B2B Keyword Strategy: Intent Over Volume

Take a specialist compliance consultancy. Over 90 days, it buys 186 clicks from the phrase-match keyword "compliance services" at £14 each. It receives 12 enquiries and passes one to sales. A second theme—"FCA compliance consultant for fintech"—produces 31 clicks at £29, six enquiries and four sales conversations.

The larger keyword looks efficient: cheaper clicks, more forms and more visible demand. Commercially, it is the worse buy.

That is the central rule of a B2B Google Ads keyword strategy. Search volume tells you how often wording appears. It does not tell you whether the searcher runs a company you can serve, has the problem you solve or is selecting a supplier.

For niche B2B, intent outranks volume. A phrase searched 20 times a month can be worth more than one searched 2,000 times if it identifies the need, buyer context and readiness to evaluate. The wider B2B paid-search playbook covers campaign structure and budget decisions. This article stays narrower: which meanings deserve keywords, and how match types should control their reach.

A B2B Google Ads keyword strategy starts with commercial meaning

A keyword is an instruction to Google about a meaning you want to target. A search term is the wording a person actually typed. Confusing the two creates false confidence: an exact-match keyword can still trigger several search terms, while one broad keyword can reach meanings you never wrote down.

Search volume is also a demand estimate, not a value score. It cannot measure account fit, buying authority, urgency, contract value or your chance of winning.

Consider "asset management software". That label can refer to investment portfolios, IT hardware, facilities or digital media. A large volume number may simply combine several markets. More searches do not resolve the ambiguity; they make it more expensive.

Specialist buyers often signal value through detail:

• A regulated task: "automate SOC 2 evidence collection"
• A vertical: "contract management software for biotech"
• A technical environment: "warehouse system that integrates with NetSuite"
• A replacement event: "alternative to spreadsheets for supplier audits"

A reported volume of zero or ten should not automatically remove those terms. Keyword tools have limited data at the long tail. Treat the estimate as confidence about available demand, not proof of commercial value.

The first selection test is simple: if this search came from a real person, would the wording give sales a credible reason to speak with them? If not, volume cannot rescue it.

Select B2B keywords from the sale backwards

Start with language attached to actual buying conversations, not a list exported from a tool.

Useful sources include closed-won discovery notes, qualified lost opportunities, contact-form messages, sales emails, customer interviews, CRM rejection reasons and product implementation calls. These reveal the words buyers use before marketing tidies them into category jargon.

Look for four signals:

A term does not need "buy" or "demo" to carry intent. In technical B2B markets, a precise workflow, regulation or integration can be a stronger buying signal than a generic transactional modifier.

We use a simple weighted score to stop volume dominating the decision:

Priority score = (commercial intent × 35%) + (ideal-client fit × 30%) + (offer fit × 25%) + (demand confidence × 10%)

Score each factor from one to five. Suppose "compliance software" scores 2, 1, 3 and 5. Its weighted result is 2.25 out of 5. "audit evidence automation for credit unions" might score 5, 5, 5 and 1, producing 4.6.

The second term loses badly on volume confidence and wins decisively on everything that produces revenue. That is the intended result.

Do not manufacture unnatural, ten-word combinations to inflate specificity. Preserve the compact phrase a buyer would use, then let match types cover credible variations. The unit of strategy is a distinct commercial meaning, not every singular, plural and word order.

Match types are reach controls, not intent labels

Google Ads’ current match-type rules work with meaning, not just strings of characters. The three positive match types set how far the system can move from the keyword’s meaning:

Our plain-English Google Ads glossary covers the wider terminology. The operational point is that punctuation changes reach, not the commercial quality of the underlying idea.

Exact match is not literal match

Exact gives the most steering, but it does not promise character-for-character matching. Google applies close variants to every positive match type and offers no opt-out. They can include misspellings, singulars, plurals, reordered words, implied words, synonyms, paraphrases and searches Google judges to have the same intent.

Use exact for category, use-case, integration and evaluation terms whose meaning is already valuable. Review the search terms that it triggers; brackets are not a substitute for judgement.

Avoid building dozens of overlong exact keywords. Extremely specific strings can receive low-search-volume status and add no useful coverage while inactive. Capture the shortest phrase that preserves the commercial meaning.

Phrase match is a controlled learning layer

Phrase match reaches the exact meaning plus searches that add or imply context. It suits niche B2B themes where buyers describe the same need in several natural ways.

The centre of the phrase still has to be unambiguous. "risk management software" could mean financial, cyber, supplier or health-and-safety risk. "supplier risk management software" gives Google—and the person reading the advert—a much firmer instruction.

Phrase match is not an old-fashioned word-order lock. The search terms report, not the quotation marks, reveals how Google interpreted it.

Broad match should be earned

Broad match has the widest reach and is the default when a keyword has no brackets or quotation marks. Google’s broad-match guidance says it can use additional signals to judge relevance and recommends pairing broad match with Smart Bidding.

