2018 will be Google's Year of the Machine

Posted by Owen Powis on 30 Jan, 2018
View comments Search News
2018 might be the Chinese Year of the Dog, but it’s going to be Google's Year of the Machine. In this year's piece I go through my predictions for 2017 and my new predicted Google focus for 2018. Want to see how I did for 2017? Read on..!

Google Assistant

Last year I wrote a piece focused on mobile and correctly predicted Google would continue to move away from big algorithm updates, and have a continued focus on the mobile index. So let’s start off with a quick recap of what I said last year.

The main thrust of the article was about mobile and how Google was going to be focusing on this to increase revenue from Adwords...

Google has access to brand new markets and a shortcut into markets they were previously struggling with. Desktop would never provide access to these as there are way too many barriers to ownership rates rising so dramatically (such as cost and internet infrastructure).

Mobile has solved a key problem for Google. 99% of revenue for Alphabet comes from Google and 77% of that comes from AdWords. That mobile traffic is key to this figure and they are going to be doing their best to keep pushing it.

And this is from an FT article on the third quarter earnings for Alphabet last year:

Strong growth in mobile search, programmatic advertising and smartphone use in Asia helped accelerate revenue at Alphabet to the fastest rate in almost five years, surpassing estimates for sales and earnings in the third quarter.


I did focus more on Africa than Asia, which was misplaced with the amount of investment Google has placed in the Asian market and the potential gains there. However, Africa is still an incredibly important market for Google and one they invested in massively during 2017, pledging to train 10 million people in Africa in online skills.

Additionally, Google has launched and continued to progress the mobile index, with mobile SEO further splitting from desktop.

One of the other noteworthy predictions was that we weren’t going to see any major updates on the desktop index anymore. With updates instead being small and frequent unnamed (by Google) corrections.

In 2017 we saw lots of small updates detected by the community and whilst there were a few new penalties put in place, for example targeting pages that used interstitials too heavily, these were not big algorithm shifts. The days of penguins, pandas and hummingbirds are over. It makes the lines even more blurred as it’s harder to point to things and say ‘after update X we know Y is now the case’. As a result of this, I would expect decreasing amounts of consensus across the SEO world as this continues. See our Google alogirithm updates in 2017 for a recap of the year's updates.

Finally, for the recap, I also talked about ‘Peak Mobile’ in some markets with Google's focus shifting to changing user behavior for those who already owned mobile devices…

In the UK and US smartphones have reached saturation or are at least very near that point. With over 70% penetration in pretty much all the key markets (Europe, US, China), growth in mobile is going to be relying more on changing user behavior of existing device owners. Therefore we can expect more focus from the search engines on user behavior.

Google went on to broker a deal with Apple and they switched from Bing to Googe to power Siri results on iOS in September 2017.

I’ll come back to this in the next part of the article where I lay out my predictions for 2018 as I think it’s strongly relevant.

The 2017 article is worth a read and covers a few other points as well. All the information I put across was an extension of the activity that was already taking place, so it was all a pretty safe bet. This year however I am going to go decidedly off piste…

The new device

There is a battle of the machines going on between Google and Amazon, with both of them vying to get themselves in your home. Google has ‘Home’ and Amazon has ‘Alexa’. Both of these are physically little more than glorified bluetooth speakers. However that speaker is linked to their respective AI offering.

Think about the last time there was an entirely new device for you to be served content from. We had desktop, then laptop and then mobile. An honorable mention goes out to tablet as well but it’s pretty much lumped in with mobile. But that’s it. Since 1998: 3. Well that’s now 4 with Google fighting a pitched battle to get into your home in yet another format. It’s Google VS Amazon or in other words Home VS Alexa.

Google is putting increasing resources into this battle, having realized that Amazon was leapfrogging ahead with Alexa. You might think this device is insignificant for Google as it has really limited potential as an ad serving platform. Which is correct. But, it’s got massive potential as a data collection platform, and crucially with every search or query made through Alexa, which is not visible to Google. So it’s losing out on that data.

Google Home is going to become increasingly integrated with other devices and they are going to keep driving the device cost down offering different model types. I also think that Google Assistant will be released, for free, for any manufacturer to use in their 3rd party device. There were a couple of examples of this popping up with select manufacturers right at the end of last year. Google doesn’t care about the hardware, it’s about getting Google Assistant into as many homes as possible. I can see it becoming a standard integration into bluetooth speakers, especially as Google have a track record for developing and releasing products free for any manufacturer to use, Android being a great example. That’s how they came to dominate mobile search and it’s how they’ll dominate voice search as well. Google make a massive $0 from Android the software, but they make billions in revenue from the searches conducted on Android devices. They also, importantly, don’t have to pay handset manufacturers for Google to be the preset search on those devices.

Looking at the recent increase in the range of Alexa products, it does look like Amazon is throwing the kitchen sink at it hardware wise and this could be in a pre-emptive strike against Google releasing Assitant out to any 3rd party to use for free. Interestingly within the EU the right to data portability will mean that consumers will be allowed to port their personal data between devices. This might sound like a win for the consumer, but it could have a dark side. It's forcing companies to put aside their differences and develop a universal data format for our personal data. Theoretically this means it will be MUCH more useful to third parties and easier to compile even bigger data pools. Also if you swap back and forth between devices you're sharing that data in more places, adding more information to their own networks.

