Skip to main content

Introduction to search term analysis

· 5 min read
Liz Argall
Liz Argall
Technical Writer and Program Manager

If you want to jump straight straight into Search Term Analysis spreadsheet:

About​

When we talk about content quality, findability is an important factor that extends beyond clean architecture. "Can the user find the data?" is a common question. Can the user find the data if they start with a Google search takes us deeper into this investigation.

At the Write the Docs Conference in Portland this year I gave a lightning talk on "Google Magic!" It showed how and get insight into your documentation in five minutes. Through this approach you can get a clearer understanding of your users and identify the most impactful ways to improve your docs.

This article goes into a bit more detail and provides a template for conducting several kinds of search term analysis.

Why do a search term analysis?​

Users who actually provide feedback are a small percentage of our user base. They are often important users, they're the super users who care and will help others use our tools and services. Statistically, however, people who give feedback are outliers.

When you conduct a search term analysis you can think through many user personas as you come up with terms and get a suite of benefits, including:

  • Identify content gaps and pain points
  • Prioritize content based on real-world queries
  • Interact with your content in fresh new ways helps you refresh your content
  • Make some of the invisible maintenance work you do more visible
    • It's one thing to make content improvements as general maintenance, it's another thing to make content improvements with a framework!
note

There are content marketing and sales tools that can help you with some of aspects of search term analysis analysis, but they're often expensive and seldom made available to technical writers. Additionally, part of the power of this analysis is the physical process of thinking through and going through user experiences.

LLM investigation​

For this new spreadsheet, I created a tab exploring how you could analyze user experiences if they ask an LLM questions. I suggest using Perplexity as it is designed to provide references. It was interesting to think about how a robustly referenced Perplexity output differs from search engine output. LLMs can statistically aggregate relationships, they can provide interesting juxtapositions, but individual topics aren't explored with as much depth. LLMs provide an answer, as Tara McMullin would say, they prioritize efficiency and authority. LLMs will never fully replace search engines, but when used well LLMs can be a launch pad for more robustly using search engines.

How to do a search term analysis​

I've put the majority of the instructions in the spreadsheet, so that it can be the source of truth.

Step 1: Set up your spreadsheet and read the instructions​

Step 2: Decide on your approach​

Once you've read through the instructions, decide on your focus. Start with the most pressing needs and potential pain points/unknowns and then iterate. While you could do this as part of a formal audit process, I recommend starting it as a casual investigation. Try it out, explore the benefits and please yourself before getting too fancy!

I recommend starting with a quick Targeted Keyword or Google Autocomplete analysis.

Step 3: Decide when to share your findings​

Unless there's a particular pain point (why can't folks find calibration information?), it's often better to chew over it, get to know your content via search term analysis over time. You may be surprised by the content improvements and insights you get exploring content in this organic, exploratory manner.

An analysis that fits in around other things might take a month or two. An analysis that focused on some burning questions might take a day or two!

Step 4: Reflect on the process​

What worked what didn't? Set a time for when to revisit your content, twice a year can be a helpful cadence. You may wish to schedule an analysis the month before a major announcement, so your content is prepared should there be a sudden surge in interest.

If you'd like to tell me how it goes, I'd love to hear about it! I'd especially love to hear about whether the LLM section is useful. Everything else in the spreadsheet is tried and tested, but that section is a new adventure. You can email me at liz@lizargall.com or you can find me on LinkedIn.

Media I'm enjoying​

Want to read more blog posts?​

Subscribe to my newsletter to be emailed whenever I do an update!