Yesterday, your faithful correspondent was part of an interesting session held by the Council of MLS (CMLS) here in Anaheim on branding or certifying MLS data. The basic premise with which the meeting began was something like the following:
- The MLS has a strong brand for trustworthiness and accuracy in the consumer’s mind
- Consumers assume that all real estate data is MLS data
- Third party data is crap
- Therefore, the MLS should brand and/or certify MLS data
Various speakers presented various pieces of evidence supporting the above. And if you accept the premise above, then branding the MLS data in some way is a no-brainer. The two methods proposed are (a) brand the destination as “MLS Trusted”, and (b) certify the listing data itself as “MLS Certified”.
But I have my reasons for wondering about the assumptions.
Survey Bias vs. Gut Feeling
The biggest reason for doubt about the assumption is that the evidence presented came mostly from two sources.
The first is a study conducted by… somebody… (WAV Group?) of brokers, asking them about their opinions of what consumers want and/or believe about real estate data. The survey bias is… well, obvious in this case. It may be the case that brokers are right on the money as far as what consumers want/believe. But one can’t help shake the feeling that opinions of brokers and agents may be influenced by the fact that they sell real estate for a living.
The second source was a consumer survey conducted by somebody (I don’t have the slides used, so don’t recall who did the survey; sorry) that showed that consumers trust MLS data, prefer the MLS website, etc. etc.
Trouble is, it became evident that these consumers were solicited online, on the MLS website itself. Meaning, Joe Public is on MLSWebsite.com looking for a house, and some sort of “Hey, take this here survey” thing pops up. Well, that’s a pretty self-selected group of respondents. They’re already on the MLS website; they already know what the MLS is; they already use the MLS data. It should be shocking if any of those respondents say negative things about MLS data.
See, my gut feeling based on nothing but the state of my tummy is that most consumers actually don’t know what the MLS is, nevermind the quality of its data. Even if they’ve heard the term, their understanding is vague at best. They have no idea about the difference between MLS data and syndicated data and so on and so forth.
Why This Might Be Relevant
So what? Why does source selection bias matter at all?
Well, the issue for me as a marketer is that what is being proposed — whether branding the destination or certifying the listing — is neither free nor easy. Many of the speakers brought up other successful branding efforts like “Intel Inside” as what the MLS should do in terms of its branding effort. Well, that campaign cost $4 billion from 1991 to 1997. Of course, the payoff of the ingredient branding campaign was domination by Intel for years.
What professional marketers understand is that branding is a hugely expensive, time consuming, but potentially valuable initiative. The key to thinking about it, of course, is whether the expected benefits outweigh the costs involved: you know, that old chestnut, “Return On Investment”.
So thinking about branding the MLS data turns on whether that branding would actually result in some sort of benefit that justifies the investment of time and money.
Therefore, the selection bias in the studies supporting the proposition that the MLS has a strong brand in the consumer’s mind is troubling. If that assumption is wrong, then spending time and money on branding or certifying data as “MLS Trusted” or “MLS Certified” is a giant sinkhole.
Proposing An Experiment
I have a suggestion. Before the industry invests significant time and resources into branding the data, why not run a short-term test to make sure that branded data has an impact in the marketplace?
Site Branding Experiment
Launch two IDX websites, identical in every way: same design, same data, same everything. IDX sites are cheap and easy to setup, so it can be done in a matter of days. Put a “MLS Trusted” tag on one site, and no tag on the other site. Monitor traffic, SEO ranking, clickthru rates, lead generation from the site, etc. See if there’s a meaningful difference between the two sites. If there is, then we know that the “MLS Trusted” brand is meaningful; if there isn’t, then we know that the brand isn’t.
Allow a website to commingle listings that are “MLS Certified” with listings that are not. Ideally, this would be the same IDX feed; tag every other listing as “MLS Certified” so as to randomize which listings are tagged and which ones are not.
Monitor clickthru rates and lead generation on a listing-by-listing basis. See if there’s a meaningful difference between certified listings and uncertified listings. Again, if there is a meaningful difference, then one can try to calculate the expected ROI from a branding/certification campaign. If there is no difference, then… well… y’know…
Testing & Data = Soul of Marketing
I think running those two experiments for 90, 120 or 180 days should provide enough actual data from actual market performance to determine whether a branding/certification campaign is worth doing. Marketing isn’t actually about creativity and design and pretty words: it’s about numbers, data, and analysis.
CMLS would be a wonderful organization to coordinate these experiments; there are any number of companies with a long track record of market research who could conduct the experiment for CMLS. And the resulting data and analysis would be wonderful to have.
Of course there are a number of other factors you want to consider about ingredient branding. But without proving the basic assumption about how consumers actually view the MLS in the marketplace, nothing else really matters.
Your thoughts/questions/hollerin’ are, as always, welcome.