Keyword Data Accuracy & Data Manipulation by SEO Tools [In-Depth Study]

The work of a search engine marketing (SEO) guide revolves round one central theme:


Especially key phrase knowledge.

We acquire it from quite a lot of third- and second-party sources, even perhaps by way of self-made monitoring instruments, to then begin crunching the numbers and finally delivering worthwhile insights to our bosses, shoppers, or prospects.

However, solely working just a few instruments and using some analytical magic just isn’t going to chop it.

We additionally have to be considerate about how we interpret knowledge from key phrase instruments and cope with any inaccuracies or inconsistencies.

Just like all software program program, every key phrase instrument has a attribute mechanism in place for gathering, aggregating, and manipulating knowledge.

Similarly, instruments’ workings have an effect on how they deal with queries and current the output key phrase knowledge.

An important a part of a marketer’s job perform is to validate whether or not the info values saved for these key phrases are represented in a constant and unambiguous type.

Meaning, is the key phrase knowledge I’m working with correct?



The easy reply:


Comparing knowledge values of various instrument suppliers for a set of key phrases already proves to include massive inconsistencies – not solely in knowledge values but additionally in if and the way your output knowledge is offered.

This examine, by my firm, OAK, makes an attempt to search out readability by exploring knowledge accuracy and reliability with respect to second- and third-party key phrase instrument knowledge.

Specifically, this examine examines the next subjects:

  • Data assortment: How do key phrase instruments acquire their knowledge?
  • Data dealing with: How do key phrase instruments manipulate knowledge?
  • Data validation: Validating key phrase knowledge values.
  • Role as an SEO Consultant.

The main objective of this examine is to develop consciousness concerning the complexity surrounding key phrase knowledge values and gear suppliers’ knowledge gathering and processing mechanisms.

Google Search Console

Let us start initially: Google Search Console.

It is a second-party instrument from Google that collects behavioral knowledge for a single area or entity and, after manipulation, injects the info into the front-end interface.



The mere incontrovertible fact that Google collects and processes the info would possibly you surprise: how near actuality are the info values of the projected knowledge?

This query poses a right away problem: Search Console knowledge just isn’t 100% validatable.

Luckily, Google is, to some extent, clear and gives varied explanations for why your knowledge values don’t mirror actuality or add up as you might anticipate.

A couple of of them are:

  • To defend the privateness of the consumer. The click on is usually not credited to the search time period. Search Console, nonetheless, does register the clicking, inflicting discrepancies between the desk and diagram knowledge.
    • The identical can apply to branded queries.
  • Clicks may come from bots.
  • In some instances, deciding on sure filter combos can even result in variations between the diagram and desk knowledge.

Unfortunately, solely the G-giant has entry to the precise knowledge values, which suggests verifying the accuracy of Search Console knowledge is a hard course of.

The reliability of key phrase knowledge will increase, nonetheless, with third-party instruments.

These are instruments like SEMrush, Ahrefs,,, and lots of others.

To discover solutions, this examine explores the mechanics of those key phrase instruments apply.

Unfortunately, the businesses working these instruments disclose little to no details about how they acquire, combination, or manipulate their knowledge.

It appears honest.

A chef doesn’t simply give away her or his world-famous recipe. Hence we try to generate insights with the assistance of the next approaches:

  • Using and evaluating the instruments.
  • Inquiring on the customer support departments.
  • Reading the FAQ sections and utility pages.

1. Data Collection: How Do Keyword Tools Collect Their Data?

In basic, there are 5 sorts of sources by which key phrase instruments accumulate their knowledge:

Google Ads API / Keyword Planner

Keyword knowledge is gathered straight from Google’s key phrase database by the Google Ads API.

As is the case with Search Console, Google Ads first manipulates the info earlier than injecting it into the database.



Clickstream Data by Aggregators & Data Brokers

Clickstream is nothing greater than knowledge derived from shoppers’ on-line browsing conduct.

Aggregators collect this knowledge in quite a lot of methods.

Large, till just lately energetic aggregators, have been, for instance, Jumpshot or Hitwise.

Wherefrom do they get their knowledge?

  • Browser extensions and plugins
    • A selfmade plugin or extension of the aggregator itself.
    • They pay exterior third occasion browser plugins to share client knowledge
  • They pay web service suppliers for entry to the info in an “anonymized” knowledge feed.

