Volume is a vanity metric
"10,000 searches a month" sounds impressive until you realize most of that volume might be navigational traffic looking for a competitor's login page, or informational queries from people who will never buy anything. Good keyword research isn't a volume hunt — it's an exercise in matching a query to a business outcome.
The right question isn't "how many people search this?" It's "if I rank #1 for this, does it move a metric I care about?"
The four intent types — and why they change everything
Every keyword carries an implicit intent. Get this classification wrong and the rest of your research is wasted.
| Intent | What the searcher wants | Example query | Right content type |
|---|---|---|---|
| Informational | To learn something | "what is keyword difficulty" | Explainer, guide |
| Commercial | To compare options before buying | "best seo tools for agencies" | Comparison, listicle |
| Transactional | To take an action now | "seobox pricing" | Pricing/product page |
| Navigational | To find a specific site or page | "seobox login" | Direct landing page |
Targeting a transactional page at an informational query (or vice versa) is one of the most common reasons content ranks but never converts — the page satisfies the algorithm's relevance check but not the human's actual need.
Difficulty is relative to your domain, not universal
A "keyword difficulty" score from any tool is an estimate of how hard it is to rank in the top results in general — based on the backlink profiles and authority of pages currently ranking. It is not a fixed number. A brand-new site chasing a keyword with a 70/100 difficulty score, dominated by ten-year-old domains with thousands of backlinks, is playing a very different game than an established site with existing topical authority chasing the same term.
Two adjustments matter more than the raw score:
- Topical relevance to your existing content. Ranking for a keyword adjacent to topics you already rank for is dramatically easier than an isolated, unrelated term — search engines weigh topical authority, not just page-level signals.
- SERP feature crowding. If the results page is dominated by an AI Overview, a video carousel, and three ads before the first organic result, the effective difficulty is much higher than the raw score suggests, regardless of backlink competition.
Finding opportunities, not just keywords
The highest-leverage keyword research doesn't start from a blank list — it starts from signals you already have:
- Content gaps. Topics your competitors rank for that you have no page targeting at all.
- Ranking gaps. Keywords where you rank on page 2–3 — close enough that a focused content or internal-linking push can realistically move you to page 1.
- Decay. Pages that used to rank and are now sliding, usually because the content went stale relative to newer competing pages.
- New keyword emergence. Rising queries in your space that don't have entrenched competition yet — the cheapest wins available, if you catch them early.
This is exactly the framework behind Seobox's opportunity engine: instead of handing you a raw keyword list, it classifies every opportunity into one of these categories against your connected Google Search Console data, so you're prioritizing by actual upside instead of alphabetical order.
A practical research workflow
- Start from a seed list — your product categories, competitor comparisons, and the questions your support team answers most often.
- Expand with related and "people also ask" queries to map the full concept space around each seed, not just the head term.
- Classify every keyword by intent before doing anything else with it.
- Score difficulty relative to your current topical authority, not the raw tool score alone.
- Cluster keywords that share intent into a single target page — trying to rank one thin page per micro-keyword variation dilutes authority and creates duplicate-intent cannibalization.
- Prioritize by business value × achievability, not by volume alone. A cluster of five commercial-intent keywords at moderate difficulty usually beats one high-volume informational keyword at extreme difficulty.
Common keyword research mistakes
- Chasing volume on informational terms with no path to conversion, at the expense of lower-volume commercial terms that actually drive revenue.
- Ignoring keyword clustering, publishing five thin pages for five near-duplicate queries instead of one comprehensive page.
- Never revisiting old research. Search behavior shifts constantly — a keyword list from eighteen months ago is missing every AI-search-era query pattern that's emerged since.
- Treating "low difficulty" as automatically "worth doing." An easy keyword with zero commercial value is still not worth a content slot.
How Seobox automates this
Seobox pulls live keyword and SERP data for every site you connect, scores difficulty against your actual domain authority (not a generic estimate), classifies intent automatically, and surfaces content-gap, decay, and ranking-gap opportunities directly from your Search Console data — so the research step that used to take hours per topic runs continuously in the background.
FAQ
How many keywords should one page target? One primary keyword plus its close intent-matched variants (a cluster), not a scattered list of unrelated terms. If two keywords require genuinely different content to satisfy, they belong on separate pages.
Should I ever target a keyword with "zero" reported search volume? Yes, cautiously. Long-tail and emerging queries frequently show as zero or "not enough data" in tools while still driving meaningful traffic in aggregate — especially in fast-moving categories like AI search itself.
How often should I refresh keyword research? Quarterly at minimum for competitive categories, and immediately after a major product or market shift. Search behavior around AI tools in particular is still moving fast enough that stale research ages out within months, not years.