View-through vs. Click-through

10 12 2009

View-throughs (VT) are great when discussing with your boss! You can show a kick-ass campaign with a tiny tiny cost per acquisition (CPA). The problem is that it is, that 10 VT sales are not always the same as 10 extra sales. And your boss might notice that after a while (at least if he is smart).

I have always had a problem with the metric view-through. What is this metric exactly describing?

Every time I am looking at my stats from my display ad campaigns I always get a little dizzy when I come to the VT’s. I mean CT’s a quite easy to understand. A user clicked at your banner and converted on your site. Boom – that’s it.

A VT on the other hand is a lot more hard to understand. It could be someone who surfed through a page where your banner was served at the bottom, and the user never actually saw you ad. Nevertheless the user converted 5 days later of other reasons. But the user still has your cookie and will be counted as a VT sale except if he is coming through another marketing campaign such as Adwords.

This problem is even bigger when you ads are served at pages with lots of traffic like the main national boulevard-press website like The Sun or eb.dk here in Denmark. There is a good change that almost all surfers get by at some point and get your cookie placed at their computer.

And that is the time when your boss is paying you a second visit asking you why your other channels suddenly are falling while your great campaign is top performing. The VT’s are simply taking credit for all the natural conversions as well even though the traffic was there before the campaign.

So should I just skip the metric and focus on CT’s? This will cause the problem that the display ad channel will look so gargantuan expensive compared to decisional channels like the search engines. So de facto this is the same as closing down the channel.

Should I just cut the display ad channel then? Actually I tried that for a while already ;-) That gave me a significant lower CPA but unfortunately my SEM-campaigns suddenly got a lot more expensive and the volume in search engines in general fell. Doh!

I have turned on the display campaigns again. I have developed a special excel sheet trying to compensate for all the flaws. I’m measuring my upper-funnel on more parameters that just conversions and all in all trying to compensate. I’m not happy though!

Guys, how are you handling the problem?


Actions

Information

4 responses

10 12 2009
Morten E. Wulff

Have you talked to Ulrik and Martin @ TraceWorks about it?

11 12 2009
Anders Bach

I reached out for the heavy artillery a called the fearless Traceworks. We will work out an attribution model and hopefully get a little more grip of realities. Thanks!

11 12 2009
uberVU - social comments

Social comments and analytics for this post…

This post was mentioned on Twitter by andersbn: New blog post: View-through vs. Click-through http://bit.ly/7brwD9

29 12 2009
Kasper

We have never paid much attention to VT sales due to the same reasons you have already mentioned. Sometimes though I use VT sales as an extra parameter on specific niche medias – but only as an vague guideline. Some 3party hosting system, like Adform, is able to distinguish between standard banner impressions and actual banners shown in the active window of the browser. If you use such a hosting system then it makes more sense to trust in VT sales. But it is still too easy to cheat the setup, and I would never recommend any one making cpo deals on VT sales – unless they sell space rockets or drilling rigs.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s




Follow

Get every new post delivered to your Inbox.

%d bloggers like this: