{"id":13076,"date":"2025-10-13T17:52:57","date_gmt":"2025-10-13T15:52:57","guid":{"rendered":"https:\/\/svaerm.com\/projects\/wdf-idf\/"},"modified":"2026-03-11T10:49:41","modified_gmt":"2026-03-11T09:49:41","slug":"wdf-idf","status":"publish","type":"post","link":"https:\/\/svaerm.com\/en\/blog\/wdf-idf\/","title":{"rendered":"Why We Advise Against SEO Optimization via WDF*IDF"},"content":{"rendered":"\n

WDF*IDF is a method popular among online marketers and copywriters in Germany for optimizing on-page SEO text. It\u2019s a concept that\u2019s been around for more than a decade and revolves around the frequency of (key)words on web pages.<\/p>\n\n\n\n

As an online marketing agency, we have our own method for creating SEO text that does not rely on WDF*IDF analysis.<\/p>\n\n\n\n

We acknowledge when other marketing professionals succeed with WDF*IDF. However, in this article we briefly outline why WDF*IDF doesn\u2019t convince us on a professional level.<\/p>\n\n\n\n

What Is WDF*IDF?<\/h2>\n\n\n\n

WDF*IDF is a mathematical approach that claims to determine an optimal frequency for all important keywords in an SEO article. This frequency is usually given as an absolute count, or in some variants as a corridor (range).<\/p>\n\n\n\n

\"\"
WDF*IDF: Does not convince us on a professional level<\/figcaption><\/figure>\n\n\n\n

Example<\/h3>\n\n\n\n

To make it more tangible, below is a fictional numerical example for a 2,500-word SEO article on the topic \u201cWordPress agency<\/a>\u201d:<\/p>\n\n\n\n

Keyword<\/strong><\/td>Frequency (absolute)<\/strong><\/td>Frequency (as corridor)<\/strong><\/td><\/tr>
WordPress<\/td>60x<\/td>45x\u201375x<\/td><\/tr>
WordPress agency<\/td>12x<\/td>9x\u201315x<\/td><\/tr>
Development<\/td>8x<\/td>6x\u201310x<\/td><\/tr>
Design<\/td>7x<\/td>6x\u20139x<\/td><\/tr>
\u2026<\/td>\u2026<\/td>\u2026<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

An online marketer or copywriter would use a list like this to write the article, watching during drafting to include the keyword \u201cWordPress<\/a>\u201d 60 times, \u201cWordPress agency<\/a>\u201d 12 times, and so on. Lists of this kind are usually fairly long and therefore include a lot of individual terms.<\/p>\n\n\n\n

If corridors are used instead of absolute counts, the article would pass the WDF*IDF SEO check as long as the number of mentions falls within the specified ranges, and could then be published.<\/p>\n\n\n\n

Where Do the Numbers Come From, and How Does WDF*IDF Work?<\/h2>\n\n\n\n

WDF stands for \u201cWithin Document Frequency.\u201d It means: How often does a keyword occur in relation to all keywords in the document?<\/p>\n\n\n\n

This is its mathematical formula:<\/p>\n\n\n\n

\"\"
Formula to calculate WDF<\/figcaption><\/figure>\n\n\n\n

i: Word
j: Document
L: Total number of words in document j
Freq(i,j): Frequency of word i in document j<\/p>\n\n\n\n

IDF stands for \u201cInverse Document Frequency\u201d and acts as a corrective:<\/p>\n\n\n\n

\"\"
Formula to calculate IDF<\/figcaption><\/figure>\n\n\n\n

ND<\/sub>: Number of documents
ft<\/sub>: Number of documents that contain term t<\/p>\n\n\n\n

Words that appear in very many texts (e.g., \u201cand,\u201d \u201cthe\u201d) count less; rare, topic-specific terms count more. This highlights the words that truly differentiate a text rather than mere frequency. In combination with WDF, this prevents generic terms from distorting the evaluation and helps prioritize relevant keywords.<\/p>\n\n\n\n

WDF and IDF are determined and multiplied to compute a term score\/relevance value, from which the recommended word frequency is then derived.<\/p>\n\n\n\n

When WDF*IDF became popular in German online marketing in the early 2010s, some users calculated keyword frequencies by hand. The inputs used were competitor texts\u2014that is, texts already ranking on Google for the intended focus keyword<\/strong> (link coming soon).<\/p>\n\n\n\n

Today, there are analysis tools that automate or further develop this calculation. One example of a tool that uses an evolved form of WDF*IDF is Surfer<\/a>.<\/p>\n\n\n\n

Surfer even specifies frequency corridors for special placements like headings and bold text, indicating how often certain keywords should appear there.<\/p>\n\n\n\n

Our Results with WDF*IDF<\/h2>\n\n\n\n

We have (re)optimized various web pages using WDF*IDF and\/or Surfer.<\/p>\n\n\n\n

Our biggest success: For several weeks, we managed to rank our page \u201csvaerm: WordPress agency<\/a>\u201d on the first page of Google across Germany for the keyword \u201cwordpress agency.\u201d Previously, this page was only visible for search queries from the Frankfurt area.<\/p>\n\n\n\n

As a result, the number of project inquiries from prospects looking for a WordPress agency measurably increased.<\/p>\n\n\n\n

Why We Still Advise Against WDF*IDF<\/h2>\n\n\n\n

(Re)optimizing pages like \u201csvaerm: WordPress agency<\/a>\u201d with WDF*IDF took an entire workweek.<\/p>\n\n\n\n

The challenge is to make the text sound natural and credibly signal E-E-A-T<\/strong> (Experience, Expertise, Authoritativeness, and Trustworthiness; link coming soon), even though there is a precise mathematical specification for how often which words should occur.<\/p>\n\n\n\n

That requires a high level of copywriting skill, because just ranking on Google isn\u2019t enough\u2014web pages also need to \u201cconvert,\u201d i.e. in B2B, generate business inquiries. For a junior copywriter, this is nearly unattainable under such strict constraints. In our experience, even with carefully engineered prompts, AI is not yet at a point<\/strong> (link coming soon) where it can automate or reliably support this process without making significant mistakes or going off topic. So, a senior SEO writer has to handle it, with a considerable investment of resources.<\/p>\n\n\n\n

\"\"
Texts optimized with WDF*IDF by junior marketers or AI usually do not generate many conversions<\/figcaption><\/figure>\n\n\n\n

Unfortunately, the success tends to be short-lived: As mentioned, WDF*IDF analysis is based on competitor texts. That\u2019s problematic because Google constantly ranks new pages in its results. Especially in competitive niches, the search results can change by the second.<\/p>\n\n\n\n

\n
\"\"<\/figure>\n<\/div>\n\n\n

The recommendations from a WDF*IDF analysis are therefore always just a snapshot and incomplete. While ranking nationwide for a keyword like \u201cwordpress agency<\/a>\u201d is a good achievement, it\u2019s not economical to invest an entire workweek of a senior SEO copywriter every month to maintain it.<\/p>\n<\/div><\/div>\n\n\n\n

WDF*IDF is probably only worthwhile in particularly static niches with little competition in SEO content marketing\u2014in other words, \u201cblue ocean<\/a>\u201d<\/strong>. But for such keywords, you usually don\u2019t need a WDF*IDF analysis in the first place because the competition is so low.<\/p>\n\n\n\n

What You Can Do Instead<\/h2>\n\n\n\n

We tailor our recommendations based on your company size:<\/p>\n\n\n\n