Broken Traffic Estimator

[AussieWebmaster]

If anyone is using the Google traffic estimator to try and determine pricing and traffic I should tell you "IT DOES NOT WORK".
If you have played with it you would realize that it throws incorrect numbers. So do not base decisions on it.
AWR - when are you guys going to fix it?

[AdWordsRep]

If anyone is using the Google traffic estimator to try and determine pricing and traffic I should tell you "IT DOES NOT WORK"...

AWR - when are you guys going to fix it?

AussieWebmaster, I can assure that work is afoot to make the Traffic Estimator more useful and accurate. I'm not able to speak to a specific timeframe, though.

A few general notes on the Traffic Estimator:

* The intent is to give advertisers a ball-park estimate of what may happen in the real world, to get them started.

* It is always best to rely on actual statistics once the account is running.

* The Traffic Estimator works to give you a snap-shot of a moment in time, in a program in which nothing stays the same from one moment to the next.

* It is important to aware that the Traffic Estimator gives you an estimate across all countries that you have targeted. However, each country is a different competitive marketplace, and traffic and position can vary hugely from country to country.

With all that said, I invite any and all to post in more detail about how the Traffic Estimator is not useful to you, and I'll make sure your comments reach the right ears.

AWR

[AussieWebmaster]

I can use it when I want to lower a bid... they say at $4.00 I am getting 140 clicks but at .50 cents I will get 150.... seems broke to me or it is true that more people click the 5 spot than the top spot.

[AdWordsRep]

Just my opinion here, but of all the information that the Traffic Estimator provides, the number of clicks per day that it suggests is the least meaningful.

Why? Well, the number of clicks you get per day for a keyword is almost entirely dependent on how well targeted the ad is to that keyword - and how well written the ad is, of course. But the Traffic Estimator will have no way of judging these factors.

For example, if you enter the keyword 'software' into your keyword list, the Traffic Estimator will probably tell you that you'll get hundreds of clicks. (Because many thousands of searches will be done each day on that keyword - or variations. And even a average CTR would likely give you hundreds of clicks - or at least the Estimator thinks so.) However, if you then show an ad about really soft cotton socks, you'll probably get only a click or two at best. And if the ad is badly written, and has poor grammar and spelling, then you'll probably get zero clicks.

Hope that makes sense.

BTW, if I thought that a keyword would work well, I wouldn't let the Traffic Estimator 'talk me out of it' with an estimate of low clicks. I'd just run the keyword at a comfortable Max CPC and see what happens in the real world.

AWR

[OptimizeOnline]

* It is important to aware that the Traffic Estimator gives you an estimate across all countries that you have targeted. However, each country is a different competitive marketplace, and traffic and position can vary hugely from country to country.

Following on from this ... I often want to find the maximum bids that are being placed for a UK campaign vs a global campaign, for example. Rather than use the Traffic Estimator my approach is:

a) open an existing client campaign for which ads will only show in the UK
b) go to any AdGroup and select 'Edit Keywords'
c) change the maximum bid to a very high level and replace the keywords with those for which maximum bids are required
d) Select 'Estimate Traffic' and extract the data I need
e) Finally, and most important of all do not save!

I would then follow the same process for a client's global campaign to obtain comparative figures.

[AussieWebmaster]

Following on from this ... I often want to find the maximum bids that are being placed for a UK campaign vs a global campaign, for example. Rather than use the Traffic Estimator my approach is:

a) open an existing client campaign for which ads will only show in the UK
b) go to any AdGroup and select 'Edit Keywords'
c) change the maximum bid to a very high level and replace the keywords with those for which maximum bids are required
d) Select 'Estimate Traffic' and extract the data I need
e) Finally, and most important of all do not save!

I would then follow the same process for a client's global campaign to obtain comparative figures.
The funny thing about it is not when you are increasing a bid but rather decreasing that the Estimator has the most problems.

[Mikkel deMib Svendsen]

I understand that it is very difficult toi estimate the number of clicks to expect from a certain listing. There are simply too many variables - and they can change any second 24/7

However, calculating how many impressions to expect should be much easier to do. Google should have keyword data going years back. By trending that data I know you would be able to give us advertisers a tool that could predict impressions very precise. It would even work to predict searches several month ahead - for example leading up to Christmas.

If I just know the impressions then I can do the math. I think I would be able to do a better job than the estimator in most cases :)

[AussieWebmaster]

I understand that it is very difficult toi estimate the number of clicks to expect from a certain listing. There are simply too many variables - and they can change any second 24/7

However, calculating how many impressions to expect should be much easier to do. Google should have keyword data going years back. By trending that data I know you would be able to give us advertisers a tool that could predict impressions very precise. It would even work to predict searches several month ahead - for example leading up to Christmas.

If I just know the impressions then I can do the math. I think I would be able to do a better job than the estimator in most cases :)
They should bring you in for a couple of mill... worth it in the long run.

[Mikkel deMib Svendsen]

They should bring you in for a couple of mill...

I don't go that cheap :)

[AussieWebmaster]

I don't go that cheap :)
I am impressed... I could have Paris for less

[The Generator]

If cutting costs is a big issue, I would look for keywords that don't have many competitors (i.e. 2 - 4) and just put a minimum bid to come in last place. However if you try this for words with many competitors, it's very probable that your ad will get fewer impressions, if any at all. :)