The short answer: track one to three keywords per landing page, not a magic number
The honest answer to how many keywords to track is a formula, not a single figure. Track one to three primary keywords for every page you actually want to rank, plus a layer of "striking distance" terms sitting on page two, plus the handful of head terms where competitors are beating you. For a 40-page site that usually lands somewhere between 90 and 170 tracked keywords. You'll see a round "22 keywords on average" quoted across blog posts, but that kind of average is pulled from people on per-keyword pricing, so it tells you what they could afford to track, not what was worth tracking.
That distinction matters more than any benchmark. Most rank trackers charge per keyword, so the tool's pricing quietly trains you to track too few of the wrong ones. But the cost of tracking a keyword and the value of tracking it are unrelated. The keywords worth checking are the ones tied to revenue and the ones where you sit on page two and can still move. A head term you'll never rank for, or a keyword already locked at position one, is mostly cost with little signal.
This is a spoke of our guide to keyword rank tracking. Here we settle the sizing question and give you a method to pick the right portfolio.
Why "22 keywords" is the wrong target
There are two failure modes, and a flat benchmark hides both.
Track too few, and you go blind on the terms that actually pay. People with a tight tracking budget tend to monitor their three or four obvious head terms ("project management software") and miss the long-tail variations and intent-specific queries that convert better and rank faster. You optimize what you can see, so a small list quietly shrinks your SEO program.
Track too many of the wrong ones, and your reports turn into noise. Two thousand keywords, where half never had a chance and a quarter are pinned at number one, means every weekly report is a wall of green and grey you stop reading. The problem is not the count. It is that nobody tiered the list, so movement that matters drowns in movement that does not.
The right number falls out of a better question: which keywords, if they moved three positions tomorrow, would change a decision you make? Track those. The rest is vanity.
The four tiers worth tracking
Build your portfolio from four buckets. Each earns its place for a different reason.
1. Revenue keywords. The terms tied directly to a page that makes you money: a pricing page, a high-intent product or service page, a comparison page. These you watch closely regardless of position, because a one-spot slip here costs real revenue. If "invoicing software for freelancers" sends you trials, it is a revenue keyword even at position two.
2. Striking-distance keywords. Queries where you currently rank roughly positions 8 to 20, the bottom of page one and page two. These are the highest-leverage terms you own, because the click-through gap between page two and the top of page one is enormous. Backlinko's analysis of more than four million Google search results found the number-one organic result earns a 27.6% click-through rate, while the entire second page of results collects just 0.63% of clicks (Backlinko, Google CTR study). A keyword stuck at position 11 is a few internal links and a content refresh away from real traffic. You find these in your Search Console performance report by filtering for queries with impressions but a low average position. Tier two is where tracking pays for itself.
3. Defensive number-one keywords. Terms you already rank first for and would lose money if you slipped. You do not need to watch these daily, but you need an alert. Tracking them at a lower cadence catches the day a competitor's refresh or a new SERP feature knocks you off the top.
4. Competitor-gap keywords. Terms a direct competitor ranks well for and you do not rank for at all. These define your roadmap. You are tracking their position more than your own, which feeds into share of voice tracking, the metric that tells you how much of a keyword set's total visibility you own versus your rivals.
| Tier | What it is | Cadence | Why track it |
|---|---|---|---|
| Revenue | Money-page keywords | Daily | A slip costs revenue now |
| Striking-distance | Positions 8–20 | Daily / every other day | Highest upside per effort |
| Defensive #1 | Locked top spots | Weekly + alert | Catch losses early |
| Competitor-gap | Rival ranks, you don't | Weekly | Roadmap and share of voice |
A sizing formula based on landing pages
Stop guessing from gut feel and size from structure. Google's SEO Starter Guide tells you to write naturally around the words a user might search for and to anticipate real queries, rather than repeat one term over and over (Google Search Central SEO Starter Guide). Ahrefs makes the same point from the keyword-research side: closely related queries that return near-identical results can be served by one page, so you cluster them instead of building a page each (Ahrefs, keyword research guide). That gives you a clean unit to count: the page.
A workable formula:
- 1 to 3 keywords per indexable landing page you care about ranking (the primary term plus one or two close variants that share the same intent).
- + your striking-distance set from Search Console, typically 20 to 60 queries for a mid-size site.
- + 10 to 30 competitor-gap terms that define near-term content work.
So for a 40-page site: roughly 40 to 120 primary keywords, plus 30 striking-distance, plus 20 gap terms, call it 90 to 170. That is your starting portfolio, and it is defensible because every entry maps to a page or a decision. Review it quarterly: promote striking-distance winners to defensive, retire terms that never moved, add new pages' keywords as you publish.
A few practical multipliers. If your audience searches across regions or languages, the same keyword is really several tracked rows once you add rank tracking by location and account for mobile versus desktop rankings, which diverge more than most people expect. A local business chasing the map pack adds another layer through local pack tracking. City-level and vernacular SERPs (Hindi, Tamil, Marathi and dozens more) multiply the count further, which is exactly why per-keyword pricing punishes location-heavy and multilingual strategies hardest.
When tracking cost shapes the wrong strategy
Here is the part the benchmarks never mention. When a rank tracker bills you per keyword, you start making editorial decisions to protect the bill. You drop the striking-distance terms because there are "too many." You skip the location and language variants. You track your four head terms and call it a portfolio. The pricing model has quietly written your SEO strategy, and it wrote a bad one.
Keywords should not be the thing you ration. DeployFlare's rank tracking is built so you can track generously: the whole striking-distance set, every location and language variant, the competitor-gap list, without watching a meter. Plans start at ₹499/month, billed in INR with UPI and GST, and the per-keyword anxiety that shrinks everyone else's portfolio is simply not part of the model. You can start with the free rank checker to test a handful of terms before committing, and the same generosity is why teams switching from a SEMrush alternative usually end up tracking two to three times more keywords for less.
The goal is not to track the maximum. It is to track the right tiers at the right cadence and never let the invoice decide what you can see. Once cost is off the table, the only question left is the useful one: which keywords would change a decision? Track those, and read them well. Our guide on how to read rank tracking data covers turning a tracked portfolio into action, and how to track keyword rankings walks through setting it up cleanly.