Sunday, November 10, 2013

An example in how to minimize the "multi-channel attribution" issue

As we all know, companies have a hard time attributing the merits of their multichannel marketing campaigns. How do you know if someone watched a TV commercial then searched for your company or if they were just searching instead?

The PR people I hired to help in my offline presence were always bragging about how successful their work was - but we never really found a way to get specific statistics on their impact to compare it to other marketing efforts. That is, until we gathered enough data so that we had a "base rate" of website visits on any given date and time to compare to times when the PR team got us a good press release or interview.

So now what we do to measure their impact is fairly straightforward: consider a Sunday night, when our typical number of site visitors is around 600 to 800. In this specific Sunday night, there was one difference in the context: our PR team got us an interview that was airing in a famous TV news program. The minute the interview was off, we monitored our statistics and were very satisfy to see the number of visitors increasing steadily until it reached 2,200.  Now, simple maths dictates that the additional impact of PR was of about 1,400-1,600 visitors.

We have also designed a change to our basic sign-up sheet online that includes asking whether the person found out about our company from the specific show that was airing that day - and so we can monitor both new visitors and subscribers coming from the PR team's efforts.

Even though such metrics are not perfect, they allow us to understand the impact of different marketing channels a lot better than we did before!

No comments: