17 replies to this topic

[kestrel]

Hi,

Does Google only look at the on-page relevancy of the landing page when determining ad quality score or does off-page relevancy come into play?

Cheers,

K

[lee.n3o]

Have a quick look at this K

http://www.cre8asite...showtopic=48718

Might help, might not...

[kestrel]

Hi Lee,

Excellent tool for taking the work out of trying different landing pages but nothing on ad score and relevancy.

Google advises your ad quality score is based partly on the keyword relevancy of the landing to the ad and the search query.

What I would like to know is does Google just look at the on-page relevancy or is off-page relevancy a factor in this equation?

Cheers,

K

[Huligan]

Here are the factors Google uses to calculate quality score.

- Keyword relevance
- Ad copy relevance
- Landing page relevance
- Click through rate (CTR)
- Keyword performance history
- Other relevance terms Google will not share

To answer the original question, Google the relevance of your keyword and ad copy play a part. Also, your CTR is included. Here is the official answer from Google.

What is a 'Quality Score' and how is it calculated?

*link drop removed

Edited by eKstreme, 03 July 2007 - 04:26 PM.

[A.N.Onym]

It's be been rumored (do I not like to spread rumors, tho) that Google also takes into account the overall linking weight (popularity) of the domain. Of course, it also takes all factors you can think of into account, too.

It has been noted earlier that heavy early spending helped boost a campaign.

In fact, quite recently I had copied an ad to a new ad group and it was shown more than a new ad, though the settings were to show them equally.

Edited by A.N.Onym, 04 July 2007 - 03:29 AM.

[Huligan]

When creating your PPC campaigns, it is recommended that you start with high bids because it will help your long-term quality score. The reason is you get a higher ad ranking which will help to give you a higher CTR. The higher CTR will help give you a better quality score. I don't believe the linking popularity of the domain matters because it would create an uneven playing field. PPC advertisers would discover this unfair policy and would react negatively. Google doesn't want this.

[Black_Knight]

I don't believe the linking popularity of the domain matters because it would create an uneven playing field.

Google was the first to factor in popularity by users, a typical Googlism of allowing user choice to counter dvertiser bids in the form of click-thru weighting. Using existing user-preference data (such as from a weightd form of Link Popularity rating, such as PageRank) makes more sense than using TrustRank, and either could b a smart move, and very Google.

PPC advertisers would discover this unfair policy and would react negatively. Google doesn't want this.

That was why the other PPC companies allowed only bid amounts to set positions. Google won, and Yahoo! eventually aded a quality factor over their initial 'free market' highest bid wins system. Of course, it didn't hurt that Advertisers knew they could make up for low factors elsewhere by paying even more, and thus ending up bidding even higher to beat the weighting of the system.

[Huligan]

Black_Knight, I think my reply was misinterpreted. PageRank only applies to organice search results. It does not apply to pay per click results. Click Thru weighting is used but only with regard to the click thrus on your PPC ad. This goes into the quality score calculation. Any measure or score for relevance is contained within the PPC world only. Google does not consider your organic ranking when scoring your PPC ads. Google wants to deliver the best results but PPC relevance is does not stretch outside of the PPC boundaries.

[Black_Knight]

I don't think I misunderstood you at all. You said:

I don't believe the linking popularity of the domain matters because it would create an uneven playing field

Which is a clear satement of your beliefs.

I present a cogent counter-argument to your claim that it would create an uneven playing field, and that Google would not consider it.

Indeed, it is where pure cash determines the rankings that there is what we traditionally call the unfair playing fild, which is precisely why SEO, which is not solely dependent on who has the most millions to throw into TV and Print advertising, has been so often claimed to have created a far more level playing field for large and small companies.

Google's AdWords, with its rankings weighted by click-thrus, not Overture's pure bid order amount was the first PPC offering to have been cited by the press to have also put small businesses on an even playing field (though we all know it isn't really true).

The offline world of business has often seen the giants of industry deliberately initiate pice-wars that result in making a loss on every sale simply to starve out the compettors wh don't have millions of dollars of cash sitting around, and can't afford to suffer a loss on every sale.

We saw the same thing in the early days of the internet with the CD and books giants all taking a loss to provide free shipping, with Amazon losing millions of dollars per month at the time just to continue to be the cheapest.

Afterthought

I am not making the claim that Goole do, currently, use PageRank or TrustRank, as even a part of the overall 'quality scoring' of landing pages, but to simply dismiss that they could, whenever they wish, would be very unwise.

