2017 predictions - Greenlight Digital

The greenlight

magazine

D E C E M B E R 2016 The Predictions Edition

2017 PREDICTIONS

1

MACHINE LEARNING, AMP AND UX: SEO PREDICTIONS FOR 2017

www.greenlightdigital.com

www.greenlightdigital.com |  +44 (0)20 7253 7000

SEO

BEA PATMAN HEAD OF SEO

Forecasting is a risky business at the best of times, but

This time around I’ve decided to hedge my bets and

there’s something about 2016 in particular that seems to

present a round-up of predictions from Greenlight’s SEO

have defied just about every expectation. For me, at least,

Consultancy team, instead of just my own speculation.

the unpredictable nature of this year hasn’t been entirely

Garnering their thoughts has turned out to be a brilliant

unwelcome, which has helped soften the disappointment

idea, because their proximity to day-to-day optimisation

of my last prediction for this magazine failing to amount

means their senses are far more keenly attuned than

to much1. That’s because this time last year I predicted a

mine to shifts in trends and behaviours. Seven key themes

big rise in native advertising, stimulated by advertisers

emerged from my poll of the team, with some more widely

attempting to circumvent the increase in consumers using

supported than others, so here I’m going to focus on the

ad blockers to suppress more traditional digital ad formats.

two most popular suggestions.

And while it hasn’t necessarily proved untrue – well, it’s hardly been the year’s marketing phenomenon either!

2


SEO SEO PREDICTION: AMP FEATURING:

DAMIAN BURKE SENIOR SEO CONSULTANT

DAN LEE SENIOR SEO CONSULTANT

This is the topic that was most unanimously supported among the team – not least because our eagle-eyed consultants clocked that six out of the last seven articles published on Google’s Webmaster Central Blog had been AMP-focused at the time I polled their thoughts. AMP, or Accelerated Mobile Pages, is an initiative designed to improve the experience of content consumption on mobile devices by stripping out weighty assets and delivering just the bare minimum of lean, efficient code. AMP itself is an open-source framework that utilises existing technologies in new ways, and we’ve already found ourselves employing if for clients in 2016. Nevertheless, the team believes that there’s more to come, with Senior SEO Consultant

3

mobile index and their trajectory towards Progressive Web Apps, I reckon we could start to see something similar on desktop as it’s fast, efficient and keeps the content somewhat within Google’s control. That said, I don’t see this working for everyone, especially on desktop, so it would also be interesting to see how Google frames it in order to convince sites to adopt.” Fellow Senior SEO Consultant, Dan Lee, agrees that we’ll see uptake of AMP continue to surge next year. Dan believes that genuinely optimising for mobile performance with dedicated solutions such as this will become

Damian Burke noting:

more important in the eyes of search engines.

“I think AMP will gather a lot of momentum in the next year. A lot of big sites are rolling out AMP, including Reddit2 and eBay, and I look forward to seeing where this goes in 2017. As consumers, we could be spending less time on traditional sites and viewing more third-party content without leaving Google’s ecosystem. We’ve already seen this on mobile results for things like movies, where you can see cast information, reviews, and so on. With Google’s impending switch from desktop index to the upcoming

one, however. Google itself has been candid


Damian’s point about adoption is an interesting about AMP not being a perfect fit for all web pages, given that dynamic content such as interactive maps, for example, simply can’t be served effectively within the code restrictions of the framework. In my experience, though, Google tends to work towards solutions to such obstacles at about the same pace that webmasters take up new technologies (the nowdefunct AJAX crawling scheme being one fairly recent example), so I can’t see there being too many remaining barriers to adoption by the end of the year.

SEO

SEO PREDICTION: MACHINE LEARNING FEATURING:

DAN CODY, SENIOR SEO EXECUTIVE Remember all the buzz about RankBrain back in 2015? It was October, and Google announced that for much of the year “a large fraction”

user’s experiences on a website, and the history of experiences on a website, in a meaningful way.”

of queries received by the search engine had been handled by a machine learning algorithm.

What I particularly like about this prediction

RankBrain uses “deep learning” to help it

is that it’s angled towards user experience,

guess the intent of queries based on a wealth

which is making up a bigger chunk of our SEO

of historic data about ambiguous searches

approach than ever as we reach the end of

and what they really meant, meaning that it

the year. It’s interesting, too, because we’ve

can understand queries that Google has never

talked for some time now about how Google

seen before and respond to them using AI-like

is gradually reducing its reliance on links as a

behaviour. The use of AI sounds like a pretty

quality signal and increasing the value placed

huge deal, but it’s important to remember that

on factors such as site speed, user interaction

RankBrain is just one component of Google’s

and heuristic usability evaluation measured

overall (Hummingbird) algorithm – one of about

directly by the algorithm. A heuristic evaluation

200, according to Google’s Gary Illyes. So, AI is

is one where a site’s UI is judged against a

certainly a factor in search these days, but it still

list of recognised usability principles or rules,

isn’t running the ship.

which are exactly the sort of thing that can

The team believes that this will change over the

be algorithmically defined as long as a page

course of 2017, however, Senior SEO Executive

can be rendered by a search engine exactly as

Dan Cody suggests:

users tend to see it. Where AI might fit into this measurement suite is in learning from other

“I think the use of artificial intelligence by Google is going to grow, taking the focus away from traditional keyword optimisation towards the searcher’s intent and (somehow) starting to incorporate that

4


user signals to inform ‘blind’ heuristic evaluation when a page can’t be fully rendered by the search engine. This is highly speculative, of course – but such is the nature of predictions!

SEO

SEO PREDICTION: EVERYTHING ELSE In the spirit of fairness (and optimum hedging of bets!), here are the other trends the Consultancy team anticipate in 2017:

STEFANIA ROMANELLI, SEO CONSULTANT

ELODIE MORVAN, SEO CONSULTANT

VISUALLY SIMILAR RESULTS FOR MOBILE

‘DIRECT’ TRAFFIC INCREASINGLY

AND DESKTOP

CANNIBALISING ‘ORGANIC’

Search results on desktop and mobile will look

Direct traffic will continue to ‘steal’ organic

more and more similar, even though different

traffic, especially on mobile, because of the way

indexes are in the pipeline.

in which mobile browsers are finishing some queries and sending users directly to websites.

SEMANTIC SEARCH Google Tools signals will move further away from keywords and closer to semantic SEO (such as topics and semantic variations). LOCAL SEO Hyperlocal SEO strategies will become more

AGATA BOWDEN, SEO EXECUTIVE

commonplace as Google compresses local map pack results.

VOICE SEARCH More than ever, we’ll be hearing much more on voice search next year – not least because “digital assistants” (the likes of Siri and Cortana) are now popping up in smart tech as well as on mobile devices.

1. 2.

5


http://www.greenlightdigital.com/blog/magazine/greenlight-magazine-predictions-2016/ https://www.reddit.com/r/changelog/comments/53q42z/ read_reddit_faster_via_google_with_amp/

www.greenlightdigital.com The Varnish Works, 3 Bravingtons Walk, King’s Cross, London, N1 9AJ +44 (0)20 7253 7000 [email protected]

6