The aggregators then promote the info to key phrase instruments resembling Ahrefs, SEMrush, and Moz, amongst others.

Browser Extension & Plugins

Keyword instruments can even straight purchase client knowledge from exterior third occasion browser plugins.



Numerous browser extensions have been developed to assist entrepreneurs.

Despite the nifty functionalities, these plugins and browser instruments typically pursue shady practices.

Providing consent earlier than utilizing an extension is frequent, however we often have restricted information concerning the practices to which we give consent.

By consenting, you may allow these instruments to:

  • Collect your on-line browsing conduct.
  • Retrieve shopper knowledge from Google Analytics, Search Console, or different monitoring software program.

And most worrisome:

  • Share the info with third events resembling aggregators or key phrase instruments.

These browser extensions might need entry to any doubtlessly delicate knowledge and are ordinarily not compliant together with your buyer or enterprise’s GDPR.



Exercising care when working with extensions is critical to ensure knowledge safety.

Some key phrase instruments have additionally developed their very own browser plugin or extension.

Moz, as an illustration, launched MozBar, an all-in-one extension with all types of helpful options.

Browser plugins developed by the established key phrase instruments presumably wouldn’t pursue any malicious practices, however they will acquire on-line conduct and use it to attune their knowledge values.

External Tools

Keyword instruments moreover retrieve knowledge by way of APIs from exterior events that acquire on-line browsing conduct knowledge, resembling GrepWords prior to now., as an illustration, obtains knowledge from Keyword Planner but additionally different third sources. They disclose some recommendation, albeit reasonably basic:

“Keyword Tool provides an external API that gives you the keyword suggestions that you would never be able to find in Google Ads. Keyword Tool uses autocomplete data while Google Ads hides valuable keywords that could be found using autocomplete.”

Own Tools

Some key phrase instruments as an alternative have their very own packages or software program set as much as accumulate key phrase knowledge.



These 5 strategies of gathering and examples solely unveil the tip of the iceberg.

There exists an unlimited internet of firms and instruments inside this world of gathering, exchanging and promoting key phrase knowledge.

At least a bit higher understanding of the image will undoubtedly assist us to appreciate that key phrase knowledge displayed on our SEO instruments, Google Sheets and dashboards are barely greater than a product of an ambiguous assemble.

To Sum Up

Keyword instruments acquire knowledge from 5 several types of sources.

It’s frequent to make use of a number of knowledge sources from completely different knowledge supply varieties.

2. Data Handling: How Do Keyword Tools Manipulate Data?

The subsequent step in adopting a extra vital stance towards key phrase knowledge is by studying how instruments combination and manipulate the info they’ve obtained.

It is near inconceivable to search out out simply how precisely instruments run this process.

The follow of getting this proprietary data is the same as making an attempt to find Coca Cola’s recipe – futile.



Instead, allow us to settle with the notion that aggregation in itself could additional taint the accuracy and reliability of information.

One may argue the reverse that aggregating knowledge can mechanically flatten out any excessive knowledge values.

After all, merging these knowledge sources right into a single coherent aggregated sort will yield a greater approximation of the common metric values.

Even although it’s a respectable stance, key phrase instruments nonetheless collect knowledge from sources that in itself could also be incalculable, biased, and incorrect of their measuring mechanics.

Regardless, key phrase instruments do greater than mixing completely different knowledge sources right into a single set of information.

Schematic view keyword tool mechanism for collecting, manipulating and querying data | SEJSchematic illustration of key phrase instruments’ workings in gathering, manipulating and querying knowledge.

Running the Data Through an Algorithm

Some instruments have developed an algorithm that capabilities as a filter for his or her collected knowledge.

For occasion, SEMrush explains:

“To ensure the highest level of accuracy, SEMrush uses its Neural Network – a combined algorithm that references various sources of data and recognizes patterns in the same way the human brain understands patterns. The data sources in our network include clickstream data in addition to our own database of backlinks and organic search engine positions. “



It seems a logical explanation that SEMrush uses its algorithm to validate the obtained external data and attune the aggregated metric values where needed.


Keyword tools can group keyword data, which proceeds in two different ways:

  • They group metric values of search term variations into one.
  • They group variations of search terms into one.

