Error prone affiliate tracking with HotelsCombined

[martin-s]

Is anyone else struggling to get affiliate click-throughs tracked accurately with Hotels Combined?

After bouncing back and forth several support emails, I've been able to confirm that:

* Only new visitors are reported with their dashboard statistics (not greatly helpful). Return visitors are not reported at all.
* A cookie is solely responsible for tying a visitor to an affiliate.
* The cookie lasts 365 days, but is replaced if a visitor is subsequently tracked as coming from another affiliate (last contact wins).
* Stats operate in near real-time (within a minute or two).

Currently their stats are only reporting around 20-25% of the click-throughs that we logged via Google Analytics.

I have now tested and proven that their cookies are not accurately tracking all visitors using three different methods:

* Different PCs
* Different browsers
* Chrome Incognito mode

Has anyone else experienced issues like this? I'm having difficulty getting them to take it seriously.

 

[chippyb]

Change program then, im using booking and seems to be performing much better than laterooms

 

[martin-s]

We tracked a similar problem for booking.com and they don't use cookies, which I'm not a fan of. The earnings were very erratic with them, we could sometimes go three weeks with nothing. At HC, the earnings are every day despite the problems with their tracking.

 

[RobM]

I thought hotelscombined were PPC and booking.com were commission based?

 

[martin-s]

Yep, that's right

 

[chippyb]

Yea, and isn't the value you getting for a click really low? was when I tried it.

See I have not had that problem to booking so far, much better than laterooms.

Now seeing bookings everyday where I didnt with laterooms. Im just surprise I have a hotel site in Google still ranking if im honest that gets traffic!

 

[martin-s]

Probably averages $0.70-$0.80 per click

 

[Edwin]

Probably averages $0.70-$0.80 per click

You need to normalise any affiliate stats back to a standard CPM because seeing just a per click value tells you nothing since it's dependent on the click count reported by an affiliate program.

If you know that a given website had 10,000 pageviews in a month, and earned £50 in commissions from affiliate program A, then you effectively earned £5 CPM from that affiliate program.

If you use the same analytics to measure another 10,000 pageviews with affiliate B's ads in place, and you earned £0.50 per click with 80 clicks, well then you just earned £40 from that traffic which is £4 CPM.

£5 > £4 so affiliate program A is "better" for your particular site.

Without normalising the stats, you'll never know.

Incidentally, domain parking programs also suffer from the same issue. They report CPM, CPC etc. but you never know what they are or are not counting as a "visitor". The only reliable way to compare parking programs is to run each one for the same amount of real traffic and see which one gives you the most MONEY for that traffic. In other words treat each parking program as a "black box" - you feed 1,000 visitors into the box, and £X comes out. You feed 1,000 visitors into a different parking company's black box, and £Y comes out.

At the end of the day, it doesn't matter how many click, visits, pageviews, sales, leads etc. are recorded - all that matters is your actual revenue per actual unit of traffic (and how that compares to the revenue you would have got by sending exactly the same amount of traffic somewhere else).

 

[damion]

* The cookie lasts 365 days, but is replaced if a visitor is subsequently tracked as coming from another affiliate (last contact wins).

Frustrating as it can be, this is a normal practice with all affiliate programs.