Google Analytics add Benchmarks - The China View
So many great news today! First Omniture's announcement about the Baidu, then the news that Google Analytics has a new Benchmarking function. Lets talk about the GA Benchmarking first, so I can focus on Omniture / Baidu in a longer post during the weekend.
Why Benchmarking?
Reporting is easy, providing actionable recommendation is less so. In an earlier past I recommended a 3 step approach based on answering 3 simple questions:
Why is this announcement so important for China?
Regular readers of this blog might remember my rants about the lack of benchmarks in China. While you can benchmark you reach (unique visitors) with some accuracy using IResearch's iUserTracker, key industry benchmarks like CPC, CTR or Bounce rate are unavailable in China. There is not even an industry organization like the IAB that conceivably could report such numbers. As long as you don't have a large sample of clients or long experience in the market, you are out of luck (just barely avoided a 4 letter word here)
How useful will it be?
Well, that depends. I just singed up for the service (you need to agree to share you own data anonymously first) and will share more comments in the future, but I see a number of limitations.
[Update: My friends at Google confirmed that the benchmark is global. No luck for the wicked]
Why Benchmarking?
Reporting is easy, providing actionable recommendation is less so. In an earlier past I recommended a 3 step approach based on answering 3 simple questions:
- Is this number good or bad?
- Why is the numbers good or bad?
- What can we do to improve this metric and is it worth the effort?
Why is this announcement so important for China?
Regular readers of this blog might remember my rants about the lack of benchmarks in China. While you can benchmark you reach (unique visitors) with some accuracy using IResearch's iUserTracker, key industry benchmarks like CPC, CTR or Bounce rate are unavailable in China. There is not even an industry organization like the IAB that conceivably could report such numbers. As long as you don't have a large sample of clients or long experience in the market, you are out of luck (just barely avoided a 4 letter word here)
How useful will it be?
Well, that depends. I just singed up for the service (you need to agree to share you own data anonymously first) and will share more comments in the future, but I see a number of limitations.
- The benchmark needs to be relevant to my industry, GA allows you to choose among a number of verticals to address this issue. So the value will depend on how close this match is.
- The benchmark needs to be relevant to my geography. Most
benchmarks are widely different in China than they are in the US (or in
Japan, or in Germany for that matter). So far it seems this problem is
not addressed. That is a key weakness for us poor web analytics souls
in China an without this feature, GA benchmark will remain a nice graph
is the system.
[Update: My friends at Google confirmed that the benchmark is global. No luck for the wicked]
Chinese marketing research lacks benchmarks, it will be a long trip to educate people getting used to it. But definitely adding more comparison choice with authentic third party data would be great. Historical and competitive, make everyone have concern in them.
Reply to this
GA benchmarks - I suspect - are phony. They are just the way Google chose to convince people that their (people's) high bounce rate is "better" than benchmark so that Google can keep on clicking (fraudulently) and can even increase fraudulent clicks and bring the bounce rate up to the "benchmark"...
Just a hunch.. but I have been a google customer long enough to trust my hunches.
Reply to this
Dear Guest, That for commenting, and welcome to Chinalytics. Can you clarify what you mean with phony? Do you expect the data to be randomly made up, or selective, or ...? As I mentioned in my post, I certainly agree that the current value of Google Analytics benchmarks is limited. We always need to qualify the data in our reports.
That being said, I see little incentive for Google to fake this data. If my bounce rate is better than the benchmark, does that mean I spend more money on AdWords? Not really, only if the AdWords or AdSense bounce rate is better than that from other traffic sources. On the contrary, the way I see it, Google has a inherent incentive for all websites to be better converters (lower bounce rate etc), since better conversion mean more money for online advertising. So if anything that should show my site as performing below the benchmark.
Am I completely of the mark here? Let me know in the comment section.
Reply to this
That's a very informative post, I understand that benchmarks are limited in China but this is also an important step forward, nobody expects perfection from the first try.
Reply to this
Google benchmarking is really an interesting thing, which is more significant that any other competitors metrics. Although it is in beta mode now but people are gaining a lot. If i talk about myself, I approached it to have an eye on pay per click industry and it really worked. Thanks Google
Reply to this