Last week, Google (as well as announcing SSL encryption for search) introduced perhaps the most significant changes to its PPC algorithm for some time. 

Google has changed the proportion of quality score weighting to make landing page relevancy more important, a moving away slightly from of click through rate (CTR) as the primary metric for an ad's success. 

I've been asking some PPC experts about how the changes will affect brands and advertisers... 

What is your view of the changes? 

Kevin Gibbons, Founder, SEOptimise

In theory, it makes sense. Google's algorithm, both for natural and paid listings, has always been about providing the users with the most relevant result possible. So if you're optimising for a long-tail query using SEO, you will naturally be looking to send users to a specific landing page which is focused around this term. 

That's good for the user, because they are being sent to a targeted landing page for the exact query they searched for and it cuts out the need for unnecessary additional browsing in order to find the information/product etc that they searched for originally. 

However, it's not a one-fits-all algorithm change in my opinion. While it makes sense in theory to have specifically targeted campaigns, sending users to your homepage isn't always due to laziness in advertising strategies, we've seen many cases where a homepage will convert at a much greater rate than a targeted landing page.

So while in theory it makes sense, in practice it may reduce the quality of the landing page from a user perspective and have a negative impact to revenues in some cases. 

Phil Pearce, Analytics Director at Jellyfish:

This change is in line with Google Adwords strategic plan to move beyond-the-click, as a means of increasing the industry average CostPerSearchQuery. 

CTR has always been (and still is) a necessary means of maximising revenue for Google, by filtering-out the good & bad advertisers; rewarding good advertisers who are generating a comparatively high volume of click, with a lower cost per click: Think Costco/Wal-Mart approach.

This tied with a Clicks-as-votes user centric marketing, has help Google skyrocket in terms of market share, brand recognition and revenue.

However, as the paid search market has matured, CPC prices have flat-lined, and increases in blended CTR are increasingly harder for Google to achieve. Clearly, paid search innovations such as Sitelinks, Ratings, Product Listing, +1 Social recommendations, all help create CTR uplift, but there is a fine line between revenue optimisation and a cluttered search page, which could lead to user leakage to Bing.

Additionally, Google is vulnerable to Comparison sites, Aggregators and Arbitrage sites who target the masses, rather than specific audience types, hence their adverts outperform brand advertisers especially on generic phrases - Think Generalist vs. Specialists.  

This creates a bizarre auction outcome, whereby pre-qualified ads (such as product prices or loan requirements) actually cause an increase in CPC for the advertiser, due to "low impressing voting" and thus a penalised account level quality score.

Lastly, Google Adwords is tied to the CPC model, as mainstream usage of CPA-only model for paid search could be seen as biased, non-consumer friendly or it could taint the Google brand.

The ~10% CTR shift from 60% to 50%, and redistribution towards ~25% relevancy and ~15% landing page (with ~10% account level QS unchanged) will lead to re-focusing of efforts from AdCopy optimisation, towards a more granular matching of Keyword-to-Adcopy-to-Landing Page.

Think user-intent matched to offer/answer fulfilled on landing page. It could lead to a more honest Advertising environment, with excessive adcopy claims matched with unrealistic landing pages, eventually downgraded in the ppc auction.

Matt Whelan, PPC Director at Guava:

This has been live in a number of countries for a while now and our colleagues there didn’t report much of an impact, so I suspect we won’t see too many changes. In general a move that is genuinely rewarding quality content has to be a good thing.

How will it affect brands and advertisers? 

Kevin Gibbons:

This is where it gets interesting, because in theory having a greater emphasis on landing page relevancy towards quality score does make sense – but it practice it's going to be a huge challenge! 

I expect this will be a lot more difficult to roll out. One of the biggest challenges is always getting things done and justifying the effort of resources against the potential reward of these changes.

So I wouldn't expect to see brands and advertisers rushing out to make major changes here until they can see the impact that this has had to their CPCs and CPAs. If these increase and make the campaign less profitable as a result, then it will clearly require more urgency.

Phil Pearce:

Pre-qualifying adcopy becomes more viable on generics and a closer attention to keywords with poor bounce rate, rather than just good CTR is increasingly important. Yes, this means the PPC manager needs to use GA a lot more! Rollup those sleeves, and start digging into that data!

A bulk review of all keyword penalties to fix the quick wins is a good plan.

Checking daily account level QS and first page bid over the last few months is a good barometer to see if there is a negative change.

Running a "QS what’s changed report" in Excel, before & after the update on Monday, should highlight any keywords that have been negatively affected, then updating the landing pages or keyword-adcopy combo for these terms should be adopted.

