Attribution Question For You

ClickEquations Blog

A Serious Look at Paid Search Marketing Strategies, Tactics & Tools

By ClickEquations | January 13, 2009

The issue of how to allocate credit across the different searches (or other visit types) that lead up to a conversion event is a deservedly hot topic.

This is true largely because the broadly used ‘last click’ allocation model (where the last keyword gets 100% of the revenue credit) is really inaccurate and inadequate.

There are several other attribution models, including first click, linear, weighted, and other hybrids. The pros/cons of each are worthy of extensive discussion. That’s not the topic here however. We have a related question and would like to get some feedback.

Should the allocation of credit ever extend beyond a single conversion?

Two examples below further frame the question. In both cases, assume there is a user-defined tracking period which applies. In other words if you’ve set a ’30-day tracking period’ in any case the look back for events would only extend back 30 days.

CASE 1

Search on KW1 –> Search on WK2 –> Conversion 1 –> Search on KW3 –> Conversion 2

Question: Should KW1 and KW2 be part of the allocation chain for Conversion 2? If so KW1 and KW2 could get partial revenue credit for both Conversion 1 and Conversion 2.

Or does Conversion 2 only allocate back to KW3?

CASE 2

Search on KW1 –> Search on KW2 –> Conversion 1 –> Conversion 2

Question: Should KW1 and KW2 be part of the allocation chain for Conversion 2?

Or should Conversion 2 be seen as an independent event (perhaps inspired by a follow on email or other interaction after Conversion 1.)

FYI, Google Adwords does allocate revenue to keywords from multiple conversions if no search happens between them.

Your Opinion?

We’d love to hear your opinion on this. Please take the survey below, and leave any other thoughts in the comments.

[wp_surveys]

http://www.webbanalysts.info Lars

This is a very good question, and not an easy one to answer. Another question is what share of the credit each search should get. rnrnAlso, in countries where Google has a search share of 70-80% it’s safe to assume that Adwords often overrides banners and other forms of advertising. In other words, Google has the luxury of “stealing” credit from other ads (as measured in many web analytics tools) simply because searching is so common.rnrnClick on banner ad #1 -> Click on banner ad #2 -> Search on Google -> All credit to Google.rnrnWe see a banner, we click on it, we don’t buy right away. A day later we decide to buy, but we can’t remember the URL. We search. We buy. Kaching Google.rnrn
http://www.webbanalysts.info Lars

This is a very good question, and not an easy one to answer. Another question is what share of the credit each search should get.

Also, in countries where Google has a search share of 70-80% it's safe to assume that Adwords often overrides banners and other forms of advertising. In other words, Google has the luxury of “stealing” credit from other ads (as measured in many web analytics tools) simply because searching is so common.

Click on banner ad #1 -> Click on banner ad #2 -> Search on Google -> All credit to Google.

We see a banner, we click on it, we don't buy right away. A day later we decide to buy, but we can't remember the URL. We search. We buy. Kaching Google.
http://www.acteva.com Greg

Your Question: Should KW1 and KW2 be part of the allocation chain for Conversion 2?

My Answer: Only if they contributed to the conversion.

This is a trick question. You have not provided enough information to give an answer.

Show me a few thousand cases of people clicking on various ads in various different sequences, who buy at different times along the way, and I'll analyze it and give you the answer.

Maybe 60% of the people who see Ad #3 will then buy, and the data shows that it makes no difference whether they ever saw Ad #1 or Ad #2. Or maybe people who saw Ad #1 are twice as likely to buy when later they are exposed to Ad #3. Without the data, it's impossible to say.

Is this a guest post by the Mad Hatter? : )

This blog is amazingly great, one of my favorites, but today's question throws me for a loop!
http://www.bgtheory.com/blog/ brad

What do you think?rnrnI’ve been pondering this for a while (thanks for the post); and I think part of this comes back to lifetime visitor values. If someone searched and converted again on the same site – makes me think that this company wasn’t top-of-mind as a direct return visitor; and makes me think that a better loyalty program needs to be put into place if it happens often (it’ll happen some regardless).rnrnIf product 2 was complimentary to product 1 — meaning that by buying the first product the company was more likely to do business again with the same product – then the initial keywords should get some attribution.rnrnIf product 2 was completely different, and the initial keyword didn’t help increase the conversion process for the 2nd buy – they the first keyword shouldn’t get any attribution.rnrnNow – how do you automate and scale that type of decision?
http://www.bgtheory.com/blog/ brad

What do you think?

I've been pondering this for a while (thanks for the post); and I think part of this comes back to lifetime visitor values. If someone searched and converted again on the same site – makes me think that this company wasn't top-of-mind as a direct return visitor; and makes me think that a better loyalty program needs to be put into place if it happens often (it'll happen some regardless).

If product 2 was complimentary to product 1 — meaning that by buying the first product the company was more likely to do business again with the same product – then the initial keywords should get some attribution.

If product 2 was completely different, and the initial keyword didn't help increase the conversion process for the 2nd buy – they the first keyword shouldn't get any attribution.

Now – how do you automate and scale that type of decision?
http://clickequations.com Craig Danuloff

Totally agree that 'Lifetime' vs 'Click' attribution/allocation is the way to solve it. Very interesting idea to make decisions based on the product – keyword relationship. Assumes some meta data that may or may not be available. But your point is exactly right, how in the world could you automate that with any precision – we're looking at the same issue for pausing keywords based on inventory levels – great idea until you realize that the keyword-sku relationship isn't linear.
http://www.bgtheory.com/blog/ brad

What do you think?

I've been pondering this for a while (thanks for the post); and I think part of this comes back to lifetime visitor values. If someone searched and converted again on the same site – makes me think that this company wasn't top-of-mind as a direct return visitor; and makes me think that a better loyalty program needs to be put into place if it happens often (it'll happen some regardless).

If product 2 was complimentary to product 1 — meaning that by buying the first product the company was more likely to do business again with the same product – then the initial keywords should get some attribution.

If product 2 was completely different, and the initial keyword didn't help increase the conversion process for the 2nd buy – they the first keyword shouldn't get any attribution.

Now – how do you automate and scale that type of decision?
http://clickequations.com Craig Danuloff

Totally agree that 'Lifetime' vs 'Click' attribution/allocation is the way to solve it. Very interesting idea to make decisions based on the product – keyword relationship. Assumes some meta data that may or may not be available. But your point is exactly right, how in the world could you automate that with any precision – we're looking at the same issue for pausing keywords based on inventory levels – great idea until you realize that the keyword-sku relationship isn't linear.

ClickEquations Blog