For all of the good information that is shared online there is just as much that could be (generously) considered questionable. Two of the areas where you see the most misinformation is in monitoring and measurement. Why? Frankly, some of the blame has to be at the feet of the challenges we've faced in traditional communications. We've had a hard time truly grasping the utility of monitoring beyond a reputation management function for years. Similarly, traditional communications measurement has been butchered seven ways to Sunday despite the great work of
Katie Paine (and others) to educate everyone. For the record, I refuse to blame that butchering on communications professionals not liking math. That feels like a thoroughly uneducated answer to me. There are a lot of other factors there that we'll explore some other time, but for the purposes of this post know that measurement (well beyond social) is being butchered.
Yesterday,
Todd Defren wrote a great post about PR measurement failures. The trouble as he points out there is that we have widespread acceptance of formulas and concepts that really don't mean a heck of a lot. The scenario as he painted it was:
“Say the client spends $100,000 on PR, in one year. For the sake of argument, let’s say PR is the biggest (or only) marketing vehicle. In that one year time period, the client gets 1M website impressions. Could you not divide 1M impressions by $100K and claim PR is driving leads to the website at a rate of 10–cents per impression?”
Wow. Interesting, eh? You can check out the comments for all of the rebuttals (including my own). Here is the hard and fast truth...
While we have generally accepted best practices for measurement none of them are without fault. My suggestion to Todd in that post was to come up with an index model that takes into account several different metrics. I'd consider that best practice, but I'll admit there's no direct ROI tie there. While there could easily be a connection made to brand reputation, the metrics are soft(er). Don't get me wrong, we should be searching for the harder metrics/calculations/formulas, etc... However, I'd much rather move the debate toward coming up with a generally accepted framework rather than the
Bataan Death March approach we take toward ROI. It wont be perfect, but at least it would be generally accepted.
Monitoring, unfortunately, happens to be in the same boat. Upfront, I think it's important we clear up one small misconception:
monitoring is NOT measurement. Monitoring helps to inform measurement, but the two terms do not mean the same thing. Now that this is out of the way, when you're going through the process of picking a monitoring provider you need to know that no tool offers 100% capture. No tool is going to offer you 100% technological stability. There will be bugs, and there will be conversations that the systems do not capture. What should you look for?
-
The tool has the ability to adapt - By adapting I mean if you find a source that the tool is not capturing can you send them that source and they can modify the "net" to begin capturing it. If the answer is no, then it is time to look for a different solution.
-
The vendor is adding sites on their own - Obviously, the onus shouldn't be on you to identify new sources for them. They should be taking it upon themselves to advancing the "net" to capture as many sites as possible.
-
Consistency - No matter how diligent you/they at updating their platform, there will be sites that are missed. Just accept it. The worst case scenario, though, is for a certain set of sites to be missed this week and then next week a whole different list is missed. If we're going to have a gap, we should know that the gap is going to be consistent so that we can potentially identify other tools that could help.
With respect to
Lexus, the pursuit of perfection in measurement and monitoring is silly. We should be pursuing what makes the most sense for our clients and our organizations. If the metrics fit the goals of the campaign and is generally accepted by those constituencies, then use that methodology. End of story. What has worked for you in the past?
By the way, thanks to
Kasey Skala for inspiring this post.