This article was originally posted on DMNews.com
We can track who views our marketing campaigns online. We can tell which ads they viewed; how long they looked, and where they went afterwards. Everything is measurable today. You’d expect me, as the person who wrote the book Sexy Little Numbers: How to grow your business using the data you already have, to tell you that all that data and all the new technologies that can analyze that data will solve all your problems.
I won’t—because the data and technology are the wrong place to start, as I learned at the beginning of my career.
Back in the early days of the Internet, when data was hard to come by, a revolutionary product hit the market. Webtrends Log Analyzer could spit out a 70-page report every day with hundreds of metrics on what was happening on your website. Once I got access to Webtrends, I set up automatic feeds for all of my clients. Every morning they would open their mailbox and find a 70-page report on all the activity on their site from the previous day. There was a lot of excitement in the beginning. “All that data–unbelievable.” “Everything is measurable!” “This is incredibly powerful.”
However, the honeymoon period was short and it was followed by a long period of silence. My clients really struggled with the data. There were too many metrics to digest. It soon became apparent that 70 pages every day was overkill. But even when we went to weekly and then biweekly reports we found very little interest. When I asked why, my clients said, “There’s no context. What does it mean if I have 5,000 visits to my site a day or that my visitors spend two minutes and 52 seconds on average on my homepage? Is that good or bad?”
Those questions pinpointed a real problem then. One that is still a problem now. We tend to be too focused on what we can measure, and not focused enough on what we should measure.
So the place to begin is by figuring out what you need to measure. What are your goals? Increased sales? Higher margins? Increased sampling? Customer retention? Then you need to prioritize those goals. If everything is important, nothing is.
You would think that managers would prioritize objectives as a matter of course, but they don’t. And that’s a real shame because when you get this wrong, everything else you do from a measurement point of view won’t be effective. You need to get agreement.
Once you have it—and only then—do you tie metrics to each and every one of your prioritized goals and figure out how you are to monitor them so you stay on track. Simple stuff right? You’d be surprised how many companies I’ve seen who don’t do this.
- Inputs: What are we investing in to grow demand?
- Outputs: What results are generated as a result of these investments?
- Outcomes: How is what we ultimately want to achieve impacted by what we invest in.
Here’s a simple, but specific example. Let’s say we sell widgets and our number one goal is to acquire new customers in order to grow demand. To do that, I might run a few print ads and spend some money buying widget-related search terms on Google. The inputs would be the money spent on print and search.
The output could be the print and search impressions we get for that investment, as well as the awareness and likeability of the print ads and the clicks and click-through rate for my search advertising.
The outcome would be any incremental new customers I acquire.
This may seem extremely obvious. Why, then, do I see so many dashboards or report cards that don’t have outcomes, only outputs. Or they fail to put the outputs in the context of the inputs. With all the technological possibilities in measurement today people often tend to forget about the basics.
Here’s another basic skill I wish I’d see in practice more often: testing. David Ogilvy argued that the most important word in the advertising lexicon was test. Yet, despite all the cheap automated testing technologies available today, many companies are forgetting about the basics of:
1. Setting up a learning agenda,
2. Capturing the learnings,
3. Distributing them as best practices company-wide, and then (of course)
4. Tracking compliance with these best practices when new executions are planned or put in market.
These are the basics and, if implemented at scale, they can make companies millions of dollars. Nevertheless, I know so many companies that seem to have lost this disciplined approach to testing and learning over the past couple of years.
The digital revolution has given us tremendous tools, but have we become smarter marketers? I’m not always sure we have. A return to basics can help a great deal.
Follow him on Twitter: @dimitrimaex