Is Click Through Rate (CTR) A Ranking Factor?

August 27th, 2015 by Jamie Richards


Moise Pointer Macro

Source: searchmetrics

Earlier this year searchmetrics released its 2015 Ranking Factor Whitepaper based on the extensive data it collects of organic search (the paper is well worth the read and can be found here).

A small section of the whitepaper focussed on click through rates with an attractive graph (above) on how CTRs of search results compare with organic ranking positions.

Searchmetrics data shows that on average the higher a result ranks in Google search, the higher the CTR of this result.

This is logical – users click on the top results more than the bottom results for a number of reasons, most often because many searchers assume the top results are always the most relevant for their query.

But what if searchers click on the bottom results more than top results and these results receive the highest CTR? Will results with the highest CTR stay at the bottom of the page (all things being equal) or will these results rise to the top?

The Argument For Yes

In 2014 Rand Fishkin of Moz conducted a quick experiment by asking his Twitter followers to search ‘IMEC lab’ in Google and click on Moz’s IMEC Lab page currently ranked 7th (we should point out that Rand currently has 264k Twitter followers so clicks likely weren’t in short supply).

moz 1

Source: Search Engine Land

At 6pm Rand’s page was at #7. 3 hours later, and Rand estimates around 175-250 clicks later, the page was #1.

moz 2

Source: Search Engine Land

But before we jump to conclusions let’s go back to some experiment basics. This experiment:

– Had no repeatable methodology

– Was not repeated

– Was not monitored for a period of time before or after the experiment

– Other important on page factors that could influence the results were not monitored (bounce rate + time on site)

– Click volume and CTR of other search results is not known

Which all raise issues with the credibility of the results. Nonetheless, ignoring all of the variables in this instance we can confidently say that as more people clicked on a result than the results above in a short period of time, the lower results moved up in a short period of time.

What Does Google Say?

google 1

Source: Twitter

Danny Sullivan, co-founder of Search Engine Land, shared a comment from Google’s former chief of search quality Udi Manber as saying, on very loose terms, that CTR does have a direct correlation to search rankings.

“The ranking itself is affected by the click data. If we discover that, for a particular query, hypothetically, 80 percent of people click on Result No.2 and only 10 percent click on Result No.1, after a while we figure out, well, probably Result 2 is the one people want. So we’ll switch it.”

The evidence seems to heavily suggest that CTR influences search ranking positions.

Or does it?

 

The Argument For No

Search Engine Land contributor Bartosz Góralewicz ran an interesting test earlier this month in which he sent bots with unique IP addresses to do the following:

– Open browser.

– Go to Google.com.

– Enter the query and click on “Google Search”.

– Search for (his) domain in the search results page. If not found, go to the second page, and so on, up to the 12th page.

– Click on the result.

– Stay on the website for ~2–4 minutes while going through random pages.

As Bartosz notes:

“After the experiment, my search rankings remained flat for two months, only to start plummeting around June 1, 2015.”

experiment 1

Source: Search Engine Land

Again however just as Rand’s test did, this experiment fell short on a number of adequate controls so the validity is somewhat subjective. It would also be a jump to correlate the click bot test to the later drop in rankings.

What Did (Someone Else At) Google Say?

But here’s an interesting point – Gary Illyes from Google stated at SMX Advanced that CTR data is not used for ranking, only experimentation and evaluation.

He noted that it is well known that bots can be setup and run to click on search results to boost SEO rankings and therefore Google ignores it.

And yes, that does contradict a previous statement from a Google employee!

Conclusion

Clearly the tests conducted so far, and mixed comments from Google, seem to contradict any conclusion one way or the other. So my opinion is to take it from a search relevance perspective – the purpose of Google’s search engine:

If people are clicking on the bottom result far more than the top result, wouldn’t it make sense for an algorithm built to deliver the search result that is best for the user to take this into consideration? I would say yes.