This grouping mechanism relies on four linguistic determinants:

  • Plural vs. singular nouns in the keyword.
  • Combinations of articles and prepositions.
  • Usage of regular, comparative, and superlative adjectives.
  • Placement of adjectives or interrogative pronouns.

Volume Grouping

Let’s begin with an example.

We have two different search terms, “door handles” and “door handle.”

Some instruments, whether or not key phrase instruments, aggregators, or different knowledge gathering instruments, merge the person quantity values into one combination whole, and show this whole for each key phrases.



For instance, teams the key phrase, whereas doesn’t.

That is what it appears like for the U.S.:

table with keywords and volume values - SEJ

Two issues stand out instantly:

  • attributes the identical quantity worth to each key phrases (plural and singular), whereas doesn’t.
  • Volume values of are considerably decrease in comparison with that of

Let’s additionally check out Ahrefs.

Ahrefs collects its knowledge from Keyword Planner, amongst different sources.

According to the customer support division, Ahrefs ungroups the key phrases that Keyword Planner teams collectively.

The following desk is just like the earlier one, however this time we now have included knowledge from Ahrefs and pulled the identical question for one more nation.



Keyword data volumes for specific keywords for 2 countries - SEJ

Two issues stand out:

  • Ahrefs knowledge values reveal completely different quantity values in comparison with each and
  • For the Netherlands, Ahrefs does assign a better worth to the singular type “door handle” in comparison with the plural sort “door handles.” It’s diametrically against’s values.
    • However, for the U.S., and Ahrefs exhibit the identical equilateral distribution.

Query makes an attempt with different key phrase units give us related outcomes. In some instances, instrument X presents the most important values, in different instances, instrument Y or Z.

One factor is certain: knowledge values are scattered, questioning the reliability of information values.

The subsequent desk lists a set of standard key phrase instruments and whether or not or not they group key phrase volumes:

Keyword grouping per keyword tool - SEJ

Keyword Grouping

Besides quantity clustering, the grouping impact additionally applies to look phrases.



The first linguistic determinant in key phrase grouping is singular vs. plural utilization of nouns.

Singular vs. Plural

Keyword instruments can group nouns into both the singular or plural type.

However, this doesn’t essentially imply the opposite model, whether or not singular or plural, doesn’t exist within the instrument’s database.

Tools select which type to show within the output.

We’ll illustrate with Keyword Planner.

Let’s suppose we need to retrieve the U.S. search quantity of the following 4 key phrases.

Keyword data volumes for specific keywords for 2 countries - SEJ

Next, we select the tab Historical Metrics exhibiting the next knowledge desk:

screenshot table keyword data keywordplanner for 2 keywords - SEJ

Two issues that stand out instantly:



  • Keyword Planner returns solely knowledge for 2 out of 4 key phrases.
  • Keyword Planner returns solely the singular type of the nouns.

When repeating this question for different nations, we are able to’t observe any logical sample between the question and the offered knowledge.

For instance, for the Netherlands, Keyword Planner serves the next desk:

screenshot table keyword data keywordplanner Dutch keywords - SEJ

For these with restricted comprehension of Dutch language:

  • “Deurklinken” (i.e., “door handles”) is plural.
  • “Deurpost” (i.e., “doorframe”) is singular.

Keyword Planner thus teams primarily based on volumes in addition to key phrases.

Looking at each the U.S. and the Netherlands, we are able to infer that Keyword Planner’s database does include the info values for each the singular and plural type.



To ensure, let’s rerun the question.

Only this time for simply the U.S. and with the plural types of the nouns:

  • “door handles”
  • “doorframes”

The outcomes:

screenshot table keyword data keywordplanner English keywords plural - SEJ

While Keyword Planner omits both the plural or singular noun within the export, its database does embrace knowledge on all 4 key phrases.

Also, the key phrase quantity values mirror the aggregated volumes for each the plural and singular type.

Likewise, there appears no clear floor on which Keyword Planner decides which type to show aside from an arbitrary one.



One can additional discover this matter by, as an illustration, evaluating a number of nations, industries, quantity ranges, and languages.

More in-depth exploration, nonetheless, is past the scope of this examine.

The truth nonetheless is, it causes an excessive amount of confusion.

Combinations of Articles & Prepositions

Tools group key phrases in cases the place search phrases comprise articles and / or prepositions.