That is a platform recommendation, not a reason to make broad the default for every niche account.

Use it when the core theme is specific, the offer genuinely serves that theme and conversion data represents a meaningful business outcome. If the account records every form submission as equally valuable, automation cannot distinguish a serious procurement enquiry from an irrelevant one. The B2B conversion-tracking setup covers that measurement work without turning this into a tracking guide.

Broader match types can already reach queries covered by narrower ones. Cloning every keyword into exact, phrase and broad does not create three clean audiences. Give each version a stated job: control, discovery or expansion.

Worked example: lower volume, cheaper opportunities

Take a UK consultancy implementing Microsoft Dynamics 365 Business Central for food manufacturers. Its average first-year contract is £48,000.

Over 90 days, its general phrase-match theme "ERP software" produces:

• 180 clicks at £12: 180 × £12 = £2,160
• Nine enquiries
• One sales-qualified opportunity
• Cost per qualified opportunity: £2,160 ÷ 1 = £2,160

A specialist group of exact and phrase keywords—[food manufacturing ERP consultant], "Business Central implementation for food manufacturers" and [ERP traceability software food manufacturing]—produces:

• 42 clicks at £28: 42 × £28 = £1,176
• Five enquiries
• Four sales-qualified opportunities
• Cost per qualified opportunity: £1,176 ÷ 4 = £294

The specialist clicks cost 133% more. The qualified opportunities cost 86% less:

1 − (£294 ÷ £2,160) = 86.4%

Nothing in those figures proves the specialist terms will scale indefinitely. They do show why CPC and search volume are poor proxies for expense. The costly click is the one that had little chance of becoming a customer.

This calculation also stops at opportunity cost. It is not ROAS until revenue and attribution are reconciled; the distinction between platform and blended ROAS matters before making that claim. If relevant searches still produce unsuitable prospects, use the separate process for diagnosing poor B2B lead quality rather than asking keyword volume to explain every failure.

How to operate a B2B keyword strategy without chasing volume

A useful keyword list is a record of commercial hypotheses. For every theme, document the buyer signal, plausible alternative meanings, evidence source, intended match type and the page capable of satisfying the search. If no page can honestly fulfil the implied need, do not buy the click.

Then apply five rules.

1. Review search terms by meaning. The keyword row shows what you selected; the search terms report shows what you paid to reach. Promote useful new language into a controlled theme. Narrow or pause a keyword when its repeated interpretations conflict with the offer.

2. Pool close variants. Sparse B2B data becomes even weaker when singulars, plurals and minor word orders are judged separately. Assess the commercial theme as a whole unless a variant reveals genuinely different intent.

3. Put sales evidence beside media evidence. Track qualified-opportunity rate and the reason sales accepted or rejected each theme. CTR can show that an advert attracts attention. It cannot establish that the underlying keyword attracts suitable companies.

4. Expand one semantic step at a time. Move from a proven exact meaning to phrase variations, then test broad expansion when measurement is trustworthy. Jumping from "food manufacturing ERP" to "business software" is not expansion; it is a change of market.

5. Use volume as a constraint. Once two themes clear the fit and intent threshold, demand estimates can decide which to test first. Volume belongs at the end of selection, not the start.

The same record should exist whether campaigns are run in-house or through ongoing Google Ads management. Without it, keyword changes become reactions to a dashboard. With it, each change has a commercial reason and a falsifiable result.

FAQ

What is the best keyword match type for B2B Google Ads?

There is no universal winner. Exact suits proven, narrow meanings. Phrase explores credible wording around them. Broad can expand a validated theme when the account has trustworthy conversion data. Start with the level of control the evidence supports.

Should we target keywords that show zero monthly searches?

Sometimes. Keep a term when it comes from real buyer language, signals strong fit and expresses a credible need. Use a compact version rather than every long-tail permutation. Google’s low-search-volume status keeps a keyword inactive until demand increases, so combine tool estimates with phrase coverage and search-term evidence.

How many keywords should a niche B2B company start with?

Start with roughly 10–30 distinct intent themes, not hundreds of cosmetic variants. The correct count is the number of commercially different needs, contexts and evaluation paths the offer can satisfy. Expand when evidence reveals a new meaning, not when the spreadsheet looks short.

Should competitor names be included?

They can indicate evaluation intent, but only when your offer is a credible alternative and the resulting advert and page can answer the comparison honestly. Competitor traffic is not automatically qualified, and volume alone is not a reason to pursue it.

Summary

• Commercial meaning, ideal-client fit and offer fit should outrank search volume.
• Build keyword themes from sales language, then use tools to estimate demand.
• Exact controls a narrow meaning; phrase discovers variations; broad expands proven territory.
• Exact match still includes close variants, so inspect search terms rather than trusting punctuation.
• Judge keyword themes by qualified opportunities, not cheap clicks or a long keyword list.

Actualyse runs precision B2B Google Ads programmes optimised for lead quality, not click volume. Book a call to talk through where yours stands.