Wordtracker offers multiple data sources for keyword research

Posted by Mal Darwen on 2 Oct, 2013
View comments Wordtracker NewsWordtracker Tools
Wordtracker's Mal discusses how to use multiple data sources for keyword data - includes Wordtracker and SEMrush keyword result comparison.

It's generally accepted that when you're gathering market data, the more data points you have, the better.

That applies well to keyword research - and given that no keyword tool is perfect, and no keyword search database has all the information from all the search engines in all the countries, then spreading the net more widely than a single source is going to be a no-brainer.

That's why Wordtracker make sure there are two data sources (three if you count the Related Keywords tool, but that’s a different story) available in the Keyword Tool, so that you can compare the keywords delivered, measure the results, and make better informed decisions about how you ultimately present your carefully crafted pages to the world - human readers and the search engine robots.

As a bit of background, it’s worth mentioning that Wordtracker’s database contains around 489 million keywords, with 14 different territories available to search in. SEMrush’s database is around 91 million keywords strong, with 9 territories to research from.

 

I'd have mentioned as well that a great place to do keyword research would have been Google Analytics, but the plug on tracking keywords that bring traffic to your site via organic search has now been firmly pulled.

Why are multiple sources useful?

It’s not complicated to start making comparisons. Taking the example keyword ‘acoustic guitar’ (I'm in the rock’n’roll industry today; I got bored of making chocolate), I search both Wordtracker and SEMrush, being careful to use the same territory for both - US in this case. SEMrush gives me 1000 keywords and Wordtracker's database offers 2000. A fair chunk of data from each source, and plenty to work from.

It's difficult to show the detailed results in a screenshot (we don't need to have a 3000 keyword long page here), but here are a couple of things that make double data helpful.

 

Wordtracker data

SEMrush data

 

 

1 - SEMrush for the head, Wordtracker for the tail.

There's generally more long tail detail in the Wordtracker data, as the SEMrush data broadly contains keywords up to three or four words long - whereas Wordtracker's keywords (partly because there are more in the list, and partly because of the nature of the database) can be longer - there is often lower volume on these keywords, but in a keyword like 'guitar lessons for beginners acoustic easy songs', you could find yourself getting traffic for 'guitar lessons for beginners','easy guitar lessons', 'acoustic guitar beginners songs' and so on. Even pages optimized for low volume keywords can attract traffic - if they're properly organized and have clean, healthy link profiles.

The SEMrush data is helpful because while it's not always as detailed as we might like to see in terms of the long tail (and also because the the values are banded estimates rather than the exact numbers that Wordtracker data reports), it can flag up really strong potential niches - for example, in the top 20 keywords there are some great ideas like 'acoustic guitar tabs', 'acoustic guitar chords', 'acoustic electric guitar' and so on - while no-one with a new or small site is going to be able to rank quickly for those terms, there are undoubtedly keywords around those niches that will make really good starting points for pages inside categories built in those terms, although they may not have huge returns in the SEMrush data.

2 - Confirm your thinking

Because it's so easy to get keywords from either source, it's also easy to confirm or deny our initial feelings about a particular keyword - it's easy to fall into the trap of using jargon rather than our customers' language on your site, but keyword research is the thing that's going to challenge your preconceptions - and from that, encourage your mindset into that of your customer. Let's take a look at some popular keywords for a quick and easy overview.

The first thing that pops out from both Wordtracker and SEMrush data is that 'acoustic electric guitar' is a very popular search in this niche. While it's a fairly short 'head' keyword term, it's one that you could probably optimize for (including the longer tail keywords on your pages), along with another keyword niche indicated in the top 30 of both lists - 'acoustic guitar lessons'. Drill into popular search niches in both sources - and you’ll find keywords inside those niches when you search.

Great. But the numbers are so different....

When anyone sees a list of keywords against a list of numbers, the strong temptation is to trust the numbers we see so that we don't have to think too hard about the rest of it. This might not actually be helpful to you in the long run - the important thing to look at is the relationships between the numbers - so to ask yourself 'Is a keyword appearing in both lists?', 'Is it roughly in the same position in both lists?' 'Do the lists correlate?'

It's not the end of the world if all three don't match up perfectly - because the raw data that feeds each of the sources in Wordtracker is differently sized and differently acquired, disparity is to be expected - but there will be some broad correlations (as we've just seen) to at least get you started.

When you're doing keyword research, that temptation I mentioned to slavishly trust numbers could actually be your undoing - as well as the keyword data itself, one of the most important elements in the process is your own common sense and your market insight - whether this is garnered from years of experience in that area or from a couple of hours analysing the competition's websites, it's a vital part of the process - all the data in the world can't replace your own mind - but apply your thinking and experience to the data, and you have more strength for your SEO arsenal.

It’s easy to find out more for yourself, though - take a free trial to check out the joys of multiple data sources.

Recent articles

How to design your website using user intent
Posted by Nicole Stelmar on 26 June 2017
The beginner’s guide to voice search engine optimization
Posted by Rebecca Appleton on 19 June 2017
PPC campaign structures: why more segmentation isn’t always better
Posted by Shane Barker on 19 June 2017
Content marketing metrics: how much do they matter?
Posted by Andy Beohar on 19 June 2017
24 Google Doc hacks and add-ons to make your life easier [infographic]
Posted by Rebecca Appleton on 14 June 2017