To illustrate, we offer an instance of

We compiled an inventory of eight key phrases to extract from’s database:

  • “legislation in the united states”
  • “legislation in united states”
  • “legislation the united states”
  • “legislation united states”
  • “legislation in the us”
  • “legislation in us”
  • “legislation the us”
  • “legislation us”

For the conspicuous reader, the checklist is as follows:

  • We used two methods to jot down down the United States: “United States” or “US”.
  • We created 4 combos of the article “the” and the preposition “in”.
    • “In the”
    • “The”
    • “In”
    • – (So neither “the” nor “in”)



Querying the info from’s database provides us the next knowledge:

screenshot table keyword data 5 english keywords - SEJ

Several issues stand out:

  • The question prompts outcomes for under 5 out of the ten key phrases. There appears to be no apparent issue that decisively impacts this specific output. The excluded key phrases:
    • “legislation in united states”
    • “legislation the united states”
    • “legislation the us”
  • Keywords with each the written type “United States” and the abbreviation “US” are offered, however, clearly not for a similar variation of articles and prepositions:
    • Listed: “legislation in us”
    • Not listed: “legislation in united states”
  • The mixture with out each the article and preposition is given for each the “US” and the “United States” variant. Still, each show different quantity values:
    • “legislation united states” — 210
    • “legislation us” — 40
  • Grouping of volumes happens “cross-keyword”. Both variants of “US” and “United States” in addition to preposition and article variants exhibit the identical metric values. It means teams the quantity values of the next key phrases:
    • “legislation in us”
    • “legislation us”
    • “legislation in the united states”
    • “legislation in the us”

Pertinent questions that come to thoughts:

  • Why is it that the mix “legislation united states” eludes clustering?
  • To what extent do articles and prepositions play an element in key phrase grouping?
  • How come the actual 4 ungrouped key phrases present clustered quantity values?
  • Is there any exact, specific mechanism in place that regulates the presentation of queried knowledge?

These are respectable inquiries to which we, sadly, don’t have a grounded reply. makes use of the Google Ads API to retrieve key phrase knowledge.

Can we then additionally anticipate the identical to occur with Keyword Planner?

We examined it by working the identical question for Keyword Planner:



screenshot table keyword data keywordplanner 5 keywords with missing values - SEJ

It produces fairly a special state of affairs.

Besides the earlier observations, we are able to additionally observe that Keyword Planner solely lists 4 out of the eight key phrases.

Also, values are given just for two out of the 4 key phrases.

Usage of Regular, Comparative & Superlative Adjectives

Adjectives or interrogative pronouns and comparative and superlative adjectives do play an element in instruments’ grouping mechanisms.

At first, it won’t appear such an enormous deal. For occasion, if we seek for “clean hotels London” or “cleanest hotels London”, the intent and the corresponding SERP outcomes are each fairly related.

In different instances, nonetheless, guests’ wants and intentions do profoundly differ. Let’s contemplate the following three key phrases:

  • “low blood pressure” – I’ve low blood stress and wish to have data on blood stress ranges which are thought of low, and maybe what to do about it.
  • “lower blood pressure” – I’ve hypertension and I wish to have data on how I can decrease my blood stress ranges.
  • “lowest blood pressure” – I most likely really feel fairly unhealthy, and I wish to know what blood stress ranges one can have with out it being life-threatening.



Looking on the instance, we are able to see variations in:

  • Audience.
  • Audience’s well being circumstances.
  • Informational wants.

This instance highlights the plain incontrovertible fact that we shouldn’t deal with these variations with the identical content material, or cluster these three key phrases into one matter bucket.

What occurs if we pull knowledge from key phrase instruments for these specific search phrases?

Inserting them into prompts the next outcomes for the UK:

screenshot table keyword data 3 english keywords - SEJ

It’s fairly clear: all metrics offered share the identical worth.



Before drawing any conclusions, allow us to first cowl the ultimate determinant.

Placement of Adjectives or Interrogative Pronouns

The variable placement of adjectives or interrogative pronouns constitutes the final linguistic determinant within the grouping mechanism.

It doesn’t occur too typically, however generally we place adjectives or interrogative pronouns on the center or finish of a phrase as an alternative of the start.

