I am trying to track conversions between two websites. One site (A) sends referrals (content site) to another site (B - e-commerce platform) which processes the actual transaction.I tried to setup up google analytics gooals (site A) to track conversions between sites A and B but it seems (maybe I am wrong)as though analytics can only track conversions on one domain.

On to the links:

* Web Metrics Demystified at Occam’s Razor by Avinach Kaushik
* Web Analytics Demystified at Occam’s Razor by Avinach Kaushik
[...]

Firstly, I love a word coined back in 2003 by Jack Aaronson (http://www.clickz.com/showPage.html?page=3622794) which is “Channibalism” :

“A basic fear felt by many executives and managers that one channel, or line of business, will steal business away from another. There are two facets to this fear: one tied to fears of consumer confusion and loyalty and one tied to managerial salary structures.”

If we take that last point about salary structure as painfully valid then we have reached the core of the multi-channel tracking problem: If I’m an on-line marketing manager in silo-marketing-cover-your-own-butt-Inc why would I want to drive a prospect to a store or call center if I don’t get my commission.

I was speaking to a colleague who recently joined an organisation that only made 5% of its sales on-line, despite being a really efficient and capable site with a target audience of IT professional. The reason - nobody got commission for on-line sales, but the personal rewards of driving people to the call center or the key account team were high.

So is this where we really stand with integrated marketing, that the company conspires to do what it best for the individual stakeholders?

Interestingly, I have been doing some work with a few organisation (mail order, travel, financial services) where they find the conversion rates AND average order values are higher if they push people to a call center at specific stages in the process. Why? Maybe people are better at persuading prospects to upgrade to “champagne and flowers on arrival” than a radio button.

So there we have it…we can dynamically create unique phone numbers for “generic” versus “brand” PPC ad’s so we can track off-line conversion, and we can bribe people to type in vanity url’s from magazine ads to get an extra 5% discount, but if we’re not getting a bigger piece of cheese at the end of the year why don’t we just push ‘em through the web funnel and life will be good.

In Web Analytics Avinash makes the point that we can trade value for insight to track off-line outcomes, but maybe nobody does if it affects their bonus?

Ho hum.

What I find utterly interesting, yet complicated to contemplate at the moment is how companies will progress in blurring the divide between offline and online data. I also read your post on non-line marketing (courtesy of David Hughes). There seems to be a big push in the advertising and marketing worlds to make this divide seamless, pooling these vasts amount of data and technologies together so that the distinction becomes even more difficult. The potential seems fantastic, yet where I struggle to make ends meet is in the reconcillation. This is why I thought your posting on predictive analytics was quite interesting.

As for 2008, I’ve only just made a New Years Resolution (yes a little late I know) and that is to learn all I can about this sector, I’ve been a generalist for quite some time, now I’ve found a subject that really stimulates the grey matter and presents constant challenges. To help this I’ve decided to buy myself a late Christmas present in the form of your book.
Thanks again

Phil (not Peter)
perennial thinker at http://think-train.com

My overall feeling is that the segment itself will continue to explode. The more I meet people, at companies big and small, the more I realize how much more data needs to be used. The good thing is that people are finally starting to see its importance. So great for Vendors, Practitioners and Consultants. It has been hard to make money on Web Analytics, this will be less of a issue for the next couple of years (for all three groups).

The other thing I see accelerating is the concept of Web Analytics 2.0 (http://www.webanalytics20.com). The essence of it is that we are going to move more aggressively beyond simply clickstream data (which is like 99% of what web analytics is today). We will realize how the other components of that model proposed are important and finally data can start to drive action rather than languishing.

The problem with Web Analytics 1.0 is that it is a exercise in data torture and reporting with long lags in taking action (if any). Data torture needs to get automated and expanded, decision making needs to get automated (think MVT, BT etc), people need to be left for smart hard things (vs what happens today!).

So two things for you to think about. Two trends that will accelerate in 2008.

On the second point of your comment, it is interesting to think of applying Game Theory to understand visitor behavior. I think the core problem is that when you say Visitor you are probably thinking of a Person, the reality is that in web analytics Visitors are “cookie_id’s”. When you come to grips with the difference you realize that there is a lot less to learn about “Visitors” then might be apparent on the surface!

