Keyword research at scale
Get 100 times the insight for the same cost with our unique process. We've broken the glass ceiling on keyword research with some well-trained machine learning models and a well-honed consultative approach, so you can scale your 5,000-word keyword research projects to 500,000 keywords.
We’re shaking up keyword research. Our service is an automated way of doing keyword research that removes the human limitations on speed and accuracy when working at scale. It gives you the same quality of research at 100 times the scale for a comparable price to traditional keyword research projects (basically what you’ll get anywhere else).
If that sounds too good to be true, it really isn’t – it’s just taken us a major investment of time and resources, and a lot of trial and error, to come up with a machine learning-assisted process that works.
Why do keyword research at scale?
Traditional keyword research gives you a good idea of search terms related to a topic, or your business. Scaled keyword research gives you more like a complete picture of the entire search landscape for your business. We call the result your Keyword Universe.
Your Keyword Universe is just that: a huge-scale, high-fidelity view of all the search terms related to your business, categorised and accessible in a way that opens up possibilities your competitors can only dream of.
The benefits of keyword research at scale
The bottom line is you get much more insight for the same investment.
And it’s not simply ‘more of a good thing’. By scaling up your keyword research insights, you can scale up the benefits exponentially – you can do things that simply aren’t possible at lower volumes.
- Find an exhaustive range of topics and themes your site ranks for, uncovering strengths and gaps you never knew existed
- Get a complete picture of all your competitors in all SERPs – not just the big players, but targeted opportunities
- Identify your competitors’ strengths and weaknesses in microscopic detail, and strategise accordingly
- See patterns between ranking topics and terms that no one else can see
- Instead of paying for menial labour, pay for expert consultancy and let machine learning take care of the grunt work
- See at a glance whether a whole business proposition, or years’ worth of content strategy, is viable
- Get more accurate data
How keyword research at scale works
The process of doing keyword research has changed relatively little in the last decade. SEOs have almost constantly been coming up with new ways to use the outputs – advanced rank tracking methods, SERP analysis, CTR modelling, techniques for content strategy. But the method of actually getting the keywords in the first place has remained pretty much the same:
Create seed keyword list > Get keywords and associated data from chosen keyword tool > Categorise keywords
The tools have changed a bit, and exact methodology varies from one business to another, but there’s been basically no getting away from the manual effort needed to gather keywords and, crucially, to categorise them. Until now, if you wanted a project to output 400,000 keywords, it was theoretically possible – if you were willing to pay your SEO agency for 100 times the person-hours.
Innovation – no, really
Keyword research at scale isn’t simply a different spin on traditional manual keyword research. We’ve taken the whole process apart and put it back together again to create a new approach. It’s a completely new proprietary service offering that no one else is doing at the moment, and we’re proud of that.
We use a variety of technologies and processes that includes machine learning, Python, various scripts, some specific APIs, and natural language processing (NLP) packages. We combine that with human expertise and industry insight to build the whole thing on a water-tight seed keyword list.
This last part is crucial. We’ve honed our process through long experimentation and training of machine learning models. And we’ve found that the adage rings true: “garbage in = garbage out”. We’ve spent years training our machine learning models and know exactly how much human guidance they need. This approach means that, instead of days and days of grunt work, the human resource you get is put to use on highly skilled consultation and analysis, defining the inputs and producing a database that works for you.
In short, it’s a combination of precise, targeted automation and thoroughly validated marketing expertise.