For occasion, one can seek for:

  • “electric scooter fast” or “fast electric scooter”
  • “how fast electric scooter” or “electric scooter how fast”

Either case carries the identical want for data.

It turns into reasonably fascinating after we add comparative or superlative adjectives to those examples and create new combos resembling “electric scooter faster”.

The level is, variations in interrogative pronouns or comparative and superlative adjectives can exhibit divergences in customers’ intent and wishes, and the kind of viewers customers belong to, as the following desk illustrates:



keyword data in a table with 4 variables - SEJ

Unfortunately, such latent data is difficult to derive from these third-party instruments’ question output, particularly when grouping is at play.

The subsequent desk by illustrates this:

screenshot table keyword data 4 english keywords - SEJ

Perhaps not surprisingly, the quantity displays aggregated values.

The conspicuous reader notices that the desk merely lists key phrases within the singular type.

Converting singular to plural type provides us the following knowledge supplied by

screenshot table keyword data for 4 english keywords with missing values - SEJ

For knowledge accuracy and reliability functions, the speedy motion right here is to validate, to the extent doable, the quantity values attributed to every key phrase.



One means to do that is by querying the identical 4 key phrases in different key phrase instruments.

Other instruments immediate completely different outcomes. For instance, SEMrush exhibits no outcomes and Keyword Planner was just like

Ahrefs and did current knowledge for all variations and, much more fascinating, with disproportionately smaller quantity values.

For occasion, the question for the UK in prompts this knowledge desk:

screenshot table keyword data - SEJ

That is a staggering 70 occasions the distinction of 310.

It’s true that the quantity worth of three.600 already displays the aggregated quantity for the set of six key phrases.

But sadly it occurs all too typically that entrepreneurs report all six – or maybe much more – variations in key phrase analyses.

We can propound the thought of selecting one variation and omitting the opposite combos. But it won’t clear up the problem.



The embedded data in key phrase variations about customers’ intents and wishes can merely differ, and thus any type is related to incorporate.

Imagine overlooking such a element, failing to see that every key phrase doubtlessly belongs to completely different clusters of key phrases destined for various pages.

And basing your visitors and monetary projections on these numbers.

It’s a compelling picture, albeit just a little amateurish. Nonetheless, one thing that occurs incessantly.

There is one statement left unattended. presents disproportionately bigger quantity values as in comparison with that of, as an illustration, Ahrefs.

Despite them each retrieving key phrase knowledge from Keyword Planner.

Apart from the instruments’ knowledge dealing with methods, what may trigger such a distinction?

Spelling Errors

Keyword instruments differ in how they cope with spelling errors.

Some, like, miss any spelling error variant in your question output.

Others, like Ahrefs and, do embrace spelling errors variants.

They each present the info values for each single key phrase in your question so long as the key phrase’s appropriate spelling variation exists in its database.



But as Ahrefs ungroups key phrase knowledge originating from Google Ads API, it does attribute distinctive metric values to every spelling error variant., however, adopts the grouped key phrases and metric values it retrieves from Keyword Planner, inflicting all spelling error variants to indicate an identical metrics.

Misspellings typically happen with model names simply susceptible to being misspelled.

Think of brand name names resembling Audemars Piguet, Breguet, Douwe Egberts, Schwarzkopf.

Let’s check out “Douwe Egberts”.

I’m Dutch, and as a local, I’m aware of the kinds of errors one could make.

For occasion:

  • Is Douwe with ou or au?
  • Is Egberts with g or ch and even with gh?
  • Is it Egbert or Egberts?

Point is: what occurs after we question an inventory of 1 single key phrase misspelled in 26 other ways?

Despite the grouping mechanisms in place, serves you each distinctive misspelled key phrase mixture:



screenshot table keyword data for 26 keywords - SEJ

Ahrefs’ scenario is a bit completely different. The question output is as follows:

screenshot table keyword data Ahrefs listing 20 keywords with missing values - SEJ

Things that stand out:

  • The question excludes 6 out of the 26 key phrases.
  • Ahrefs appears to independently attribute knowledge values per metric.
    • Except for the primary end result, which is the proper spelling variation, all different key phrase variations have both one or a number of metric knowledge values lacking.

Keep in thoughts that it solely works with the ‘list’ mode. The ‘explore’ perform will solely serve the proper spelling variation.

