Cost Per Acquisition by Match Type Mix: What Happens to Local Google Ads CPA When You Shift Toward Broad

The Setup: Why Match Type Mix Is the Wrong Place to Start

Every few months a local business owner gets a tip — from a rep, a Reddit thread, or an agency — that broad match has gotten smarter and they should 'open up' their campaigns. CPA climbs. They blame broad match. They flip back to exact. CPA drops. Case closed, right?

Not quite. Match type is the distribution valve; negative keywords are the filter. Open the valve without improving the filter and you get dirty water. The match type mix matters, but it is almost never the root cause of CPA inflation — the negative keyword gap is.

This article builds a practical framework for understanding what actually happens to CPA as you shift that mix, so you can make the decision deliberately rather than reactively.

A Labeled Model: CPA Trajectories Across Three Match Type Scenarios

We can't hand you a universal CPA number — it varies wildly by category, market, and bid strategy. What we can do is model the trajectory using a consistent illustrative baseline.

Illustrative model — treat these as directional, not measured benchmarks:

Assume a local HVAC business running roughly 500 clicks per month at a $4.00 average CPC (a rough mid-range estimate for competitive local services; actual CPCs vary significantly by market). That's a $2,000/month ad spend.

| Scenario | Match Mix | Est. Conversion Rate | Modeled CPA | |---|---|---|---| | A — Exact-heavy | 80% exact, 20% phrase | ~6–8% | ~$50–$67 | | B — Balanced | 50% exact, 30% phrase, 20% broad | ~5–7% | ~$57–$80 | | C — Broad-heavy (weak negatives) | 20% exact, 30% phrase, 50% broad | ~3–5% | ~$80–$133 | | D — Broad-heavy (strong negatives) | 20% exact, 30% phrase, 50% broad | ~5–7% | ~$57–$80 |

The critical observation: Scenario C and Scenario D have the exact same match type mix. The CPA gap between them is driven entirely by negative keyword hygiene — not by broad match itself.

Conversion rate compression in Scenario C comes from irrelevant search terms eating clicks: 'HVAC certification courses,' 'HVAC parts wholesale,' 'DIY HVAC repair.' None of those convert for a service business. They inflate spend and deflate conversion rate simultaneously — a double hit to CPA.

The Mechanism: Why the Negative Keyword Gap Is the Real Culprit

Broad match — especially paired with Smart Bidding — works by finding intent signals beyond exact keyword strings. When it works, it surfaces buyers you'd have missed. When it fails for local businesses, it's almost always because:

1. Geographic bleed. Broad match can trigger on queries from outside your service area, or queries with ambiguous location intent. A plumber in Cleveland doesn't want impressions on 'plumber near me' from a user 40 miles out. 2. Funnel-stage mismatch. Broad match pulls in upper-funnel research queries. Someone searching 'how long does a furnace last' is not ready to book a repair call today. 3. Category bleed. Queries that share words with your service but belong to a different buyer entirely (the HVAC example above).

The negative keyword list is the only lever that addresses all three. Match type ratio changes move which bucket traffic falls into — but without exclusions, you are just rerouting dirty traffic, not filtering it.

This is the same principle behind what we cover in Blended ROAS by Traffic Source: The Hidden Drag — unfiltered traffic sources drag down blended performance even when individual 'good' traffic performs well.

One Solid Benchmark Worth Citing

Google's own internal data, referenced in multiple Search Central and Google Ads Help resources, indicates that broad match combined with Smart Bidding and a strong first-party data signal tends to find comparable or better conversion volume than exact match alone — but that finding comes with a critical asterisk: it assumes the campaign has conversion history and a properly configured negative keyword list.

For local businesses that are earlier in their campaign maturity — limited conversion data, thin negative lists — that condition is rarely met. The benchmark is real; the precondition is the part that gets omitted in the pitch.

A separate widely-cited industry pattern (referenced across multiple PPC practitioner analyses, though specific figures vary by source) is that irrelevant search terms can account for 20–40% of broad match spend in campaigns without active negative keyword management. We'll call this a practitioner estimate rather than a precise citation, because the exact figure shifts by vertical and campaign structure — but the directional reality is consistent: the waste is material.

The Audit Sequence: How to Diagnose Your Match Type Mix Problem

Before you change a single match type setting, run this sequence:

Step 1 — Pull the Search Terms Report (last 90 days) Filter to the broad and phrase match campaigns. Export every search term that triggered an impression. Do not skip this; it is the only ground truth.

Step 2 — Tag each term into one of four buckets:

• ✅ Relevant, converting — keep, consider adding as exact match
• ✅ Relevant, not yet converting — monitor; may need landing page work (see Google Ads Conversion Lag by Local Service Category for realistic timelines by vertical)
• ❌ Irrelevant, never will convert — add to negative list immediately
• ⚠️ Ambiguous — flag for review

Step 3 — Calculate your irrelevant spend ratio For every 'irrelevant' term, pull its impressions and clicks. Multiply clicks × your average CPC. Sum it. Divide by total spend for the period. If this number exceeds ~15%, your negative keyword gap is the CPA driver — not your match type ratio.

Step 4 — Build your negative list before you shift mix Add exact and phrase negatives for every irrelevant bucket term. Add broad negatives for category-bleed themes (e.g., 'DIY,' 'course,' 'certification,' 'wholesale,' 'parts'). Apply at the campaign and account level where appropriate.

Step 5 — Then, and only then, test the mix shift If you want to move toward more broad match, do it in a controlled test: one campaign, 30–60 days, same bid strategy, with the new negative infrastructure in place. Compare CPA and ROAS against your exact-heavy baseline — not against industry benchmarks, against your own data.

This sequencing matters for bid strategy decisions too. The interaction between match type and bid strategy is real — we break down that relationship in Portfolio vs Single Bid Strategy: Local Business CPA.

When Broad Match Actually Makes Sense for Local

Broad match is not the enemy. It genuinely earns its place when:

• Your campaign has 30+ conversions per month — Smart Bidding needs signal to function; below that threshold, it is guessing
• You have a robust negative list already in place — ideally built from at least 60–90 days of search term data
• Your service area targeting is tight and verified — radius targeting + location bid adjustments reduce geographic bleed
• You are deliberately prospecting for volume, with the understanding that CPA may be higher than exact-match traffic and you are managing to ROAS across the full funnel, not CPA on individual clicks

The last point is important. Exact-heavy campaigns often look great on CPA but leave volume on the table. A blended approach — exact for proven money terms, phrase and broad for discovery — can improve total revenue even if blended CPA ticks up slightly. The goal is profitable revenue, not the lowest possible CPA on a thin slice of traffic.

Bottom Line: Fix the Filter Before You Move the Valve

The match type mix debate is largely a distraction from the real lever: negative keyword discipline. CPA inflation in broad-heavy campaigns is almost always traceable to irrelevant search terms burning budget, not to some inherent flaw in broad match itself.

Run the audit sequence above before making any match type changes. If your irrelevant spend ratio is above ~15%, you have found your CPA problem — and it is fixable without touching your match type mix at all.

If you want a second set of eyes on your search terms report and match type structure, book a call with the Nika Spark team. We'll show you exactly where your CPA is leaking and what the fix looks like in your specific account — before we talk about anything else.