Frontlist keywords
Research and keyword recommendations for forthcoming titles, with the reader's needs and the book's position in its category in view.
Services / Amazon keywords
Keywords built around how readers find your books
Kadaxis researches the language readers use, checks how Amazon responds, and develops keywords for each title. Publishers receive practical recommendations that fit their existing metadata workflows.
The work
Backlist optimization
Review existing keywords and develop fresh sets for a backlist segment or a wider catalogue. Each book is researched individually.
Search visibility
Title-by-title reporting on whether keywords surface the intended book in Amazon results, giving your team a practical view of what is doing work.
Keywords work when they give a book a route into search
Our research found that about 8% of publisher-assigned keywords surfaced their book in the first two pages of Amazon Books results. Across thousands of books, Kadaxis-built keywords average about 40% on that visibility measure.
The difference starts with a reader's question: what does this book offer, and how would someone look for it? We use that language to build and check title-specific keyword sets.
Read the keyword visibility researchWhat your team supplies and receives
Start with your title information
We agree the title list and request the relevant ISBNs, title and author information, and existing keyword metadata. Any additional material depends on the work being commissioned.
Receive recommendations ready for review
Deliverables can include keyword sets formatted for your ONIX workflow, supporting research, and visibility reports. Your team reviews and approves the recommendations, then loads them through its existing systems.
Scope, timing, and the reporting format are agreed around your list and publishing workflow.
How we work with publishersFrom the research
“the median book ranked for zero of its assigned keywords.”
ONIX keywords work. Most keywords that publishers create don't.
“Use search phrases as the default.”
Should book keywords be single words or search phrases?
Questions publishers ask
Should keywords be single words or phrases?
In our research, reader-search phrases performed substantially better than isolated metadata words. The useful test is whether a phrase matches a reader's intent and surfaces the book, rather than whether it is a broad category label. Read the comparison.
Doesn't Amazon recombine single words?
Amazon can use words from multiple metadata fields. That does not make every combination relevant or competitive. Our tests compared the phrases readers might use with the words already supplied in publisher metadata. See the results.
What if a specific phrase has little recorded search volume?
A volume estimate is one signal. We also examine the reader need, the book's fit, and the results Amazon actually returns. A specific phrase needs a plausible discovery route; specificity alone does not make it useful. Read what makes a keyword effective.
Isn't ranking driven by sales?
Sales and other signals influence ranking, while metadata helps determine whether a book is eligible to appear for a query. Visibility reporting shows that search route; it is not a measure of sales lift. Read about relevance and ranking.
Will changing keywords hurt our ranking?
A metadata change cannot guarantee a particular rank. We review the current title information and recommend relevant terms for your team to approve. If you have concerns about a specific change, we can discuss the title and its existing search visibility.
How do you use AI, and who approves the changes?
AI supports the research and synthesis alongside market signals and editorial judgment. Kadaxis provides recommendations; your team retains approval and implementation. Read how we use AI in our research write-ups.
Discuss keyword work for your list
Tell Chris about your publisher or imprint and the titles you have in mind. We will discuss the scope, the information needed, and the right form of delivery.
Send Chris a message