Omitting ‘PPP’ Data

Google Ads API omits key phrase knowledge involving ‘PPP’ subjects.

It signifies that different instruments retrieving knowledge from Keyword Planner additionally face this limitation until they enrich their database with knowledge coming from different sources.



For safety functions, Google disallows key phrase instruments to retrieve key phrase knowledge revolving round Porn, Pills, and Poker.

Think of key phrases like “cannabis” or “full house” but additionally key phrases resembling “Koffiemachine huren”.

“Huren” is Dutch for renting. But it additionally means “whores” in German.

While this ruling just isn’t a matter of direct manipulation, it does complicate entrepreneurs’ knowledge assortment and analyses.

A choice of key phrase instruments and whether or not they present”PPP”-data to your question:

table with names of keyword tool provider and data values - SEJ

The above examples illustrate the chaotic nature of key phrase instrument mechanisms and the hazards they impose on SEO consultants’ work.

To Sum Up

  • Keyword instruments don’t essentially present all key phrase variations and corresponding metric values.
    • Potential determinants: Tool’s performance, security or safety measures, or lacking knowledge within the database.
  • To our information as outsiders, it appears the actual show of combos of key phrase and metric worth variations is randomly “chosen”.
  • Grouping applies to each the numerical values and search phrases.
  • Linguistic determinants for key phrase grouping:
    • Plural vs. singular.
    • Usage or non-usage of articles and prepositions.
    • Placement of adjectives or interrogative pronouns.
    • Usage of comparative and superlative adjective.
  • Grouping happens each inside a specific group and throughout group variations.
  • Grouping happens at random.



3. Data Validation: Validating Keyword Data Values

Keyword knowledge validation is feasible, but with out entry to uncorrupted knowledge, it turns into an act of discovering the closest approximation to the key phrase’s precise knowledge values.

One choice is to benchmark key phrase impression knowledge values from Search Console to quantity values of third-party key phrase instruments.

Search Console knowledge isn’t 100% dependable both, however it’s as shut as we are able to get.



Early in 2020, we designed a examine to find out the accuracy of key phrase knowledge with a set of 160 key phrases from quite a lot of industries.

The examine tackled these two questions:

  • For every key phrase instrument, what’s the common deviation % of key phrase quantity values for the entire set of key phrases in comparison with Search Console impression knowledge values?
  • For every key phrase instrument, what’s the variance of all deviation % for the entire set of key phrases?

The former provides us insights into the diploma of accuracy for any given key phrase’s quantity worth.

The latter query determines to what extent the deviation % of every key phrase is unfold out from the common deviation worth.

As values can each deviate negatively and positively, it doesn’t suffice to merely present the common deviation %.

It is the mix, nonetheless, of each scores that yield one of the best leads to figuring out the info values’ accuracy and reliability.

As these visualizations illustrate, we see that solely both variance or common deviations can immediate faulty illustration of the scenario:

4 plots that visualise different combinations of variance and average deviation - SEJ

Measuring the variance of the common deviation % permits us to find out the scatteredness of every key phrase’s deviation share.



Large dispersions hints to decrease accuracy and thus reliability of the key phrase quantity values.

To put it in a different way, the extra substantial the variance, the upper the chance of choosing a key phrase from the info set showcasing a extra inaccurate quantity worth than the common quantity worth deviation.

These have been the important thing findings:

  • Twinwords quantity knowledge confirmed the most important constructive common deviation to Search Console impressions: +37.13%.
  • quantity knowledge confirmed the most important adverse common deviation to Search Console impressions: -34.71%.

chart with average deviation scores of keyword data values of keyword tooling providers - SEJ

  • The frontrunners with the most important variance
    • Twinwords: 5,259
    • 5,256
    • Keyword Planner: 5,188
  • The frontrunners with the smallest variance
    • Serpstat: 0.124
    • 0.149
    • Ahrefs: 0.153

chart with variance scores of keyword data values of keyword tooling providers - SEJ

Ideally, instrument suppliers exhibit numbers near zero for each the common deviation and the variance of the common deviation.