 

Under the influence of OOH

August 20th, 2015 by Emily Holgate


20.08.15 in-copy

Outdoor industry body, Route, has recently undertaken a study into the time spent consuming media channels vs. the media channel’s share of the astronomical £16.2bn UK advertising spend-pie. The study was inspired by research undertaken by Wall Street analyst Mary Meeker, who infamously draws parallels between US ad spend vs. the average American’s daily media consumption, dividing opinion across the media industry.

The UK media study was conducted by the Outdoor measurement body, so it would be fair to assume a level of bias towards their medium – however, contrary to any pre-determined scepticism, all data used to compile the study has been taken directly from each media channel’s industry body (BARB, RAJAR, NRS, etc.), creating a level playing field and providing weight to Route’s findings.

Which media channel commands the most eyeball-time? You guessed it! Out-of-home (OOH) steals share with over a quarter of media time spent consuming this medium (28%). And it’s unsurprising really, as studies show the average Brit spends 3 hours and 10 minutes per day out of home and in a public place.

Commutes are longer than ever, with the average daily commute weighing in at 1 hour and 14 minutes, as young professionals prioritise higher paid jobs over a swift commute, with the ‘upmarket and ambitious’ travelling up to 30 minutes extra per day for a 20% higher salary. This increase in time spent “out and about” results in an average daily consumption of 71 out-of-home adverts per day, with the average for men exceeding this at 82 ads per day.

What is surprising is how this tallies up against ad-spend, with the media channel achieving 28% share of media time, receiving only 10% of advertising spend share.

So, how does this stack up against other channels?

  • TV receives an enormous 45% of ad-spend, whilst achieving just over a fifth of media time share
  • National press ad-spend outweighs time spent reading the papers, with 7% of the consumers time spent getting printy fingers, whilst an inflated 10% of advertising spend is being placed within this medium
  • Radio is another undervalued channel, with the % ad-spend half that of the media consumption share (10% media time share vs. 5% ad-spend share)
  • Magazines and regional press both display fairly equal levels of time share vs. ad-spend share
  • Internet is a bit of an anomaly – the study shows a fairly equal ratio between time and ad-spend, however the ad-spend figures exclude paid search, which equated to £3.7bn in 2014

Of course there are various factors that influence channel ad-spend…

A factor restricting the OOH markets ad-spend is the shortage of inventory available, with a finite volume of panels, which are increasingly being sold-out months in advance of in-charge.

Additionally, OOH is often construed as one of the least measurable channels, as post campaign delivery figures aren’t available as they are in TV and online. To the contrary – Route provides a coverage and frequency measurement pre-campaign, which is based on the GPS tracking of 30,000 people across the UK, collecting billions of data points, which are then multiplied using a super-complex algorithm to determine how many impacts each and every panel across the UK will deliver. This may seem far-fetched, however if we consider the measurement of TV, the BARB survey is based on the recalled TV consumption of 5,000 people, and multiplied to be representative of the UK population – and no one ever questions X Factor viewing figures!

Finally, perhaps the most crucial point is how much importance we should place on time spent consuming a channel, when each channel has unique attributes that engage the audience in different ways. Different mediums command varying levels of attention, recall, trust, cut-through and enjoyment, and possibly most importantly, context.

20.08.15

Why did Google change its name to Alphabet?

August 12th, 2015 by Elliot Jones


Alphabet spaghetti

On Monday 10th August 2015, Google did something seemingly odd. They announced that they are changing their name to Alphabet.

The Alphabet business will cover the other, slightly less well known or orthodox endeavours – driverless cars, Google X, Nest etc.

It will also include Google itself – all of which will be headed up by Eric Schmidt as Executive Chairman.

So why are they doing this? Boredom? Not likely – in fact the biggest and simplest reason will be clarity. The reality will be that investors will be able to see, on one side, the huge revenue generated by the core business, and the not insubstantial spending and investment on the other, quirkier oddities and ideas the internet giant comes up with.

