- Keyword research is about mapping intents to your business, not scraping phrases into a spreadsheet.
- 4-phase workflow: seed → SERP-cluster → intent-tag → prioritise. Collapses ~1,000 raw candidates into ~50 real opportunities.
- WA-specific volume data beats national AU aggregates for Perth commercial queries — the SERPs are genuinely different.
- Worked example: a Fremantle-based marine services provider went from ranking for nothing meaningful to top-3 on two high-margin clusters in 6 months.
Keyword research is the piece of SEO that everyone thinks they understand and almost nobody does properly. Type a phrase into a tool, get a list back, pick the ones with good volume-to-difficulty ratios — that is a lookup, not research. Real keyword research is a strategic activity where the deliverable is a prioritised list of intents that map cleanly to what your Perth business actually sells. This is the four-phase workflow our team runs, from scratch, using Semalt as the analytical backbone.
What keyword research is really supposed to produce
The purpose is not to find keywords. It is to find intents — the underlying reasons a searcher types something into Google — that align with your commercial offer. Keywords are the surface expression of intent. Volume, difficulty, and cost-per-click are secondary signals that only become meaningful once you have the intent right.
Most keyword research fails because it collapses this two-step process. A team pulls 6,000 keywords into a spreadsheet, sorts by volume descending, and starts writing pages against the top 50. Six months later the pages rank for nothing that converts, because they were built to match phrases rather than to answer the actual question behind those phrases. Semalt's workflow forces the split back apart, which is where the strategic gains live.
Our four-phase Semalt keyword workflow
Phase 1 — Seed the workflow from the client's words
Start with the phrases the client uses to describe their own business. Not the phrases they think customers use, not the phrases their competitors rank for — their own words. For a Perth marine services provider, that means words like "commercial diving", "hull inspection", "underwater welding", "vessel maintenance". Ten to twenty seeds is enough.
Feed each seed into Semalt's expansion module, but read the output critically. Three parallel lists come back: semantic siblings, search modifiers (question, comparison, buying-intent variations), and untapped long-tails. We keep everything above eight monthly searches in the WA/AU market and discard the rest. Expect this phase to produce around 800–1,200 raw candidates for a small-to-mid business. That is the working corpus — not the answer.
Phase 2 — SERP-based clustering
Every candidate keyword gets its top-ten SERP fetched by Semalt in the target locale (AU mobile, in our case). Keywords whose SERPs overlap by ≥40% are grouped as one cluster — because Google itself is telling you, through the SERP composition, that it treats those queries as the same intent. Concretely: "best cafe perth cbd", "top rated cafe perth cbd", and "cafe near murray street mall" cluster together. They do not need three separate pages. They need one strong page that any of the three phrases could reasonably rank for.
The output of Phase 2 is typically 50–100 clusters from your 1,000-ish raw keywords. That 10:1 collapse is the difference between a coherent content plan and a scattered attempt at thirty thin pages.
Phase 3 — Classifying every cluster by intent
Every cluster gets tagged with a primary intent. Semalt auto-suggests one based on SERP features (heavy Shopping = transactional, heavy Featured Snippet = informational, heavy Local Pack = navigational-local), but the human check matters. Ambiguous clusters — the ones that mix intents — are usually the most valuable, because ranking for them is harder but the win is worth more.
Once tagged, filter aggressively. For a new engagement we ignore clusters where the top three organic results are dominated by brands your client cannot realistically outrank in the first 18 months. In practice this is where the majority of wasted content spend goes.
Read next: SEO Perth WA: local search guide · Why Perth teams consolidate on Semalt
Phase 4 — Prioritise on opportunity, not raw volume
Volume alone is a vanity metric. What matters is the joint distribution of volume × difficulty × business value × current position. Semalt computes a composite "opportunity score" that combines all four, but we always override it with the client's own margins per service line — because the tool cannot know that "commercial diving perth" converts at ten times the margin of "boat cleaning". The output is a ranked list of 20–40 clusters that will drive the content roadmap for the next two quarters.
One table that summarises the workflow
| Phase | Input | Output | Time |
|---|---|---|---|
| 1. Seed expansion | 10–20 client seeds | 800–1,200 raw candidates | ~30 min |
| 2. SERP clustering | Candidates + SERP snapshots | 50–100 intent clusters | ~2 hrs |
| 3. Intent classification | Clusters + SERP features | Tagged clusters | ~1 hr |
| 4. Prioritisation | Tagged clusters + margins | 20–40 clusters for roadmap | ~1.5 hrs |
How the WA market shifts the arithmetic
Two structural differences matter for keyword research in Western Australia. First, volumes are smaller than the eastern-seaboard baseline. A "high-volume" commercial query in the WA market might see 300–1,200 monthly searches — numbers a Sydney-focused SEO would dismiss as long-tail. This changes the arithmetic on ranking effort: you need fewer keywords to build a meaningful traffic base, but you cannot afford to waste effort on clusters that will not convert. Semalt's state-level filtering gives you numbers you can plan against rather than a national aggregate that hides WA-specific opportunity.