These findings present in any other case. Specifically:

  • The grouping impact drives the big variance and common deviation rating by the frontrunners.
    • Both and Twinword get their knowledge straight from Google. And since Google Ads applies grouping to key phrase and knowledge values, Twinword and mechanically undertake this impact.
  • Keywords with the most important knowledge worth deviations additionally seemed to be key phrases with grouped knowledge values.
  • Serpstat, Ahrefs, and current variance numbers near zero. These instruments don’t apply any clustering.
  • Serpstat and present significantly decrease common deviations. It means that quantity knowledge is on common decrease than what you’d anticipate in accordance with the Search Console.
  • Although Searchmetrics’ key phrase quantity values barely deviate on common to Search Console’s impression values, the person knowledge values are additional faraway from the imply, suggesting a better diploma of inconsistency in key phrase knowledge values.
  • The numbers of Ahrefs and KWFinder exhibit the closest approximation to the key phrase’s precise knowledge values.

Data values from third-party key phrase instruments fluctuate extensively and appear to fail in offering unambiguity or consistency.

The findings additional give the plausibility to the concept mechanisms in dealing with knowledge queries, and gathering or manipulating knowledge, can add to delivering faulty key phrase knowledge.

Can We Then Validate the Accuracy of Keyword Data in Another Way?

Together with Sander Tamaëla, a Dutch Freelance SEO-expert, we got here up with a solution to validate the accuracy of third-party key phrase quantity values with the assistance of Google Search Console and Google Trends knowledge.



The concept was as follows:

  • We picked one noun and chosen each the plural and singular type.
    • We had validated the accuracy of search volumes with GSC month-to-month common impression values.
  • Then we retrieved the quantity knowledge from two or three random key phrase instruments.
  • We then positioned these two key phrases in Google Trends.

With this setup, we may decide the relative curiosity between the 2 key phrases.

Our assumption right here was that Google Trends’s relative curiosity scores mirror the purest knowledge values.

As such, the relative curiosity rating ought to mirror a ratio just like that of impression values in Search Console.

  • Next, we expanded the set with key phrases – for which we all know we now have an correct approximation of the impression worth – from a number of quantity ranges.
  • Then we compiled a coaching set.

The concept was to find out per quantity vary the deviation per key phrase quantity worth primarily based on the relative curiosity scores of Google Trends.



Unfortunately, issues didn’t work out as deliberate.

After we had challenged the belief that Google Trends knowledge depicts correct values, we found that Google Trends isn’t completely dependable both.

To check the reliability of Google Trends we arrange the following check:

  • We chosen 5 key phrases with very related month-to-month impression values in Search Console.
  • We then added these 5 key phrases in Google Trends.
    • We made positive that we had chosen the identical 12 month interval for Search Console as for Google Trends: 1 December 2018 till 30 November 2019.

One of the five-keyword units, in Dutch:

Table with keyword data from search console - SEJ

The subsequent chart illustrates every key phrase’s impression worth deviation from the imply:

chart that visualises a plot of data points with small variance and small average deviation - SEJ

The month-to-month impressions common deviation in percentages was just one.92%.



Unfortunately, it is just doable to pick as much as 5 key phrases in Google Trends, limiting our pattern to 5 key phrases.

Such a pattern measurement is statistically not a major illustration of the inhabitants. The solely various was to repeat the check setup with completely different units of key phrases.

If Google Trends is dependable, we might have anticipated that the common curiosity ratio between the key phrases in Google Trends is just about the identical.

What was the end result?

For these 5 key phrases, we noticed ratio ranges of relative curiosity rating that have been various disproportionally:

Google Trends screenshot - SEJ

Google Trends’ common curiosity scores:

Table with keyword data with Google Trends data values - SEJ

The ratio of three out of 5 corresponds to the ratio of Search Console impressions.



But the remaining two key phrases differ considerably, with a mean deviation in percentages of 31,57%.

chart that visualises a plot of data points with large variance and large average deviation - SEJ

Again, with a pattern measurement of 5, the common deviation output just isn’t important.

But by repeatedly testing the setup for various key phrase units, we noticed the same sample.

Two different examples of Google Trends’ relative curiosity scores for five-keyword units:

Chart that visualizes plots of average deviation scores - SEJ

To put in perspective, the ratio common deviation percentages of Search Console are respectively 2.73% and 1.62%.



Google Trends’ common deviations thus present considerably bigger percentages than these for Search Console’s impression worth ratios.

Can we then draw any conclusions right here?

As outlined at first, Search Console isn’t all the time exhibiting essentially the most correct illustration of actuality.