Of course it also means Larry Page can get truly stuck into the stuff he clearly loves – dreaming up new ideas. But he’s cleverly managed to do this whilst also being the big cheese (he’s now CEO of Alphabet) and can follow his dreams whilst still ensuring that he’s up there with Eric Schmidt in terms of the business.

In reality, he’ll be evaluating and allocating funds to projects – which is what Google already does, actually – just with far greater transparency, keeping the financial world much happier than when Google hid so much of its spending habits from the eyes of investors. So, perhaps Larry, at least, has the best of both worlds.

The truth then is that it’s not the name that matters at all. The reason the search engine’s stock price rose after the announcement was nothing to do with the name, but rather the implications it would have on the way the business is run. So the 5% increase in share prices, though not bad for a name change announcement, is a product of slightly more detailed financial nuances.

To you and me though, Google’s still an advertising company. You’ll still ‘Google’ a company to find their address. And unless you work in the City or Wall Street, the name change is unlikely to really mean much at all – until, that is, it follows the Skynet conglomerate model and really does take over the world.

Remember, ‘Don’t be evil…’

How To Create Great Content That Actually Works: A Case Study

August 7th, 2015 by Sarah Gough


©iStock.com/weareboutique

The web is saturated with articles that claim to teach content marketing success. Agencies, marketers, business professionals and bloggers alike have all thrown in their two cents’ worth, offering their own top tips for generating engaging, effective content. What I’ve found, however, is that none of this advice is accompanied by demonstrable success. How can we know that these techniques work without actually seeing the proof?

Content guidance generally focuses around the same five or six ‘usuals’, and while write for your audience and include a great headline are both noteworthy pieces of advice, it’s nothing that we don’t know already. What we want to see is an actual, tangible example of bloody good content that has worked its socks off to return some results.

So that’s what we’ve done: a guide to creating great content that actually works, exemplified by a recent project for a client.

The clientThe client

As one of the UK’s largest online retailers of home furnishings, our client provided plenty of scope. However, with focus turning to one of their sites in particular (they have a specialist site for each product), we found ourselves scratching our heads over the content potential of roller blinds.

For this type of product, though not particularly sexy, there’s quite a lot to be said. Unfortunately, it’s all been said before. And while your content might be super relevant, highly informative, beautifully written and complete with aspirational imagery, it’s unlikely to take off because, well, it’s repetitive. We know this because we experienced it ourselves.

We tried educational articles about the best kind of blind for each room in the home; we capitalised on seasonality and trends with Halloween-inspired content; we pulled together a detailed story of the history of blinds, complete with reliable sources and thought-provoking images. We tried articles, listicles, case studies and infographics. We furiously promoted via social, connected with industry experts and charmed authoritative bloggers. Though we got some traction, we didn’t see the huge results that we were aiming for. We needed a new approach.

SEO informed by PPCPPC informing SEO

While the SEO team were mulling things over, it’s fair to say that the PPC guys were smashing it. It seemed ridiculous to not take advantage of their successes in the client’s paid activity. After all, if the conversion is high on paid traffic this will usually transfer over to organic.

So the PPC team pulled the latest campaign data and one thing became clear – blackout roller blinds was our target. It was the highest converting keyword across all campaigns alongside being a high margin product for the client. We decided to work up an idea focusing on blackout blinds, but we needed to bear in mind the results of previous content – ‘Top 10 Benefits Of Blackout Blinds’ and ‘Our Favourite Blackout Blind Designs’ may be relevant and hit all the keywords, but they’re too predictable, too dull, and definitely not shareable.

Thinking sidewaysThinking sideways

While our long term aim was to increase our ranking for blackout blinds, we knew the best way to build a strong foundation was to steer clear of generic, salesy campaigns and instead generate genuinely engaging content that people would read, share and remember and, hopefully, would gain us some valuable backlinks. As such, we started to think around the subject, rather than focusing purely on the product.

