Keyword Research

Keyword research is the practice of identifying the terms people use when searching for products, services, and information, and assessing which of those terms are worth pursuing. It underpins both organic and paid search work, and its output shapes site structure, content planning, page copy, and campaign targeting. Done well, it replaces internal assumptions about vocabulary with evidence about how customers actually describe their problems.

The core discovery problem is that organizations name things differently from their customers. Internal product names, category labels inherited from an operations system, and industry vocabulary all feel natural inside a business and are invisible in search demand. Research routinely reveals that the term generating the most volume is one nobody in the company uses, and that the phrase featured prominently across the website has essentially no search volume at all. Correcting that mismatch is often the single highest-return output of the exercise.

Three attributes determine whether a term is worth pursuing. Volume indicates how many people search it, though reported figures are estimates that group similar variants and should be read as relative rather than absolute. Difficulty estimates how hard ranking would be given who currently occupies the results. Intent describes what the searcher wants, and it matters more than either of the others: a high-volume informational query may generate traffic that never converts, while a lower-volume query with clear purchase intent produces revenue.

Intent is best assessed by examining the results the search engine currently returns rather than by inspecting the words alone. If a query returns product listing pages, the engine has determined the intent is transactional, and an article will not rank however well written. If it returns guides and comparisons, a product page will struggle. This examination also reveals the format that succeeds for a term, including whether results are dominated by video, by discussion threads, or by a featured snippet that absorbs most of the clicks before any site is visited.

Research output should be organized into topics rather than treated as a list of individual terms. Search engines interpret meaning rather than matching strings, so a single well-built page typically ranks for many related queries, and building separate pages for close variants creates internal competition and thin content. Grouping terms by the underlying question they express, then mapping one page to each group, produces a structure that both search engines and visitors can navigate.

Research also has to account for how much of the demand a business can realistically capture, which is where many content plans overreach. Queries dominated by large publishers, marketplaces, or comparison sites with years of accumulated authority are not winnable in a reasonable timeframe for most businesses, however well the content is written. Assessing the competitive composition of a results page, and specifically whether the ranking sites are peers or institutions, is what separates a plan that produces traffic within a year from one that produces frustration. Where the head terms are genuinely closed, the productive route is depth in adjacent, more specific territory rather than repeated attempts at the obvious target.

The commercial framing is what turns a keyword list into a plan. Terms should be prioritized by expected business value rather than by volume, which means weighting them by intent, by how well the business can serve that intent, and by realistic prospects of ranking given current authority. In a marketing services engagement this prioritization drives both the content roadmap and paid search targeting, and the resulting vocabulary is equally valuable to product design work, since navigation labels and category names taken from real search language consistently outperform terminology invented internally.