However, the designed setup to check Google Trends’ knowledge accuracy and reliability supplied proof suggesting that knowledge from Google Trends isn’t constant or unambiguous both.

Does this imply that we are able to now not use these instruments? Or maybe just some?

Not essentially.

But, it doesn’t damage to pay attention to the demerits from key phrase instruments.

4. Role as an SEO Consultant

The main objective of this examine is to develop consciousness concerning the complexity surrounding the info values of key phrase instruments.

The subsequent step after consciousness is to include vital considering permitting us to acknowledge any defective habits.

Common pitfalls to keep away from:

  • Taking quantity knowledge values as granted.
  • Merging key phrase volumes from a number of instruments with out additional checks.
  • Skipping the spellings verify.
  • Ignoring the grouping impact or not validating groupings.
  • Inferring exhausting conclusions out of your key phrase quantity knowledge calculations.
  • Not offering a reliability clause to your findings within the communication to your buyer or prospects.



We can not afford to take knowledge from key phrase instruments with no consideration.

To construct experience and supply stable, dependable recommendation, we must set requirements for the way we work with key phrase knowledge.

How Will This Impact Your Role as an SEO Consultant?

I’d argue that it begins with establishing a larger sense of accountability.

Remember the sooner instance of overlooking a minor element?

Imagine that occurs.

You give this killer EnergyPoint presentation. The prospects on the desk are fully baffled by your story; you simply landed a brand new shopper!

A couple of months cross by, and also you uncover that the full quantity quantity of your key phrase knowledge set is barely 60% of the full quantity you initially communicated to your shopper.

Assuming your evaluation included just about all current key phrases related to the enterprise, such a mistake is troublesome, even perhaps inconceivable to rectify.

Especially in case your shopper’s case is restricted to a distinct segment or product cluster, you merely received’t discover different related key phrases to shut the quantity hole.



To keep away from such disasters we advocate to include the following worthwhile practices:

  • Spend extra time in your knowledge evaluation. A appropriately carried out key phrase evaluation takes time. Quality ≠ amount.
  • Validate your key phrase knowledge values.
  • Double-check your knowledge for irregularities
  • Have your ‘facts’ straight.
  • Do you need to make a presentation and draw conclusions? Make positive you a minimum of have a correct contextual story able to assist your claims.

Your boss or shopper won’t perceive why particular person efforts through the evaluation need to take a considerable period of time.

Be open and clear to shoppers and prospects concerning the required efforts to make sure the continual supply of high quality. It creates belief and fosters mutual bonding.

Telling your shopper beforehand is thus indisputably higher than explaining your mistake afterward.

That will irreversibly compromise the connection together with your shopper.

Final Notes

  • This examine’s purpose is to not place key phrase instruments in a nasty gentle.
  • Neither do I argue that key phrase instruments are in any means poor. The cause I’ve supplied the examples is only to evoke a way of consciousness surrounding the accuracy and reliability of key phrase knowledge.
  • This examine didn’t embrace different serps resembling Bing, Yandex, and Yahoo.




The examine’s setup was as follows.

We chosen a set of 160 key phrases from varied Google Search Console accounts. The choice of key phrases trusted whether or not all the following circumstances have been happy:

  • The key phrase should have had a prime 3 SERP rating for 12 consecutive months with none short-term dips reaching decrease rankings.
    • This facilitates an as shut as doable approximation of the true common month-to-month impressions rely, primarily based on a 12 month interval.
  • The key phrase’s month-to-month impression rely is 1000 or greater.
    • This will increase the chance that every collaborating key phrase instrument’s database accommodates knowledge on the chosen key phrase (long-tail key phrases are much less prone to be registered in key phrase instrument databases).
  • The key phrase shouldn’t be topic to seasonality.
    • It will increase the chance of constant prime SERP rankings all year long.
  • We additionally made positive that the 12 month interval of GSC knowledge matched the 12-month interval with which key phrase instruments calculate their month-to-month averages.

These standards have been set to be able to set up an correct recording of calculated month-to-month impression values.

Most key phrase instruments calculate their month-to-month common volumes in the same vein.

More Resources:

Image Credits



Featured Image: Created by writer, June 2020
Infographic: Created by writer, May, 2020
Screenshots taken by writer, April & May, 2020

Source link