[Huligan]

My "uneven playing field" comment was in reference to Google using PageRank and other organic measures in their AdWords calculations. Google doesn't do this and you are not saying they do. You are saying what I restated in my previous reply which was Google uses click thrus on an advertiser's PPC ads when calculating quality score which affects bids and ad position. We're saying the same thing.

[Black_Knight]

Not precisely the same thing.

You say:

I don't believe the linking popularity of the domain matters because it would create an uneven playing field

and I firmly disagree with your basis/rationale for your disbelief.

More impotantly perhaps, you say:

PPC advertisers would discover this unfair policy and would react negatively. Google doesn't want this.

And I highlighted that Google took that risk before when no accepted wisdom of PPC provision would support the idea of risking upsetting advertisers by not letting the highest bid win. Google created a system where you could pay less the more pupular the CTR was, thus obviously costing Google a lot of revenue potential. Billions of dollars worth.

Google loves for the broad masses of the internet to weight the scales. Google does want this. Google's hareholders probably don't want this, and that i likely the only reason TrustRank factors may never be applied to PPC ranking.

[Huligan]

Ugh... OK, you keep referencing my first post after I've made follow up replies. Here are my statements and beliefs. Please ignore my previous posts if you choose to reply.

1. Currently, Google does not use PageRank in their quality score calculations for determing keyword minimum bids. PageRank is a measure of non-PPC elements related to an individual's website.

2. When talking about Google's future moves, I believe it's fruitless because it would be pure speculation.

3. I know Google's history in pay per click. I agree that using just bids in determining ad position is not good. This is part of the reason why I didn't use Yahoo/Overture for a long time. I'm definitely in favor of the way Google is using quality score in determing minimum bids and ad ranking. No surprises Yahoo's new system now includes a similar measure.

[projectphp]

Please ignore my previous posts if you choose to reply.

Probably a good lesson in that...

thus obviously costing Google a lot of revenue potential. Billions of dollars worth.

Is that true? $10 * .1% CTR = $10 CPM. $1 * 1.3 CTR = $13 CPM. For every 1,000 searches, Google makes $3 more.

I think someone at Google did this CPC to CPM (cost per thousand impressions) measure and said "Hang on, it is Cost Per Click. No click no revenue. We need to adjust for that to maximise profit per impression".

Effective CPM (CTR * bid) is the way to maximise profit. It also counters the benefit advertisers get from ad qualifiers (DO NOT CLICK THIS AD unless..." and things like putting phone numbers in the ad (no click, no ranking) counter productive.

[Black_Knight]

2. When talking about Google's future moves, I believe it's fruitless because it would be pure speculation.

That post was made in the past. If it has no speculative merit, it has no value in the present.

More than that though, all marketing *is* speculation.

Are you looking at keyword data that includes the 25% of all searches that have never, ever, been made before? Of course not. Early on, you base keyword selection on past trends and speculate on the likelihood of that trend continuing.

If you are really smart, or simply dealing with a new type of product, you create your own memes and buzz-words, because you have to, and so help create that 25% of searches that are using new terms of the moment.

And of course, with SEM you are usually operating in a time-lapse environment, where to measure the effects of a change you make now, may take weeks. Longer with the extended buying processes of high-ticket-prce items and items that have a log-term commitment.

Don't dismiss speculation, it is a large part (and art) of your career. How good you are at it will determine your success in any Markting based career.

Is that true? $10 * .1% CTR = $10 CPM. $1 * 1.3 CTR = $13 CPM. For every 1,000 searches, Google makes $3 more.

Indeed, assuming such a massive diference in CTR it can ake more money. But is that realistic? Were you ever running PPC campaigns on Ovrture just prior to the launch of AdWords? The price differences were often just pennies between the top 20 bidders, and indeed, a lot were bidding identical amounts and only the factor of 'who'd bid first' was determining the order.

We knew even back then that much of the frenzy to be #1 wasn't justified, and that position 3 (Overture showed the top 3 above most SERPs on clients' search engines, like Y!). The CTR difference was affected by order though, where the top-most would always attract extra clicks simply fo being the first, (but might often attract less conversions simply becuse it was the first, and thus no comparison data was available to potential buyers until they went to later results...)

Certainly, whil it was something I, as an internet user, loved Google for adding, it was something a lot of PPC clients did not appreciate. To them, this was like an auction, where your bid was subject to a secret multipier by te auctioneer based on how much he liked you.

