Google’s Keyword Planner has been a staple for both advertisers and SEOs to get an idea of traffic volume for keywords. This helps identify the keywords with the most traffic as well as identify keywords that have lower volume, which often means less competition. But a major changes to Google’s Keyword Planner has made those traffic volume estimates much less valuable.
Last week, Google made a change to the tool where instead of showing individual keyword estimates for each keyword or keyword phrase, Google is now lumping in the data together, meaning Google will show identical estimates for similar keywords or keyword phrases – advertisers will recognize this as Google combining search variants. You can no longer see individual estimates to see which of those has the highest volume and which the lowest, and valuable aspect for both advertisers and SEOs who are choosing which keywords to target based on volume .
This also has another major implication. For those that don’t notice the change – or worse, pulling the data from tools that haven’t updated to take into account the change – this means that some advertisers and SEOs are grossly overestimating those numbers, since many tools will combine data, and there is no notification alert on the results to show that how Google calculates average monthly searches has been changed. So the data could erroneously be combined, instead of noting that the estimates Google is provided are combined already for many, leading SEOs and advertisers to think search volume is far higher than it really is.
For example, here are the results for two keywords – SEO and search engine optimization. Instead of reporting individually, they are now showing identical data, even though previously it showed the data individually. For someone not paying attention, they could believe that these two individually have a search volume of 1,346,000 average monthly searches total. But in reality, it is SEO and search engine optimization TOGETHER that have 673,000 monthly searches.
However, there is even more problems. There could also be ADDITIONAL keywords that are included in that 673,000 figure, but the Keyword Planner does not state this. So estimates could be significantly off for someone thinking the total search volume just for “search engine optimization” keyword alone is 673,000 per month.
The Keyword Planner now seems to combine many search variants, including:
- plurals with non-plurals for any word in the keyword phrase
- acronyms with longhand version
- stemming variants: -er, -ing, -ized, -ed etc keywords (ie. designer, designing, designed)
- words that can be spelled with or without space (ie. car park and carpark)
- words with and without punctuation (ie. kid toys and kid’s toys)
In some ways, it does make sense that Google is showing close variants with the same data. But for some advertisers and SEOs, sometimes those close variants can have vastly different “real world” results, even if Google is now lumping them together.
This change could be connected to the fact Google also began restricting access to the Keyword Planner tool to those with active AdWords campaigns only, a decision they later seemed to backtrack on, changing the “requirement” to a “bug”. And with Google using close variants in AdWords, then perhaps this change makes sense. But for those who use it as a keyword volume tool, the value of this tool just diminished significantly.
Jennifer Slegg
Latest posts by Jennifer Slegg (see all)
- No Plans for Google to Mark HTTP as Insecure in Search Results - September 22, 2017
- Google: Do HTTPS Migrations Separate From Other Major Changes - September 22, 2017
- Google: Rankings Should Remain Stable With HTTPS Migrations - September 21, 2017
- Google: Value (or Not) of Doing Link Audits - September 20, 2017
- Google Indexes AMP Version for Mobile First When No Regular Mobile Page - September 19, 2017
Joe says
Is there a setting we can use to seoerate the two?
Jennifer Slegg says
Nope 🙁
Joe says
Doesn’t this hinder our PPC friends too?
Jennifer Slegg says
AdWords does search variants now for advertisers.
Steve says
I discovered this a couple of days ago. Two people at AdWords have given me different views: one of them (via Twitter DM) thinks it’s a bug but is having trouble confirming internally, while another (via phone support) ignored my question and said that if I want accurate data, I should run some actual campaigns and get the impression numbers, and to also give AdWords’ Campaign Experiments a try.
Classic Google: two people giving two different answers, while one of them says you now have to pay to get the data you want. Rubbish,
I’ll report back if I get a solid answer from someone either way. Have you had it confirmed to you by someone at Google, Jennifer?
Troy Fix says
It’s under “Close Keyword Variations” – https://support.google.com/adwords/answer/2497836?hl=en.
In Google’s example, they say: For example, if your phrase match keyword is “kid’s scooter,” you’d still want to show your ad when someone searches for “kids scooter” or “kid scooters.”
My issue is it isn’t consistent. If you look at “bachelor degree” & “bachelor degrees”, they have 90K searches/month. The keywords “bachelors degree” & “bachelors degrees” have 18K searches/month.
I don’t disagree with what Google is trying to do. However, the tool isn’t giving consistent data, So when everyone tries to prioritize or do projections, they’re going to be waaaay off.
Oli says
I don’t think this is necessarily a completely bad thing, semantic indexing means we should be talking about topics over keywords, this helps give us traffic levels without getting stuck in the keyword mindset.
While it’s better to look at specific keywords for sure, this might actually push us towards a higher focus on content instead of keywords.
Dawid Sakowski says
You still could get some useful data from “Detailed forecast” feature of Keyword Planner.
Add both keywords to your plan, set your bid to $100 and download including impressions, clicks and forecasts.
Your example (seo, search engine optimization) gives the following results:
– [seo] –> impressions 476,174.97
– [search engine optimization] –> 43,825.05
jason says
Would you say that this follows Google’s transition with its original Hummingbird update, where they focus less on the keyword itself and more on the true user intent?
Lorenzo Solis says
It has a positive point of view. Now we’re going to know how different terms are semantically connected.
Suresh says
Hope we can see that data using the SEMRUSH.
C says
Thanks for this post! Great to know, I use that tool a lot. Seems like every day Google pushes users more and more towards PPC. Sad stuff, but I guess it makes sense from business point of view.
Ash Nallawalla says
Good pickup. This isn’t happening with some sets of noncommercial keywords, e.g. raaf, royal australian air force, – but happens for USAF, united states air force.
MeshCloud says
I have used google keyword planner to send over 200,000 words for a project to determine the highest volume ‘ideas’ and concepts. I find this interesting and possibly helpful, because I was frustrated that I would have to send every variant of a word in order to get a relative idea of it’s popularity.
I guess I was using the tool already in the way that google now intends for it to be used. That is to find a language to describe something using popular words and topics. I’m not looking for exact keywords to stuff in an article. But rather to determine if people use the word abstract or modern more when describing a funky vase.