My client at JNJ sent me this article from Brand Week about the clamoring for ROI metrics around Word-of-Mouth marketing (WOM).
We both agreed that it was over the top and silly. His initial position (and he’s a REALLY smart guy) was that WOM and Social Media consultants are trying to put metrics on top of something that “we all know” can’t really be measured.
As the conversation progressed, however, we realized that the consultants are simply responding to market pressure. Namely, direct marketers who are demanding ROI so they can “sell internally” a WOM initiative.
Now, I’m not saying that ROI isn’t important. After all, we’re all in business to make money. I’m just saying we need to think in a different paradigm…and we’re not there (though at Bunko, we’re trying) to figure out what really moves the revenue needle.
There are going to be a lot of “social media” consultants coming online soon (and this could hurt the industry), so when you ask them (or me, hopefully!) about ROI, really listen to the response.
After all, as Seth might say, the DM tactics that yield ROI are interactions and we spend too much time there. Not enough time on the other 4 components.
But that’s a post for another day.
Comments
Adam Schorr said on 7.13.2008 at 1:00 PM
Jer:
This is a really critical issue. Thanks for posting on this.
Metrics in general can be very dangerous. The old saw is "what gets measured gets done". And this is probably true.
But it's a sword that cuts both ways.
I've seen too many instances where people create metrics that are easy. Easy to calculate, easy to explain... but which do not create the right incentives. Often, the most important contributors to success are the most difficult to measure.
One more thing: Following metrics is just a form of empiricism. Radical empiricism can be very dangerous. Sometimes you have to follow a sound theory even though you either don't know what to measure or don't have the apparatus with which to measure it.
So now on to WOM: I think the theory behind it is sound. Not that we should stop there of course. Everything is open to further study and analysis whether it is of the empirical or theoretical sort. (See the recent WSJ article on the long tail theory. http://online.wsj.com/article/SB121493784638920147.html) But I think it's worth engaging in some WOM activity even if you haven't a clue how to measure it, simply because it makes so much damn sense.
But I do think it can be measured. The question is how deep we go on the measurement. Marketers today have all sorts of different metrics, not all of which extend as far as sales. Plenty of marketers will look at brand equity measures such as awareness (aided and unaided), advertising recall... and be pleased that the numbers are holding or rising even if they don't have a marketing mix model to tie those numbers to actual sales.
PR works the same way. Forget about crisis management which everyone knows they just have to do. Typical brand PR is very difficult to measure. We look at impressions. Maybe we're sophisticated enough to classify impressions as positive or negative or neutral. But we don't know if impressions are double counted (someone sees Oprah mentioning your product and also reads a favorable editorial in Cosmo). We don't know the value of an impression... And yet, there are lots of big PR firms paying for expensive office space in Manhattan.
You WOM guys have to come up with some metrics along those lines. You're not going to tie it to sales. Not in the foreseeable future. But how about a very simple metric: how many hits you get on Google. I bet you can even find some simple software that will classify those hits based on additional keywords into positive, negative, neutral.
And why not some sort of survey? These are easy to implement and relatively cheap. Buy me a few beers and give me an hour of your time and I'll bet we can come up with a handful of good questions to ask. For example: I would imagine that if WOM is working right, that over time what a consumer knows about my brand will come more from their peers (however they define "peer") than from my official sources (tv commercial, website...) So why not ask respondents what percent of their info about a brand comes from their peers and what percent comes from the company...
We can do this. We just have to be smart about it.
But I don't think we can tie it to sales.
jer979 said on 7.15.2008 at 9:47 PM
Adam-some solid points here. Thanks for sharing. You may want to check out Saul Colt's perspective on this issues, I think you'll like it.
saulcolt.blogspot.com/.../wom-campaigns-f