What Really Drives Web Advertising ROI

The old adage “When you only have a hammer, everything looks like a nail” could certainly apply to web advertising. Like a hammer, click thru rate (CTR) has been the core metric used by many Marketers to measure web ad effectiveness since the early days of the web.

Does Click Through Rate Predict Sales Lift?

But with growing evidence that CTR is not the “hammer” of web effectiveness it was once deemed, marketers are learning that  no metric may be more meaningful than a poor one.

What Marketers Should Be Measuring

Marketers should be measuring brand building and business outcomes. Growing brand equity and consumer engagement, along with volume and share, should be at the top of the priority list for every CMO. Ultimately, these need to be translated into a return on Marketing investment (ROMI) measure, just like any other investment.

But because it has been so hard to measure these things, even on the web, Marketers have often settled for what they thought was the next best thing—in this case, CTR. And what made CTR particularly appealing was that it measured a consumer action. What could be better than that ?

The Real Impact of Web Advertising

Improvements in measuring and monitoring consumers web activities, coupled with off-line purchase panel data, now gives us insight into what really works.

How can web advertising ROI really be measured ?

  • Web Behavior – Consumers are provided software which, when downloaded, monitors and provides a comprehensive view of their web behavior (consumers opt-in and understand this when they agree to use the software).
  • Off-Line Purchase Behavior – These consumers also participate in an off-line purchase panel. They shop as they normally would shop. Once home, they scan their purchases into a purchase panel database using a handheld wand.
  • Fused Data – The web behavior and off-line purchase panel data sets are then fused together so that the they can be analyzed to determine cause and effect relationships.
  • Test vs. Control Analysis – Ancova (analysis of co-variance) analysis is then performed on the test (those consumers seeing an ad) vs. control (those not seeing one) to determine the single variable impact of the advertising.

So what does the data tell us?  Nielsen reviewed 28 CPG campaigns using the above methodology (disclosure: I work for The Nielsen Company). Surprise: there is virtually no relationship between CTR’s and volume growth (correlation < .10). What is clear is that campaign reach had a significant impact on likelihood for success: higher reach campaigns were more likely to drive business growth than lower reach campaigns.

Click Through Rates - Not Predictive of Sales

Learnings for Marketers

CTR is not a good metric in predicting business growth for web advertising. Additionally, it’s clear that Marketers should put more emphasis on developing and running campaigns with broad reach, as campaigns with higher reach tended to deliver higher ROI’s.  And while the data says nothing about the creative itself, you have to believe that the primary driver of business results is the quality of the value proposition and creative– is it special, different and better ?

There’s another important learning here though. In the quest for bringing greater accountability to the Marketing function—a laudable goal—Marketers also need to be careful not to adopt metrics for the sake of having metrics. As the data above shows, even consumer behavior based metrics may not be meaningful.

Which brings us back to the original point. The CMO’s job is to build the brand and the business, and not just to hammer a bunch of nails. 

So what’s in your brand’s tool kit?

Get free updates of Randall Beard’s Blog by e-mail

Get free updates of Randall Beard’s Blog by RSS reader

2 Responses to What Really Drives Web Advertising ROI

  1. Eric Jens says:

    Hi Randall,

    Interesting learning. As I look at the chart, the other question that would be interesting to answer with the dataset is: “what is the correlation between CTR’s and achieving payout?”


    • beardrs says:

      Hi Eric — Thanks for reading the blog and commenting. I’ve looked at quite a lot of data on CTR, and one compelling figure is this: only 16% of the internet population represents 80% of ad clicks. CTR is not only not correlated with ad campaign sales lift, it’s also not correlated with subsequent brand search activity or brand website visits. My takeaway is that there is a fairly small group of web users who click thru and whether they go on to do anything meaningful beyond clicking or not, they’re not large enough to make an appreciable difference on any relevant brand metric. I have to believe this includes payout as well. Thanks again for your comments and keep them coming. Happy holidays !

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: