Is Keyword Research Part of On-Page SEO? Where It Really Fits

The short version: No, keyword research isn't technically one of the on-page SEO elements (title tags, headers, internal links, image alt text), but it's the input every one of those elements depends on. Do on-page SEO without keyword research first and you're polishing a page that was never aimed at anything. I've watched agencies bill the two as separate line items for years and it still confuses clients, so let's sort out where the line sits.

The short answer, then the nuance

On-page SEO, strictly defined, is what you do to the page itself: title tags, meta descriptions, H1s and H2s, URL structure, internal linking, image alt text, content structure, page speed touches you control from the CMS. Keyword research is what you do before any of that, to figure out which words and phrases you're optimising those elements for.

So no, keyword research isn't a checklist item inside on-page SEO. It's the research phase that on-page SEO acts on. Most SEO checklists lump them together because in practice nobody writes a title tag without knowing the target term first, so the two feel fused even though they're structurally separate jobs.

Here's the bit that trips people up: if you get the keyword research wrong, no amount of on-page perfection saves you. I've seen pages with flawless title tags, clean H1 hierarchy, internal links pointing every which way, that ranked for nothing because the keyword behind all that structure had almost no search volume and no commercial relevance. The on-page work was technically correct. It was also pointless.

What on-page SEO covers, item by item

To draw the line, it helps to list what's inside on-page SEO's remit:

• Title tags (ideally under 60 characters, with the primary keyword near the front)
• Meta descriptions (150 to 160 characters, written to earn the click, not just describe the page)
• H1 and subheading structure (one H1, logical H2/H3 nesting)
• URL slug (short, readable, keyword included)
• Body content itself: how the topic is covered, depth, structure, answer-first paragraphs
• Internal linking (which pages link to this one, and what anchor text they use)
• Image alt text and file names
• Schema markup where relevant
• Page speed and Core Web Vitals, to the extent you control them at the page level

Every single one of those items needs a keyword decision made before you touch it. You cannot write a title tag without knowing the target phrase. You cannot decide internal link anchor text without knowing which term you're trying to rank for on the destination page. That's why keyword research feels like part of on-page SEO even though, structurally, it's the step before it.

Where on-page SEO keyword research fits in the sequence

If you're building a content plan or auditing an old page, the order is:

1. Keyword research: find the term, check the volume, check the intent behind it
2. Content planning: decide what the page needs to say to satisfy that intent
3. On-page SEO: write the title, structure the headers, build the internal links, all around that keyword
4. Technical checks: speed, mobile rendering, indexability

Skip step one and you're doing step three blind. I have a full breakdown of what keyword research and analysis involve step by step if you want the mechanics of that first stage laid out, because it's more than typing a phrase into a tool and eyeballing the volume number.

A real example from a client audit

Two years ago I audited a site for a small accountancy firm in Kent. They'd had a content agency write 40 blog posts over 18 months, every one of them "on-page optimised" according to the agency's own report: title tags present, meta descriptions written, headers structured, internal links in place. Textbook on-page SEO by the checklist definition.

Traffic across all 40 posts combined was under 200 organic visits a month. I pulled the keyword each post had been built around and ran them through Ahrefs. Of the 40 target keywords, 31 had a monthly search volume of zero to 10 in the UK. The agency had picked keywords by brainstorming what the client "should" rank for, not by checking whether anyone searched for it.

We rewrote nine of those posts around actual search terms (things like "self assessment deadline UK" and "how much does an accountant cost small business", both with genuine monthly volume in the hundreds) and kept the on-page structure almost identical, same header logic, same internal linking approach. Within four months those nine posts were pulling more traffic than the original 40 combined. The on-page SEO hadn't changed much. The keyword underneath it had.

That's the uncomfortable part nobody likes to say out loud: an agency can do everything on their on-page checklist correctly and still deliver nothing, because the checklist never checked whether the keyword was worth targeting in the first place. On-page SEO can be executed flawlessly around a completely wrong choice. It happens more than the industry admits, because "we optimised the on-page elements" sounds like progress on a report even when it isn't.

How often you should be checking the numbers

One thing that came out of that Kent audit: the agency had checked search volume once, at the very start of the contract, and then never again for 18 months. Search behaviour shifts. New competitors enter. What had ten searches a month in year one can have a hundred two years later, or the reverse. I wrote a longer piece on how many times you should check keyword search volume before writing, and the short version is: check it before you commit to the topic, check it again right before you publish if more than a few weeks have passed, and check it again at the six-month mark to see if it's still worth the page real estate.

Does the tool you use for keyword research matter?

People obsess over which tool gives the "true" search volume, Ahrefs versus Semrush versus Google's own Keyword Planner, and the honest reality is they all disagree with each other by a wide margin and none of them is showing you real-time reality, they're all modelled estimates. I've gone into why that's the wrong question to be asking about keyword research tools in more depth, but for the purposes of this post: whichever tool you pick, use it consistently, trust the relative comparison between keywords more than the absolute number, and move on. Spending three hours arguing about whether a keyword has 320 or 480 monthly searches is time you could spend writing the page.

Why the on-page versus keyword-research distinction matters for how you plan your time

This isn't just semantics for a blog post title. It matters for how you budget your time or your team's time. If you treat keyword research as "part of" on-page SEO, you'll rush it, because on-page work feels like the tangible, checkable task (did we write the meta description, yes or no) and keyword research feels like the soft, upstream bit you can skim.

In practice, keyword research should take longer than most people give it. For a single important commercial page, I'll spend 45 minutes to an hour checking volume, intent, competitor pages ranking for it, and related terms I might be able to fold in. The on-page write-up for that same page, title tag, meta, header structure, might take 15 minutes once the keyword decision is made. The ratio is backwards from how most content teams allocate their time.

This matters just as much outside blog posts. If you're building out a YouTube video keyword research process, the on-page equivalent (title, description, tags) is trivial to write once you know the search term people type into YouTube search or Google. The research is the hard, valuable part. The on-page fields are just where you put the answer.

A quick way to check if your on-page SEO is built on solid keyword research

If you want to audit your own site fast:

• Pull a list of your top 20 pages by traffic potential (or lowest performing pages you've invested in)
• For each one, note the exact keyword the title tag was written around
• Run each keyword through a tool and check current monthly search volume
• Flag anything under 20 monthly searches (in most markets that's barely worth a dedicated page unless it's high commercial intent)
• For flagged pages, decide: merge it into a stronger page, rewrite it around a better keyword, or leave it if it's serving a non-search purpose

That's a half-day job for a site with 50 pages. It's the single highest-use audit I run for new clients, because it tells you within an afternoon whether your on-page work has been aimed at anything real.

If you're choosing, see my roundup of Google keyword planner alternatives.

Where this fits into the bigger picture

If you're new to this whole area and want the fuller context of what keyword research is and why it matters in digital marketing more broadly, that's worth reading before you touch on-page work at all. And if you're a blogger rather than running a commercial site, the stakes are slightly different but the sequence is identical, which is why I wrote separately about why keyword research matters for bloggers trying to grow traffic.

If you're weighing up whether to bring in outside help for this rather than muddle through it yourself, it's worth knowing what that costs before you commit, because the range between a freelancer and an agency retainer is wide and not always tied to quality: I've laid out real numbers in my guide on what an AI consultant costs for small business SEO and marketing work.

If that's your next decision, start with Keywords Everywhere alternatives.

The one-line answer to keep in your back pocket

Keyword research is not technically an on-page SEO element. It's the decision that every on-page element then has to serve. Get the decision wrong and the elements are just well-formatted noise.

If you write about research or data, see my live-checked list of science blogs that accept guest posts.

I've also written about updating your SEO without a developer.

Frequently asked questions

Is keyword research the same thing as on-page SEO?

No. On-page SEO refers to the elements you control directly on a page, title tags, headers, internal links, alt text. Keyword research is the research step that happens before those elements are written, and it determines what they should say.

Do SEO checklists usually include keyword research under on-page SEO?

Many checklists list it as step one of an on-page process because in practice you always need a keyword before you write a title tag. Structurally though, it's a separate discipline that feeds into on-page work rather than being one of its elements.

Can you do on-page SEO without keyword research?

You can, but it usually means guessing at what to optimise for, and guessed keywords frequently have low or no search volume, meaning perfectly executed on-page work ends up ranking for terms nobody searches for.

How much time should I spend on keyword research versus on-page SEO for one page?

For an important commercial page, budget 45 minutes to an hour for proper keyword research (checking volume, intent, and competing pages) versus roughly 15 minutes for the actual on-page write-up once the keyword decision is made.

Official documentation

• Google Search Central documentation
• Google's spam policies

Worth a read if this is your situation: our guest posting guidelines.