Enabling Google plus1 buttons on landing pages and help improve Ad destination to social recommendation.

Utilising landing page trustmarks, or even Google Trusted ecommerce stores beta in USA is a good plan.

Matt Whelan:

The change is designed to reward engaging landing pages, and a vast majority of advertisers are already doing this right.

If you don’t have an engaging landing page you end up paying for traffic that is only going to bounce away, so this was already pretty high on the agenda for most marketing departments.

It's far more likely to impact affiliate sites, many of whom are still driving traffic to thin sites with little content via PPC and then referring that traffic to the brands,  and this is without a doubt Google’s intent, to better facilitate the positioning of brands in good positions in the SERP at the expense of sites that add little value.

It’s important to remember that Google won’t roll out any change that negatively impacts its revenue from search, so in aggregate this won’t really mean cheaper clicks, it simply means that established brands, who are probably already doing landing pages right, might see a small increase in position and hence volume as some smaller advertisers get penalised.

Will the greater importance placed on landing pages provide a challenge for PPC agencies? 

Kevin Gibbons:

Yes, but again, more in the sense of pushing clients into implementing changes, especially for large accounts where the number of keywords being bid on are into the tens of thousands.

Plus they will need to weigh up where time is most effectively spent, at the moment it's unclear how significant the impact of this change will be, and it will affect each advertiser in different ways.

I'd expect to see PPC agencies managing the effort involved by taking an incremental approach towards prioritising landing page changes to top performing campaigns and adgroups, rolling this out gradually and reviewing the impact in terms of performance before pushing this out on a wider scale.

It will need testing heavily, with the most important metric still being cost per acquisition, as opposed than quality score or CPC, so you need to consider how an improvement in landing page relevancy impacts your CPAs and overall revenue generated from the campaign.

I'd also expect to see dynamic keyword insertion into landing pages becoming a more popular PPC tactic in order to boost quality scores based on passing each search query into the optimisation of landing pages via URL parameters. 

Phil Pearce:

CTR is still the strongest signal in the auction, so don't panic! The world is not ending, it's not an overnight shift. But ensuring that your GA account is connected to Adwords, landing pages are URLs recorded in GA are clean of url parameters, site speed is enabled and installed correctly, will all help.

Utilising the inbuilt reports for landing page analysis, setting up ppc dashboards within GA, and creating campaign alerts, and will all help the client maximise sales though a combination of CPC management, landing page optimisation, user experience testing and clever analytics.

Matt Whelan:

Only very rarely do agencies get full control over the page they are sending traffic to, so there is a risk here.

However, most agencies will always have been looking to help their clients improve landing pages, as this improves the value of the traffic they send and ultimately helps justify their fee. I don’t think there’s any greater risk now than there has ever been.

Graham Charlton

Published 24 October, 2011 by Graham Charlton

Graham Charlton is editor in chief at SaleCycle, and former editor at Econsultancy. Follow him on Twitter or connect via Linkedin.

2566 more posts from this author

You might be interested in

Comments (6)

Save or Cancel
Mike Fantis

Mike Fantis, Head of Paid Search at DAC Group/London (Formerly Make It Rain)

This change makes perfect sense. We balance the management of accounts across CTR vs. Conversion Rates, but Conversion is king!

High CTR doesn't necessarily mean a high conversion rate so it makes sense for Google to move away from this metric (albeit very slowly).

From an agency perspective it is relatively easy to drive CTR but pre qualifying users has to be the priority.

Landing page control can be an issue but this will support agency recommendations and MVT.

This shouldn't cause any change in account structure or strategy, you should still target relevance across keyword-ad-LP.

Cynics will associate high CTR with more revenue for Google and believe any movement away from CTR will be minimal. However, better conversion rates will also mean higher budget allocation.

almost 7 years ago


Simon Jarrett Williams - Carat Manchester

It's about time Google too Landing pages into the consideration of Quality. It's been one of those shady areas of search for years, "Landing Page Quality - Good" - the grey area of our PPC.

For too long now have agencies and advertisers been sending irrelevant traffic to Mobile devices with a poor consideration for the user.

Yes the CPC's may be cheaper and overall it may reduce the CPA - but Google have finally taken it upon themselves to actually recognise the clever client whom builds a site for purpose - it was only a matter of time as it fits perfectly within their model of "Serving the right advert at the correct time"

I cant wait to see the landscape change - albeit in line with our best practice at Carat, there are some exciting challenges ahead and of course - there is a need for better practice in Search, not explicitly just PPC.

almost 7 years ago


Josh Schneider

Finally, Google is smarting up! It has always been stressed to advertisers, by Google, to create more optimized landing pages, only know they are putting their foot down. There are so many large brands advertising and not having proper landing pages or very relevant content. It is extremely vital that your ad messages match your landing page, have little to no exit points and have a very clear call-to-action. This has always been a "best practice" in the industry of online advertising. Only now has Google decided to measure the Quality score on this. I am very happy and I hope that businesses begin to optimize their landing page strategies and designs. It will certainly make better conversions and better user experiences.

almost 7 years ago

Mike Fantis

Mike Fantis, Head of Paid Search at DAC Group/London (Formerly Make It Rain)

@Phil - Great tools.

