Ok Google – how do I optimize for voice search?

Ok Google – how do I optimize for voice search?

28th April 2015

In May, I’m speaking at SMX London on how SEO is moving away from a traditional keyword approach to a broader focus on entities and conversation. For me one of the biggest factors in this shift is the increasing adoption of voice search and, because I seemingly can’t focus on two content topics at a time, this month’s post is a preview of my talk.

Firstly, it’s important to say that voice search is not new, in fact, Google first included a ‘search by voice’ feature in it’s iOS app as early as 2008, however what is new is a significant increase in adoption, with Google reporting a more than double increase over 2014.

Aside from a big marketing push on Google’s part, also contributing to the increase of adoption is that Google Now is actually starting to get really good; as an SEO, I find the way it can build on, and understand, compound queries genuinely exciting. (I’m a riot at dinner parties).

For example if I say ‘OK Google, who is the Prime Minister of the UK?” it will chime up David Cameron, as you might expect. Then if I say “What school did he go to?” it will understand I am talking about Cameron and return Eton, then when I ask “what other Prime Ministers else went there?” it will return a list of the other career politicians Eton has churned out so many of.

This move to genuinely engaging in conversation with our phones is a colossal shift, both ideologically and technologically, and is set to completely change the way we engage with our devices long term.

What is even more exciting, though, is that we are in a position to influence it.

So how can we optimize for the voice search (r)evolution?

Well I think you can break it down in to three main areas; mobile, local and the Knowledge Graph.


The lion-share of voice search is conducted on a mobile; you can get desktop voice search, but nobody seems to really care all that much. It’s probably worth noting that I’m basing this purely on intuition and after some searching couldn’t find any research on it just yet, but we can reasonably assume that this is the case.

By that assumption then, the same algorithmic rules apply for voice search than they do to any mobile optimization. If you still need to create a mobile friendly experience the Google guides are probably the best place to start.


Studies have shown that mobile queries have strong local intent (I’ve seen anything between 40%-80%) so if we can reasonably assume that voice search is happening primarily on a mobile device, then we can reasonably infer that a sizeable chunk of these will have local intent. More so, this is the kind of intent we want to make sure we’re properly optimised for as on average, a mobile query results in 2 follow up actions and for transaction queries, 50% of these are in store.

Moz has a comprehensive Local SEO guide to get started. Though if local has always seemed a marginal part of your SEO strategy; it really is time this changed (consider incorporating iBeacons in to this, too).

Knowledge Graph

In the latest version of the Google Search App for iOS and Android, not only can you say your question out loud, but your search app can speak your answer right back to you. And, using Google’s Knowledge Graph, your search app gives you smarter answers loud and clear. [Source]

Google will chat back to you when the query has a knowledge graph result; she’s currently fairly good at having a natter about the weather, giving directions, music, events, famous people and brands and as the knowledge graph expands, so will her vocabulary. Equally, as the web increasingly adopts a common markup format – the more this data can be transferred between applications. Again, if you’re still at the beginning of your markup journey, the Google guides are probably the best place to start.

So as we know, to optimise for voice search, we need to optimise for Knowledge Graph, meaning that we need to mark up as much of our data as possible.

Though, it’s no coincidence that only 0.3% of domains use schema – implementing mark up can be complicated, resource heavy and expensive.

Fortunately there some shortcuts; if you saw my talk at BrightonSEO I tried to address this by discussing the possibility of injecting JSON-LD using Google Tag Manager. There’s also the Data Highlighter Tool and multiple plugins that help if you’re using WordPress. (I’ve used the Raven one in the past).

In summary, voice search is important right now and it’s going to be even more important in the future so we need to start considering it in our strategies. However this consideration includes nothing additional beyond what we should be doing already. That is, have a technically robust site that is mobile friendly and semantically marked up and, if you’re a local business or have physical locations, make sure your local SEO is performing as well as it could be too. If nothing else the importance of voice search can only go to add an additional line of justification to our business cases as to why we need investment for all of the above.

Actually, before I finish this post I should probably also make a nod at this point to Siri and Cortesha? Courtney? Carlos? Ah, yes, Cortana, as they are all equally players in this field, but for me the ultimate message stays the same – voice search is happening now and if we don’t build it in to our way of thinking we, along with our clumsy thumbs, are going to get left behind.


Written By
Kirsty Hulse is Head of SEO Best Practice at Linkdex and has six years experience defining search strategies for some of the world's biggest brands, as well as small E-commerce start ups.
  • This field is for validation purposes and should be left unchanged.