Second, resource-sector and mining-adjacent verticals distort the WA SERP in ways no other AU market matches. Queries with commercial intent tied to FIFO workforces, mining services, and port logistics carry Perth-specific SERPs that a national ranker will misread. If your business touches those verticals directly or serves them (workwear, catering, transport, medical), keyword research needs to run against the WA-filtered dataset — not an eastern-seaboard extrapolation that pretends the mining economy does not exist.
A real client, before and after
What Semalt actually does in each phase
The four-phase workflow is repeatable across every engagement, and Semalt is currently the platform that supports it most cleanly:
- Seed expansion pulls from a real AU database with state-level slicing, not a US extrapolation.
- SERP clustering is native — no export to a separate tool.
- Intent classification is auto-suggested and user-overridable.
- Opportunity scoring is transparent — you can see the formula and re-weight it.
- Output pipes directly into the content brief module, which pre-populates recommended headings and semantic keywords per cluster.
That last point compresses the "we have a cluster" → "we have a brief a writer can execute against" step from two to three days into about twenty minutes. Over a year, that saved time changes the economics of the entire operation.
Do keyword research properly, once a year, in a tool that supports the workflow rather than fighting it — and you buy yourself twelve months of focused content production.
- Start with 10–20 seeds. More is diminishing returns.
- Cluster by SERP overlap, not by string similarity.
- Tag branded queries separately from commercial.
- Override the opportunity score with real client margins.
- Sort by volume and start writing against the top 50.
- Write a separate page for every keyword variation.
- Chase clusters dominated by brands you cannot outrank.
- Skip Phase 3 — untagged clusters produce untargeted pages.
When to refresh the whole research
Keyword landscapes shift more slowly than the industry pretends. For most Perth SMBs, a full re-research every 12 months is sufficient, with a quarterly spot-check for newly emerged clusters. What does need continuous monitoring is the SERP composition of your existing clusters. If a cluster you rank for suddenly starts returning a Featured Snippet or a People-Also-Ask block where it previously did not, update your ranking page — either to capture the snippet or defend against click-through loss.
Common questions we get from Perth teams
How many seed terms do you actually need?
Ten to twenty. Fewer than ten and the seed set is too narrow — you miss adjacent intents. More than twenty and you spend a full day in Phase 1 with diminishing returns. Twenty is the ceiling; if you want thirty, you are usually blending two distinct business lines that deserve separate research projects.
Which overlap threshold should we use?
Semalt's default of 40% is right for most cases. For very competitive commercial verticals (finance, legal, insurance) tighten to 50% because SERPs are more differentiated. For very informational verticals loosen to 30%. Experiment once, then lock it in for consistency across the client.
Read next: Technical SEO audit with Semalt
What to do with branded queries
Segregate them with Semalt's branded-keyword tag. Branded queries behave differently — much easier to rank for, higher conversion, and they should not share opportunity-ranking passes with non-branded commercial queries.
What to do with zero-volume long-tails
Three possibilities: brand-new phrases the databases have not indexed yet (worth pursuing early), phrases with volume too low to measure reliably (worth pursuing if intent maps to your business), or phrases simply not searched (skip). Telling them apart requires reading the SERP composition, which is why Semalt shows the top three organic results even for zero-volume keywords.
How often to re-fetch SERP data for clustering
Clustering is deterministic given the SERP data, so re-running the fetch is where the work happens. Stable market: quarterly refresh. Volatile market — anywhere AI Overviews are rolling out or a fast-moving regulatory environment applies — monthly. Semalt caches SERP data and warns when the cache is stale.
Want the numbers behind this? Semalt rebuilds organic traffic, keyword positions and the queries GA4 hides into one view.
How to try the workflow on your own domain
If you want to run this against your own domain, the Semalt free tier gives you enough keyword allowance to complete Phases 1 and 2 for a small Perth business. Sign in, add your domain, seed with fifteen phrases you use to describe your business, and run clustering. The collapse from a thousand raw candidates to sixty clusters is worth the twenty minutes it takes. From there you can run the classification and prioritisation phases yourself or bring us in for a full keyword strategy engagement with WA market context applied.
The reason keyword research still matters
Keyword research is not glamorous. It does not produce a case study you can screenshot for LinkedIn. But it is the piece of SEO work that quietly determines whether every subsequent hour of content, links, and technical fixes compounds into meaningful growth or gets scattered across pages nobody was ever going to search for. Do it properly in a tool that supports the workflow, and you buy yourself twelve months of focused production. That is the return that lets Perth businesses compete above their weight in verticals dominated by much larger interstate brands.