The machines are coming from your home

I wrote an article recently about Natural Language Processing, which we now use within Wordtracker, which touched on some of the current limitations of these offerings. They are a long way from perfect and if you know what you are looking for, specifically the things these types of AI are typically bad at, it’s very easy to trip them up with simple questions.

One of those question types is comparisons. You can ask ‘how far away is the moon’ or ‘how far away is the sun’ and get an answer without a problem. However ask ‘which is furthest away, the moon or the sun’ and although the information is there it’s too complex for Alexa to process. It takes a leap in understanding of both the question, processing of information and expected result to be able to respond, and this is simply out of current reach.

It’s this ability to process the information that’s already present for a more useful result which I think Google is going to try its best to leverage. Last year I touched on ‘micromoments’ which are the moments when a consumer pulls out their phone and checks information mid-decision. That point where you are walking down the street and want to know where the nearest restaurant is. These moments are incredibly valuable as they are hugely actionable - you are ready to make a decision right there and then and your next action will likely be to commit or purchase.

“Mobile has forever changed the way we live, and it’s forever changed what we expect of brands. It’s fractured the consumer journey into hundreds of real-time, intent-driven micro-moments. Each one is a critical opportunity for brands to shape our decisions and preferences.”

Source: https://www.thinkwithgoogle.com/collections/micromoments-guide.htm

This is just one touchpoint where Google hopes to be able to utilize AI to better deliver on. This ‘in the moment’ advertising is the context. However, there are also optimizations of the ad served to make sure that it’s the right ad as well as the right time. For example, based on past behavior they might look at what colour ads, placement or type you have responded to best previously. They then roll all that information into one to serve you an ad at that point that is contextual (meaning served in the correct context such as a restaurant ad as you are walking down the street looking at restaurants) and personalised (as in tailored to you specifically based on past ads you have responded well to).

Use of AI allows for taking huge amounts of data from multiple different behaviors, touchpoints and importantly, patterns and roll this information into an ad.

A big part of the opportunity for marketers is how AI will help us fully realize personalization—and relevance—at scale. With platforms like Search and YouTube reaching billions of people every day, digital ad platforms finally can achieve communication at scale. This scale, combined with customization possible through AI, means we’ll soon be able to tailor campaigns to consumer intent in the moment. It will be like having a million planners in your pocket.

We’re getting closer to a point where campaigns and customer interactions can be made more relevant end-to-end—from planning to creative messaging to media targeting to the retail experience. We will be able to take into account all the signals we have at the customer level, so we can consider not only things like a consumer's color and tone preferences, but also purchase history and contextual relevance. And all of this will be optimized on the fly in real time.


This is a mission statement from the Google VP of marketing, Marvin Chow, in September 2017 and I strongly believe it’s where they are going to be focusing a huge amount of effort and resources this year. Adwords is where Alphabet gets almost 80% of its revenue after all.

Taking AI a step further

Something not mentioned in the 2017 article - and where I am probably heading pretty deep off-piste - is predictive behaviour based upon modeling from using AI to process and understand multiple user data. In other words, it’s not just looking at your past behavior but the past behavior of people like you.

For instance there are correlations between people and the products and services they like. So Google will be increasingly modeling you and trying to ascertain what you may be interested in from what type of person they think you are. On a simplistic level this is showing you ads for pet insurance after you googled ‘dog food’. People who purchase dog food also purchase pet insurance. There is a clear correlation. However things get a bit more weird when you take into account the multiple touch points. For instance you visit the vets and start getting ads for a specific pet insurance policy.

You’re already targeted based on past behavior, and that can then be combined with the ads that the people who meet other key criteria for you such as your age, gender, income etc have responded well to. Now you’re being pushed a single highly targeted product - advertising that is both contextual and personalized.

By the way, if you don’t want Google to know where you are all the time you need to go into your timeline within Google maps and turn off the tracking services. This is for both Android and iOS devices. This is the tracking on my device:

However I still continue to get notifications when I arrive or leave some places (particularly my local supermarket) asking me to ‘rate my experience’. This is without using Google Maps (I know the way to the supermarket!) so Google appears to still be tracking me through my device despite having it clearly set up not to. Which is comforting.

Following on from this, understanding people and patterns through AI will be Google's biggest driver in 2018; it follows on perfectly from the mobile adoption focus. They need mobile adoption for tracking and will be trying to use this to their advantage.

So what do you think Google’s focus in 2018 will be?

If you think I’m right, wrong or anywhere in between please do have your say in the comments….

Do you think Google is tracking us too much?

Will AI be taking over advertising?

Does it even matter?

Recent articles

Google launches new personalisation options in Search
Posted by Edith MacLeod on 27 November 2023
Google adds small business filter to Search and Maps
Posted by Edith MacLeod on 21 November 2023
Google releases Nov 2023 reviews update
Posted by Edith MacLeod on 9 November 2023
Interactive content: engaging your audience in the digital age
Posted by Brian Shelton on 8 November 2023
Google releases November 2023 core update
Posted by Edith MacLeod on 3 November 2023