I think the balance between CTR/QS vs. CPS/Conv Rate is a great subject, maybe econsultancy could organise a round table/debate ;)

almost 7 years ago

Phil Pearce

Phil Pearce, Senior Web Analyst at Accelerate Agency Ltd

For all the PPC people out there - Here is a Free copy of my interpretation of Adwords algo formula using Excel 2007:


Yes, an Econsultancy PPC round table (or even a new PPC Tuesday, monthly event) in London over early Dec or late January - would be a great idea! The last ppc event was 4months ago, and lots has changed over this time.

I agree, CPA`s and Conversion rates, are what drives business decisions. CTR and QS are metrics-by-proxy in the attainment of sales at a lower cost-per-sale.

Compulsive clicks in top positions, and the physics of moving the mouse from middle of the screen, to the right-hand-side, will mean that Ads in lower positions on left-hand-side, will generally have better conversion rates.

But the rules of the Adwords game mean that; even-if Conversion is King, Impressions voting is still Google Ace in the pack - at least in the medium-term.

Long-term, yes sacrificing CTR in the attainment of increasing advertiser conversion rates would in-theory, maintain and possibly increase the RevenuePerQuery for Google. Clearly, Bid = Conversion rate * (CPA * Margin).

But, this is a difficult trick to pull-off, as it requires a standardised CMS infrastructure, and reliable website metrics data. Few website have enabled microdata (used by keyword-less ads), and conversion data is frequently non-standard due to installation errors and clients website installation preferences.

TIP1: To detect Landing page QS penalties in bulk (without viewing every single bidword) - see this Adwords API function:

TIP2: As @KevGibbo mentioned dynamic {keyword} insertion IS triggered when Adbot crawl ppc landing pages.
Hence a destination url:{keyword}
is crawled as:
Thus, this technique will increase ppc landing page relevancy for dynamically generated content pages.
Although, it best to avoid getting these dynamic ppc pages indexed in Google organic.

Tip3: Utilising other available ValueTrack functions on landing pages, will further enhance content relevancy and performance. For example, visits from search, will behave differently from visits from content network ads. Also, content landing pages can utilise ss_placement_category=/entertainment/celebrities
Here is another example:{keyword}&s_match={matchtype}{ifcontent:c}&s_network={ifsearch:search}{ifcontent:content}&s_placement_category={target}

Also, monitoring for increased activity from the 2 AdwordsBots in server logs, will indicate that ppc landing pages have been through the Adwords automated scanning
* AdsBot-Google+(
* AdsBot-Google-Mobile+( +iPhone+Safari
Both Adbots just load the raw HTML and do not drop cookies, hence they ignore GWSO A/B tests. They send a blank referrer, with no gclid on the landing page.

For testing purposes, if you truly want to see the effect of landing page QS on CPC`s - you can actually block the AdBot via robots.txt
User-agent: AdsBot-Google
Disallow: /ppc-landing-pages/test.htm # QS experiment to Block AdBot from ppc dynamic pages
User-agent: *
Disallow: /ppc-landing-pages/ # Block organic GoogleBot from dynamic SiteSearch pages

Obviously, this would trigger the PageSpeed penalty, relevancy penalty, and would set the QS weighting to 0% out-of ~15% for Landing page QS component. However, in the "interests of science" a landing page control (allowed) and landing page experiment (blocked) would confirm the impact of this change on a like-for-like page basis.



P.S Here are some useful links, related to the above:

* Official Adwords documentation on the PPC algorithm, says:
Relevancy is determined by:
1. Relevance of bidword to the Ads in its AdGroup
2. Relevance of bidword and the matched Ad to the raw SearchQuery

* PPC landing page should:
1. Feature relevant and original content
2. Be transparent
3. Be easy to navigate
Ideally: have content that matches the SearchQuery

almost 7 years ago


Daniel FIgueroa

Very interesting... I am sure these changes are for the best as i can see it will improove user's experience.

Something i would like to know is, is the new algorithm already applied worldwide?

Here in Mexico i have not noticed any important changes yet, so im guessing not.

almost 7 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.