The reason I ask is this. I was playing about and tested a few adgroups, all with the same exact matched keyword. Adgroup A's [widget] has a QS of 7 with a min first page bid of 0.10 Adgroup B's [widget] has a QS of 7 with a min first page bid of 0.15 Adgroup C's [widget] has a QS of 9 with a min first page bid of 0.10 Adgroup D's [widget] has a QS of 6 with a min first page bid of 0.15 Adgroup E's [widget] has a QS of 7 with a min first page bid of 0.20 Now the reason for these changes is because they each have different Ads and different landing pages, but I would have thought that for the same keyword, surely the higher the Quality Score, the lower the first page bid would be? This doesn't appear to be the case. In the above - there is one where the keyword has a QS of 6, yet it's min first page bid is actually lower (.15) than anothers QS of 7 (0.20) Your thoughts?
Min. first page bid depends not only on your quality score. It also, depends on competitors quality score and bid for the particular keyword.
That doesn't follow, if you look at the above post. All the above adgroups were produced within minutes of each other - very unlikely competitor's QS would have changed multiple times in those 2-3 minutes.
I think the point is that, unless the "widget" in question is a highly volatile market and competitors are changing bids daily and by considerable amounts, the competitors' QS and bids can be considered fairly stable. The min first page bid is supposed to be what is required to get you onto the bottom of page 1, i.e. usually position 8, and that position is supposedly determined by QS x Max CPC bid, called Ad Rank. The variables for landing page, display url, ad performance, etc. are supposedly factored into the QS so Ad Rank should be more or less the same yet, from the OP's data, the Ad Rank varies between 70 and 140 for the same, exact match keyword. Yes, one would have thought but I think the fact is that the 1st Page Estimate is, and always has been, extremely inaccurate (though it does seem to be improving!) especially for the first day or two that an adgroup is running. So, how long did you run each ad? It would also be interesting to know the actual ave position each appeared in assuming you bid either the same for all or the minimums as detailed above.
Ha! Got it in before me. Okay, so I think that "within minutes" may be part of the problem. As I indicate above, the min bid estimate does seem very inaccurate to begin with. It would be interesting to know if the estimate and quality score "harmonize" after a few days but I am not sure there is any way of objectively testing (since they would all be competing against each other and effecting each other's CTR etc. )
For this test, the ads didn't run at all - they were only uploaded to see what Google would give for the QS and min first page bid in each instance. I'm going to do another one of these but will make it a fairer 'test' - I'll have say 2 different headlines, 2 different line 1s, 2 line 2s, 2 display URLs and 2-3 landing pages - and see the results for them all. Ultimately I'm wanting to know which of my websites, Google prefers and which landing pages in each of those websites, Google prefer - then which of the Ads Google prefer. Conversion rates and CTR will come into play and takeover but it's just to get off to the best possible start for new adgroups and campaigns. I.e Google may prefer Landing page A over B yet B might convert better - all this is just for the beginning of each adgroup. I'm wondering, if you put an Ad into an adgroup which Google doesn't particularly like, and gives, say a QS of 6 - it may not be possible to max out the QS to 10/10 - even with a very good CTR? In other words, each ad will be given a potential and no matter what, it won't exceed this. It would be good to know, because if you could do something like the above and work out straight away, what it's starting off QS is going to be, it will give you some idea as to whether to keep it, if a high ranking is something you are looking for. One thing that does help, is that in my weekly reports I get to see where ads in a particular adgroup rank. Say I have 2 ads and Ad A's average rank is 2.4, yet Ad B's average rank is 6.7, Ad A will more than likely receive a higher CTR and it might be listed higher, because Google prefers it and so rewards it with a better Quality Score. Split testing is great, but if Ad A is around position 2 and Ad B is in position 7, when you come to check in a weeks time you can't just go off CTR as CTR A vs B won't be a true figure, due to their rank differences. Fortunately, I treat Conversion cost and rate more than I do CTR, but both are very important factors.
Hmm... and ad groups were exactly the same, meaning other keywords if any, ads etc? In that case I'll have to agree with bjewelled - min. first page bid is just inaccurate for new keywords.