Here is a post that might be interesting in that context (though predictive analytics does not exactly have much in common with game theory):

http://www.kaushik.net/avinash/2007/09/data-mining-and-predictive-analytics-on-web-data-works-nyet.html

Thanks again,

Avinash.

I wanted to ask you what do you forecast for the web analytics/metrics world for 2008? more of the same or something exciting on the horizon?

Do you think that Game theory can help to understand visitor bahavior?

Best Regards and keep it up….

Dabei sollte die perfekte Metrik folgendes sein:

unkomplex
relevant
aktuell
sofort nützlich

Zum Blogeintrag: Link
[...]

One of the few blogs that I actually follow and read is this site, called Occam’s Razor by Avinash Kaushik. In one of his latest posts, he talks about Web Analytics and Web Metrics, two things that I am very interested in. I truly admire people who can make complex things easy to understand and this site wins my admiration…

Its a great post. Really covers broad criteria and relavant examples. thanks for information.

I am looking for a part 2 of this post that delves into looking at a specific tool say google analytics and explore and look into what that information tells you. Then another post about how exactly you derive your conclusions from the reports and go about fixing some typical common problems.

Am I asking for too much.

I guess so. Like you I am also greedy.

If you re not planning to post it then I ll digg for more information and see what I can post on my blog.

cheers

vineet nair

http://internetbusinessjunkie.blogspot.com

Another awesome post….really useful insight and your book is fantastic,especially with your constant updates over here.

cheers,
deric

My summary for you:

See if you can find any external benchmarks that you can use to see how others are doing and use that as a critical input. If you can’t find external benchmarks then use your own historical performance and trends there as a guide, just remember to add improvements expected due to investments in people / tools / processes.

Now here is the cheat answer in reply to Robbin’s question about how to create goals specifically for conversion rate….

From Three Interviews, Six Interesting Web Analytics Questions.

1) A: Sign up for the shop.org annual study and look at what your competitors are doing. Use that as a initial discussion starter of what your conversion rate should be.

1) B: Type “fireclick index” into Google and look at last year’s worth of data for conversion rate for the web or for one of the six vertical industries that they provide. It is free. Use that as a starting point for discussion of what your goal should be.

2) Plot out your conversion rates (segmented by your core acquisition strategies - DM, Email, PPC, Display, whatever) for the last year and see where things are trending. Bring this to your fireclick/shop.org discussion.

3) Finally see where in your acquisition strategy or site optimization you are making increased investments. If you just hired a SEM Goddess pump up the goal by 50% for that stream of traffic (Goddess will deliver). If you are implementing MVT then see what that will do.

1 + 2 + 3 = An intelligent discussion.

You’ll come up with a goal for the next three months. It might be wrong but persist and repeat the process three months later, you’ll do better this time. In six months when you do it you’ll nail it.

Give yourself permission to be wrong, trust me you’ll get better so fast.

Hope this helps!

-Avinash.

I’m going to archive this post for reading again. Thanks

It is crucial to set metrics and polish them with your process. I’m wondering what should we do with goals? You can have the best metrics in the world but if your goals are not well define, there are useless. So, how do you set goal? you can use litteraure or historic data, but as you said, each webiste is unique and evolve, so what can we do? Should we look only on variation between two periods? What is a significant variation? is it 15% or 50% ?

What I’ve run into the most is when, in an effort to aim at “Relevant,” the discussion stomps on “Uncomplexity” (or simplicity!) like a cockroach skittering hastily across the kitchen floor.

Rather than stopping at, “Given our business’s strategy, is this metric relevant?” the question becomes, “Given that, at the end of the day, we are concerned with revenue and profitability, how can we ensure that all of our metrics track directly back to one of these two corporate KPIs.” The result is metrics that either never see the light of day, or metrics that include so many leaps of logic in their calculations that nobody pays attention to them.

Based on several recent experiences, I blogged a highly tongue-in-cheek post on a “simple framework for establishing corporate metrics:” http://tinyurl.com/2eca7x. It’s broader than web metrics, but most of your posts can be applied much more broadly! It dives a bit deeper into how NOT to search for relevancy in metrics.

When I saw the ClickTracks screen shot, it reminded me of my wish that Google Analytics offered the same kind of keyword trend report. How much more productive it is to let the software illuminate the meaningful and actionable trends, as opposed to a static matrix that’s simply a snapshot in time.

–dave