14 Nov 2013

Earned media might be invisible to your web analytics

One of the questions that consistently comes up about the power of earned media and PR is how accurately it can be measured. The answer to this question depends on how you’re gathering data. If you’re relying solely on clickstream data and web analytics, chances are your data is woefully inaccurate – and you may be making poor marketing decisions based on that information.

Here’s an example. If we look in our web analytics at conversions for the quarter to date, we see that using standard last-touch attribution, Search rules the roost, followed by direct traffic (which is Google’s catch-all for “we don’t know”), then referral traffic, social media, and email:

Screenshot_11_14_13__6_17_AM

So far, so good. The logical conclusion from this data would be to beef up our search efforts, since it’s clearly working well, right? Perhaps if we want to get ambitious, we could look at where traffic sources come from before they convert to see if it reflects what our conversion data looks like. In this case, search still rules the roost, though other channels bring in more traffic than they do conversions, which means we shouldn’t discount them from our strategy, particularly if new audience is an important metric:

Screenshot_11_14_13__6_22_AM

Right then, I’ll go and tell our CEO, Todd Defren, that we’re betting the cart in the coming year on search. This would be a reasonable assumption based on what you’ve seen, wouldn’t it?

Except I’d be wrong. You’d be wrong if that were your guess. Here’s why. We take our own advice and ask this most important question when you fill out the contact form on our website:

Contact_-_SHIFT_Communications_PR_Agency___Boston___New_York___San_Francisco-4

This is arguably one of the most important questions you can ask – and it’s important that it be a free-form list, not a pre-populated list (which can create biases). What do we find in our data, in our database, when we look at the responses to this question? A lot of noise. In order to process it, I went through the responses, one by one, and tagged them by which channel I thought matched most closely to the answer given.

In the data was an entirely new category that is completely invisible to web analytics: earned offline media. Responses such as “A friend told me”, “Colleague”, “Saw you at a conference”, “Recommendation”, and “Word of Mouth” filled this category, all things that are invisible to web analytics. I built a new chart based on the tagging, and here’s the breakdown:

Screenshot_11_14_13__6_30_AM

Notice how search has now declined to one third of conversions? Notice how offline earned media, the orange slice, is a full 20% of our conversions for the quarter? That was completely invisible to our web analytics, and had we made 2014 budget allocations and decisions without taking that into account, we would have overfunded some marketing channels and underfunded others. The reality is that earned media – our PR efforts on behalf of SHIFT, from events to brand building – account for 20% of our lead generation. Had we relied solely on our web analytics, we would never have seen that and might have made the seemingly logical choice to prune our internal PR budget – and then wondered why revenues in 2014 were down the same 20%.

Take away two lessons from this: first, always ask people how they heard about you. You will often get responses that will surprise you and enlighten you about what’s really working. Second, don’t blindly rely on web analytics and clickstream data. There is more to understanding your audience than just what they click on – and a significant portion of your revenue could rely on that understanding, as ours does.

Disclaimer: As with all things data, your mileage may and should vary. This is based on our internal data and is representative of only our data and audience.

Christopher S. Penn
Vice President, Marketing Technology

Be Ballsy. Work at SHIFT. Apply now!

2 comments
jmcvearry
jmcvearry

Great Piece.. Definitely think the new focus on actual customer experience is going to make looking at data through this lens increasingly important.