Comments for Last Exit Westend http://www.westendintie.net/michael/blog A text repository about Web Analytics and Digital Services. Tue, 07 Dec 2010 14:37:45 +0000 hourly 1 http://wordpress.org/?v=3.0.1 Comment on Why the idea of a “correct” revenue attribution model doesn’t resonate with me by Chris Brinkworth - TagMan http://www.westendintie.net/michael/blog/2010/11/why-the-idea-of-a-correct-revenue-attribution-model-doesnt-resonate-with-me/comment-page-1/#comment-639 Chris Brinkworth - TagMan Tue, 07 Dec 2010 14:37:45 +0000 http://www.westendintie.net/michael/blog/?p=331#comment-639 Michael, Yes. You can use direct to site and repeat visits also. Email me on chris.brinkworth@tagman.com for 'on stage' presentation from client and how they use the data with their offline channels also. Best, Chris Michael,

Yes. You can use direct to site and repeat visits also.

Email me on chris.brinkworth@tagman.com for ‘on stage’ presentation from client and how they use the data with their offline channels also.

Best,

Chris

]]> Comment on Why the idea of a “correct” revenue attribution model doesn’t resonate with me by Michael Dlugosch http://www.westendintie.net/michael/blog/2010/11/why-the-idea-of-a-correct-revenue-attribution-model-doesnt-resonate-with-me/comment-page-1/#comment-633 Michael Dlugosch Thu, 02 Dec 2010 07:30:32 +0000 http://www.westendintie.net/michael/blog/?p=331#comment-633 Hi Steve, you hit the nail on the head. The attempt of attributing hard facts ("Cash!") to basically unlikely conversion events based on a mixture of soft factors ("... what does it mean if the conversion rate is 0,05% higher for the campaign. Is that a success?") just blurs the problem, but doesn't solve it. But imagine what a proper control group setup would have to look like these days: a panel of mystery shoppers who are obliged to NOT join any social network. Eventually even allured with monetary incentives: "And? What do you do for a living?" - "Oh, I'm a professional social media denier." My oh my... :) Thanks, Michael Hi Steve,

you hit the nail on the head.
The attempt of attributing hard facts (“Cash!”) to basically unlikely conversion events based on a mixture of soft factors (“… what does it mean if the conversion rate is 0,05% higher for the campaign. Is that a success?”) just blurs the problem, but doesn’t solve it.
But imagine what a proper control group setup would have to look like these days: a panel of mystery shoppers who are obliged to NOT join any social network. Eventually even allured with monetary incentives: “And? What do you do for a living?” – “Oh, I’m a professional social media denier.” My oh my… :)

Thanks,
Michael

]]> Comment on Why the idea of a “correct” revenue attribution model doesn’t resonate with me by Steve http://www.westendintie.net/michael/blog/2010/11/why-the-idea-of-a-correct-revenue-attribution-model-doesnt-resonate-with-me/comment-page-1/#comment-612 Steve Tue, 30 Nov 2010 21:29:23 +0000 http://www.westendintie.net/michael/blog/?p=331#comment-612 Mister Dlugosch :), Great points and I fully agree. I don't think first click or last click attribution are 100% valid as indications of what influenced the visitor to buy either. I think all the tools trying to measure this are flawed in that sense at least - it's simply another case of tools measuring things simply because they can. View throughs for instance. Overrated for this purpose in my opinion. Additionally AIDA or IADA as you discussed are also largely irrelevant as they are only effective when measuring the linear conversion campaign as you said and even then comparisons are difficult to reconcile due to seasonal fluctuations or other factors you can't control. I think the only sure way to measure the effect of a campaigns influence on the intent of the purchaser is by going back to the old direct database marketing method of keeping a control group of customers who you're sure have never been subjected to any campaigns. That way you have an apple to apple comparison of people who were and weren't subject to your ads and bought product/service. You can then see by comparing conversion rates if your ads truly made a difference and thus attribute success or failure to them. What practitioners might want to ask themselves is have they built an effective customer control group to nurture successful campaigning? rather than can I measure first/last click attribution? Of course creating these control groups in this day and age is getting harder all the time due to increasing exposure. Pesky social media! :) Cheers Steve Mister Dlugosch :) ,

Great points and I fully agree.

I don’t think first click or last click attribution are 100% valid as indications of what influenced the visitor to buy either. I think all the tools trying to measure this are flawed in that sense at least – it’s simply another case of tools measuring things simply because they can. View throughs for instance. Overrated for this purpose in my opinion.

Additionally AIDA or IADA as you discussed are also largely irrelevant as they are only effective when measuring the linear conversion campaign as you said and even then comparisons are difficult to reconcile due to seasonal fluctuations or other factors you can’t control.

I think the only sure way to measure the effect of a campaigns influence on the intent of the purchaser is by going back to the old direct database marketing method of keeping a control group of customers who you’re sure have never been subjected to any campaigns.

That way you have an apple to apple comparison of people who were and weren’t subject to your ads and bought product/service. You can then see by comparing conversion rates if your ads truly made a difference and thus attribute success or failure to them.

What practitioners might want to ask themselves is have they built an effective customer control group to nurture successful campaigning? rather than can I measure first/last click attribution? Of course creating these control groups in this day and age is getting harder all the time due to increasing exposure. Pesky social media! :)

Cheers
Steve

]]> Comment on Why the idea of a “correct” revenue attribution model doesn’t resonate with me by Michael Dlugosch http://www.westendintie.net/michael/blog/2010/11/why-the-idea-of-a-correct-revenue-attribution-model-doesnt-resonate-with-me/comment-page-1/#comment-605 Michael Dlugosch Tue, 30 Nov 2010 08:02:04 +0000 http://www.westendintie.net/michael/blog/?p=331#comment-605 Thanks for your comment, Chris. I have taken a look at the article. Yes, it makes perfect sense, and would allow as well to score pages people came across according to their relative importance within the path. How about repeat visits? Any possibility to attribute conversion paths across visits as well? Thanks for your comment, Chris. I have taken a look at the article. Yes, it makes perfect sense, and would allow as well to score pages people came across according to their relative importance within the path.
How about repeat visits? Any possibility to attribute conversion paths across visits as well?

]]> Comment on Why the idea of a “correct” revenue attribution model doesn’t resonate with me by Chris Brinkworth - TagMan http://www.westendintie.net/michael/blog/2010/11/why-the-idea-of-a-correct-revenue-attribution-model-doesnt-resonate-with-me/comment-page-1/#comment-598 Chris Brinkworth - TagMan Mon, 29 Nov 2010 15:01:45 +0000 http://www.westendintie.net/michael/blog/?p=331#comment-598 Michael, superb insight. Do, if you have the chance - read the econsultancy blog post about where TagMan client's are now planning on 'paths' - not channels, as I think you will find it useful in your pondering on what 'model' should resonate. There is no one size fits all. <a href="http://econsultancy.com/us/blog/6764-optimise-paths-to-conversion-not-channels" target="_blank" rel="nofollow">econsultancy.com/us/blog/6764-optimise-paths-to-conversion-not-channels</a> Michael, superb insight. Do, if you have the chance – read the econsultancy blog post about where TagMan client’s are now planning on ‘paths’ – not channels, as I think you will find it useful in your pondering on what ‘model’ should resonate. There is no one size fits all.

econsultancy.com/us/blog/6764-optimise-paths-to-conversion-not-channels

]]>