Who’s actually buying blackout blinds? How old are they? Do they have children? What are their interests? Why do they need a new blackout blind? Obviously, those in the market for blackout blinds want to darken a room, either for themselves or their kids. They’ll be looking for a way to prevent being woken by the sun, promoting a better night’s sleep. Perhaps they’d benefit from other advice about sleeping well?

We decided to go forward with a campaign about getting a good night’s sleep. Within this, we’d include the benefits of blackout blinds, but the content wouldn’t focus purely on this. Instead, it would inform people of everything and anything about sleeping soundly.

The making ofThe making of

The content developed into a detailed infographic with several sections, including ‘Your Recommended Hours Of Sleep’, ‘The Causes Of Too Little Sleep’ and ‘The Benefits Of A Good Night’s Sleep’. The design demonstrated technical and scientific truths through bright colours and childish graphics: an engaging balance of fact and fiction.

When complete we hosted the content onsite, ensuring all SEO best practices were covered off, including all of the usual on page essentials aka optimised title, H1, copy, and an appropriate meta description. Distribution then began, including social promotion, contacting relevant sites that may wish to share, and submitting to visual content distributors (the latter often function on a submit/accept basis, and generally have a backlog of around two weeks while the host approves). We also utilised Visual.ly, which allows users to set up their own profile – kind of like Facebook, but for data visualization and infographics. While we knew that all outbound links from Visual.ly were “nofollow”, it was still a worthwhile platform as there was exposure potential. A lot of it, we soon discovered.

SuccessSuccess

A week after our release we tracked results to report back to the client. While content is a slow burner and it can take weeks– if not months – to see real results, we wanted to keep the client informed of any early signals. We discovered the site had received some new backlinks, including a link from DesignTAXI, one third of a global creative network, and named as Forbes’ top five design and creative sites. It was a big deal.

Little did we know at the time, but TAXI’s content curators trawl the web to discover the freshest creative to share onsite and via social. One of their go-to sources is Visual.ly, where they stumbled across our content and decided to share it. Our client was now featured on one of the world’s top design sites, and exposed to upwards of 360k Facebook likes and 420k Twitter followers. We had asserted them as experts that offer more than reams of roller blind product pages. They were now the proud owner of great content that others were actively sharing, and were demonstrating to search engines that they were deserving of high quality features and links. Best of all, this link wasn’t chased or paid: it was earned.

LearningsLearnings

From this, we’ve been able to compile our very own list of top tips for content marketing… and best of all, there’s not a “create a strong headline” in sight!

1. The knowledge of others is a benefit to you, and you should be open to stepping out of your comfort zone and into others’ areas of expertise. For this campaign, PPC data was extremely advantageous for our content. The most successful strategy is an integrated one, so utilise the paid search experts, the tech geeks, the PR pundits, the traditional media gurus, the creatives, and anyone else that happens to be hanging around – it really works.

2. Focusing on one piece of great content is far more beneficial than churning out stuff ‘just because’. This campaign required data analysis, brainstorming, research, careful planning, copywriting, design, proofing and editing, not to mention the distribution, promotion and then monitoring and reporting. Any good content marketer will complete these tasks exceptionally well for one piece of work, rather than half-heartedly for several.

3. An obvious one, but don’t be afraid to promote your content. Utilise different methods to share it, whether that be dropping an industry expert a tweet, submitting to a relevant blog or simply picking up the phone. If your content is well researched, well executed and relevant, you have absolutely no reason not to share it.

4. Don’t overlook the potential of channels with “nofollow” tags. Just because search engines ignore the link, doesn’t mean everyone else will. In our case, submitting to Visual.ly put us in an excellent position to be spotted by DesignTAXI, thus gaining a really valuable link from an authoritative domain.

You can find out more about Boutique’s SEO and content service by tweeting us @WeAreBoutiqueUK or giving us a call!