Getting back on topic however, think about why Google introduced any scoring for Landing Pages anyway? Why would Gogle want to turn down some bidders and disable their bids? That certainly does cost money for Google at least in the short-term.

The reason you need to question for yourself about why Google did that, is because if the current factors don't achieve whatever you think they wanted t, it is almost certain that they'll introduce still tougher criteria in the future, and PageRank or TrustRank would be some of the leading contenders in my book in terms of how incredibly easy it could be to add them into the ranking system, and how effective they'd almost certainly be in providing better results to users, and reducing complaints about Google's 'commercial evils'.

[hamletb]

Directly from the horse's mouth. Bill Slawski just blogged about a new patent that gives us all the insight we need about the Quality Score algorithm.

The patent applications lists examples of 44 different factors that might be used in a quality score that doesn’t focus upon click through rates. These include such things as:
# How many times a user selects a given ad in a given session.
# A duration of time, from an ad result selection, until the user issues another search query. This may include time spent on other pages (reached via a search result click or ad click) subsequent to a given ad click.
# A ratio of the time, from a given ad result selection until the user issues another search query, as compared to all other times from ad result selections until the user issued another search query.
# Time spent, given an ad result selection, on viewing other results for the search query, but not on the given ad result.
# How many searches (i.e., a unique issued search query) that occur in a given session prior to a given search result or ad selection;
# How many searches that occur in a given session after a given search result or ad selection.
# Rather than searches, how many result page views that occur for a given search query before a given selection. This can be computed within the query (i.e., just for a unique query), or for the entire session;
# Rather than searches, how many search result page views that occur for a given search query after this selection. This can be computed within the query (i.e., just for the unique query), or for the entire session;

If you use paid advertisements through Google, these patent applications may be worth delving deeper into. It is pretty interesting to see all of the user behavior considerations that may go into determining a score and placement for an ad.

You can find out more here http://www.seobythesea.com/?p=717

[bragadocchio]

Thanks, Hamlet.

The patent applications list over 44 different examples of things that they might look, and I only listed a few of those.

There is a lot to those patent applications, and a lot of math involved. How are your math skills? I've been starting to brush up on mine.

No guarantee that what is described in the patent applications is what Google is doing these days, but the good thing about them are that they are from Google.

[hamletb]

There is a lot to those patent applications, and a lot of math involved. How are your math skills? I've been starting to brush up on mine.

Bill,

I am refreshing on linear algebra and graph theory. Last time I attended linear algebra classes was on 1993! Without the Math is really hard to make sense of these patents.

Dr. Garcia has some really nice Math and IR tutorials.

[kestrel]

It would kinda of make sense if off-page factors were included in ad score calculations. It would certainly make sense for the consumer.

My one concern would be for the advertiser who didn't factor natural SEO into their budget and suddenly finds themeselve plummeting because older sites more established in the SERPs have more trust/page rank.

Is that fair on the advertiser? Isn't the point of PPC (for the advertiser) that anyone can get started overnight and make a living without having to wait months/years for natural results?

It would certainly make it less a attractive option for the paid advertiser if they found they had to do off page SEO as well on-page SEO for their PPC campaigns.

(And what about all the redirecting tracking techniques used by PPCers? How will they be affected?)

Is that fair on the consumer? Is that enough reason one way or another for Google to factor in/not factor in page/trust rank?

Its easy to say - If they ain't doing it now they may well do in the future so might as well get on the case. It won't hurt and if they decide not to, your efforts won't have been wasted because you'll stand a chance in the natural SERPs.

But how do you explain to a client that they need to be increasing their budget to cover work for something that might not happen/is only speculation?

Yes we speculate with the natural SERPs all the time, but not so much if you've been practising 100% Google freindly SEO. We usually only speculate with things like paid links might getting discounted in the future and when they did, fair enough, we knew we were sailing close to the wind in the first place.

So the conversation would be something like "look, most of our links our paid. We need to build more natural ones because there's a chance the paid ones might get discounted."

But I've never heard anyone speculate that authorative links, orginal content, best practice HTML will get discounted for instance.

If Google decides to use page/trust rank in Adwords then there's no prior warning and not a lot the paid advertiser can do. There'll probably be a bunch of disgruntled PPCers wishing they'd stuck to natural SEO though

And what would happen to the terms SEO and SEM? Does SEM become part of SEO or the other way round?

K

Edited by kestrel, 08 August 2007 - 10:20 AM.