Keyword Research Ninjas Tactics – SMX London 2011

Keyword research tools are powerful and plentiful and your competitors are using them for the same reasons as you are. This session explores keyword research “outside of the box”, mining unique or unusual sources of data for insights and opportunities that others overlook.

Richard Baxter

The problem with keyword research data is it’s a bit of a jumble, anyone can make a list, but how do you make it actionable for SEO?

How do you get from a dataset to IA?

If you can find a way to group and segment, and take away noise you can learn about priorities. By breaking down the following search query you can identify segments and groups within the phrase.

Buy used audi s3 black London

Action => condition => brand => model => colour => location

By identifying these segments you can then scale the search query to find a wider range of keywords.

The original query becomes:

Buy used audi s3 black London
Find new bmw z3 green Manchester – and so on…

Identifying the search demand curve

Make a pivot chart in excel – SEOGadget has a helpful guide

Keyword Strategy

If you have data that you can filter and make sense of then you can create keyword groups and ultimately a keyword strategy

Awesome Excel queries:



How do I categorise keyword data?

Create a table in excel, then use an array formula – (more info contact Richard Baxter on his Twitter account

An array formula is a formula that works with an array or series of data values rather than a single data value.

Increase your data – improve perspective

  • Downloading all of analytics
  • Use suggest API’s to increase your KW list
  • Capture search engine rankings
  • Build a data set

Why this is awesome?
This data reveals more about the mechanics of a search and allows you to understand the intent and nature of search queries a lot better.

  • How do people search for products in my indust
  • What terms are most important to my site
  • Where are the quick wins
  • Are there new types of search

Richard has created a list of his favourite tools with small descriptions

Christine Churchill

Christine sets the tone of her presentation with her opening sentence…

“Tools will make your life easy and keep you sane”

Christine continues to identify the best keyword research tools and some of their benefits.

Google Keyword Tool

People might have noticed a change in September. The tool used to pull data from Google and partner sites, but now is just Google which led to massive drop in search volume.

  • More info if logging in (more kwd’s and trend info)
  • Ability to compare desktop or mobile search phrases (filters)

Google Instant
This has a direct impact on user behaviour so it is worth spending the time analysing what keywords are suggested. I have personal experience in this and is definitely worth tracking the related and suggested keywords. Rishi has done a post previously about how you can manipulate Google Instant


Scrapes all suggestions from Google Instant and is a great way to download these into excel files.

Lasse Clarke Storgaard

Lasse declares himself more of a search philosopher than a practitioner and his first gem is:

“Our customer is our wife”

Lasse talks about search-economics and says it is not always important to be the first or the fastest but how well you carried out your strategy. Using a supermarket shelf analogy Lasse describe the consumer decision process to the position of products on a shelf.

Lasse states that there are 3 mindsets in the online space environment:

  • Purchase 6-8%
  • Consideration 10-15%
  • Research 60-80%

Although the vast majority of users start off with research driven search queries you still want your client to be present on the top shelf.

Understanding that ecommerce hasn’t evolved much is key, it still has research and purchase stages so it is important to position yourself at both of the stages and to also consider the intent of the user at every step.

Kevin Gibbons

How do you optimise for long tail traffic? Especially when Google themselves say between 20 – 25% of queries have never been recorded before?

You need to understand the nature of the beast, most searches are long tail, in fact 94% taken up by long tail compared to head so if you are just focusing on head in ranking reports then you are not seeing bigger picture.

Kevin outlines 10 steps to targeting long tail traffic process:

1)    Find common search trends 2)    Answer FAQs within niche (google instant, how to queries, yahoo answers, quora) 3)    Pick out popular themed keywords in GA 4)    Use segments to analyse long tail terms (focus on 4 or more keywords) 5)    Use PPC & impression share data 6)    Use multiple tools to verify results (strategiser by wordtracker, hitwise) 7)    Estimate avg CTR (SEOmoz and Econsultancy predict between 36-42% CTR for 1st position) WMT more accurate to your industry 8)    Use excel to predict traffic vales (use click distribution to help manage expectations of clients) 9)    Filter keywords into themed group

10)    Don’t overthink it and go too long tail


Recommended by panel to analyse internal site search, however be careful when you use these terms to create content. Make sure you use clean URLs without search query nodes and create unique content.

Look at how they are searching and look where they are searching, analytics can tell you what page they were on when they used site search. Could you identify their problem and provide the solution on the page?

Want to extract Google suggest results? Use a front end tool like ubbersuggest – or hire a developer from

Scraping Google for rankings? Make sure you distribute ranking request by range of IP addresses or just use existing tools such as AWR

Tags: Keyword Research

Leave a Reply

Your email address will not be published. Required fields are marked *