" and "plasma tv deals"...what more do you want? Geez. Paid search is still the highest ROI vehicle (next to inhouse email lists) and you're not happy that we can't qualify the users first?
It's not the search engines stuck, it's the online public overly concerned about privacy. Believe me, if there wouldn't be any backlash from users to allow search engines to either pass that data through or allow advertisers to pass data to them, then it would be happening as we speak.
Old-line direct marketers will be the first to tell you that while the list is one important factor, there are other contributing factors that impact success and ROI like the offer itself, the creative execution, etc.
So rather than wait for a holy grail that probably really doesn't exist, I would recommend you take advantage of the tools the search engines do provide to fine tune your paid search programs and improve conversions the old-fashioned way -- with marketing smarts.
I believe the problem that Steve has correctly identified is more fundamental than marketing to the apparently already converted. It has to do with what Clay Shirky refers to as the distinction between information overload and filtering. I could (for example) provide the hyperlink to Shirky's video on the problem, but I won't (mainly to illustrate the point). Go find it yourself. The point is: stop thinking search and key words and all that other needle-in-a-haystack stuff. Think about "findability" and "addressability" which in the world of search is clearly problematic. Want to find Shirkey's video? Try "Clay Shirky information overload" or some such. Maybe you will and maybe you won't find it.
Nor does so called semantic search solve the problem: too much ambiguity in language. What to do?
Arguably (and this only one of presumably many possible solutions) is to privatize the Web, which is to say channelize it along addressable lines. It can be done, literally, with single words: kind of like key words except that each key word has a channel, and each channel has sub channels (separated by dots in the dot-com name space, that modify the words in various ways.
It can be done and in fact it was done, way back when, in anticipation of overload and, thus, filtering. And not only that, anybody can use the available word-based architecture that was layered over the Web in the dot-com name space, starting in 1999. Want to find it?
It's there. Really! Check it out on the search engines.
Wait! I have an idea! What if there were a way to collect qualified buyers of specialty products into a stream, by presenting, say, some kind of online editorial content aimed specifically at their needs? And then you could additionally qualify these buyers by seeing exactly what content drew their attention, capturing and converting them by placing ads right next to that highly-topical content? Even better: in that kind of specific context, nobody would be creeped-out to see relevant advertising!
Couldn't do it for free, though. (grin)
There lies the rub - delivering QUALIFIED buyers. If we could only just sell and not have to INVEST all that time and money on marketing, bringing buyer and seller together, and keeping in touch with them throughout the buying process. Messy stuff marketing. I don't think it has dawned on people yet that social media is just that - social.
Yes, whilst it difficult to locate a list of buyers for your product and services has marketing really changed?
It always been about well scripted headlines, copy and layout.
and...
it's about
1. The List;
2. The Relationship you have with your list;
and
3. The Offer.
Get these right and the numbers start falling your way.
Cheers Kurt - http://www.kurtjohansen.com - Australia's Email Marketing Guru.
Ouch - that's a little radical. The search engine shouldn't have access to that data (unless you are Google analytics), and that could cause some major problems. The only way to get around that is to have a behavioral targeting network that follows what you've bought - and is done with the permission of the retailer... and consumers :)
Also, that would only be apropo for a product that keeps on getting bought again and again. Some products are bought once, and not afterwards. It would almost require a cross-website amazon.com recommendation-like engine...
Lemme guess, Steve's got something up ;)
What this article fails to mention is that the price that people pay per click is based on conversion rates and if their conversion rates suddenly skyrocket because they are only reaching people that have purchased previously, then their bid amount would also skyrocket. So the net real result would be no change in cost per sale (once you remove market inefficiencies).
Sidenote: also worth mentioning the fact that not all commerce that results from online research is conducted online. Most complicated items - business software, cars, houses and most anything over a few grand - is still purchased offline. But it makes perfect sense for advertisers to still pay to reach them for sales lead generation.