Phone Call Conversion Duration Threshold vs Lead Quality: What Your Call Length Data Says About Real CPA

Why the Default 60-Second Threshold Is Probably Lying to You

Google Ads ships with a 60-second minimum call duration as its out-of-the-box conversion standard. For a SaaS company with a long sales cycle, that might be fine. For a local plumber, roofer, or med-spa, it is almost certainly over-reporting conversions.

Think about what actually happens in those calls:

• 10–30 seconds: Wrong number, voicemail, or a price-shopper who hangs up the moment they hear your rates.
• 30–60 seconds: Could still be a tire-kicker, a vendor call, or someone confirming your address — not a booked job.
• 60–90 seconds: Getting warmer. You have probably had a real exchange, but appointment booking often happens in this window, not before it.
• 90–120+ seconds: Typically reflects a substantive qualifying conversation — the kind that precedes a real job.

When your conversion column counts 60-second calls, Google's smart bidding algorithm trains on that signal. If a meaningful share of those calls are junk, you are teaching an optimization engine to buy more junk. This is the core problem.

The Framework: Threshold Shift Modeling

Before touching your account, you need to understand the math. Here is a structured way to think through it — using a labeled illustrative model, not industry averages pulled from a press release.

Illustrative model inputs (a local HVAC company, not a real client):

| Metric | Value (illustrative) | |---|---| | Monthly ad spend | $5,000 | | Calls generated | 100 | | Reported conversions @ 60 sec | 70 | | Reported CPA @ 60 sec | $71 |

Now apply duration filtering logic — typical for a service business, based on general call-center and lead quality research:

| Duration Threshold | Est. Calls Qualifying | Reported Conversions | Reported CPA | |---|---|---|---| | 60 sec (default) | 70% of calls | 70 | $71 | | 90 sec | ~55% of calls | 55 | $91 | | 120 sec | ~40% of calls | 40 | $125 |

These drop-off percentages are illustrative estimates based on general call duration distribution patterns — not published benchmarks. Your actual data will vary by category, offer, and call handling quality.

The number that matters is not CPA — it is CPA per booked job. If the 70 reported 60-second conversions only produced 30 booked appointments, your real cost per booked appointment is $167, not $71. Tightening the threshold to 120 seconds might drop reported conversions to 40 — but if 35 of those become booked jobs, your real cost per booked job is now $143. That is a better business outcome with a higher reported CPA. This is the counterintuitive insight most local ad managers miss.

What the Research Actually Supports

There is a real, well-documented relationship between call duration and conversion intent. BrightLocal's Local Consumer Review Survey (regularly updated) consistently shows that when consumers call a local business, a large majority have strong purchase intent — they are not browsing. That intent signal is strongest for calls that go beyond an initial greeting exchange.

Separately, call tracking platforms like CallRail have published general guidance (referenced in their product documentation and blog) noting that calls under 30 seconds are almost universally non-converting, and that qualified service calls for home services and healthcare tend to average 2–4 minutes. That range suggests a 90–120 second threshold is a more honest proxy for genuine lead quality than the default 60 seconds.

We are not citing a single magic percentage here because the right threshold is category-dependent. A locksmith emergency call converts fast — maybe 75 seconds of genuine booking intent. A roofing inspection inquiry may need 150 seconds before the homeowner commits to scheduling. The framework matters more than any universal number.

How Threshold Changes Corrupt (or Correct) Smart Bidding

This is where the stakes get serious. Google's Target CPA and Target ROAS bidding strategies are only as good as the conversion signal you feed them.

When you use a 60-second threshold and 30% of those conversions are low-quality calls:

• Smart bidding optimizes toward the audience segments, times of day, and keywords that generate those junk calls.
• Your bid adjustments tilt toward efficiency metrics that do not reflect revenue.
• You may actually be suppressing bids on your best keywords because they attract longer, higher-intent calls that are harder to generate at scale.

When you tighten to 90 or 120 seconds:

• The conversion pool shrinks, but it becomes a cleaner signal.
• Smart bidding re-calibrates toward the traffic that generates real conversations.
• Over a 4–6 week learning period, you typically see reported CPA rise before it stabilizes — do not panic and revert during that window.

This dynamic is closely related to the issue we cover in What Pausing Google Ads Really Costs Local Businesses — disrupting the algorithm's learning state has compounding costs that extend well beyond the pause period itself.

One practical guard: Before changing your threshold, pull 90 days of call data from your call tracking platform or Google Ads call report. Filter by duration. If fewer than 10% of your calls exceed 120 seconds, something is wrong with call handling — fix that first, or you will starve smart bidding of data entirely.

A Step-by-Step Process to Find Your Right Threshold

Step 1 — Audit your call duration distribution. Download your call report from Google Ads (Campaigns > Dimensions > Call Details). Bucket calls into 0–30s, 30–60s, 60–90s, 90–120s, 120s+. You want at least 60 days of data, ideally 90.

Step 2 — Spot-check call quality by duration bucket. Listen to a sample of calls in each bucket. You are looking for: Did the prospect ask a qualifying question? Did your team offer to book? Did either party mention price? Five calls per bucket is enough to identify obvious patterns.

Step 3 — Map duration to booked outcomes. If your CRM or booking system records call source, match call timestamps to jobs booked. Even a rough match on date and phone number tells you a lot. This is the closest you will get to a real conversion-to-revenue signal — more useful than any external benchmark.

Step 4 — Set a new threshold and document the change date. Change the threshold in Google Ads under Conversions > Call from ads > Duration. Note the exact date. You will need this to interpret performance shifts cleanly. Expect a Smart Bidding learning period of 2–4 weeks.

Step 5 — Evaluate at 6 weeks, not 2. Your reported CPA will almost certainly rise. That is expected and probably correct. The question to ask at 6 weeks: Has revenue per conversion improved? If the answer is yes, the higher reported CPA is a feature, not a bug. This connects directly to the payback-period framing in Marketing Payback Period by Channel: Local Ad Spend — the correct denominator is always downstream revenue, not the first cost metric Google surfaces.

For context on how form-based lead quality interacts with similar conversion definition challenges, Single-Step vs Multi-Step Forms: Local Service CPL Guide walks through a parallel problem on the landing page side.

What to Watch for by Business Category

There is no single right threshold. Here are rough directional estimates — treat these as starting hypotheses to test against your own call data, not published benchmarks:

• Emergency services (locksmith, plumber, HVAC emergency): 75–90 seconds. Calls are shorter because urgency is high and decisions are fast.
• Home improvement (roofing, windows, remodeling): 90–120 seconds. These jobs involve scheduling an estimate, which takes back-and-forth.
• Healthcare / medical spa / dental: 90–120 seconds. HIPAA-adjacent categories often involve insurance verification and scheduling complexity.
• Legal / financial services: 120+ seconds. High-ticket, high-stakes — anything shorter is likely an inquiry, not a commitment.
• Restaurants / retail: Call conversion tracking is generally less relevant; this framework applies mainly to appointment- or estimate-driven service businesses.

One signal that should prompt an immediate review: If your average call duration across all incoming calls is under 90 seconds, your call handling process — not your threshold setting — is the real problem. No algorithm adjustment fixes a team that is not converting callers.

The Bottom Line: CPA Is a Ratio. Control Both Numbers.

Your reported CPA is spend divided by conversions. Most local businesses obsess over lowering it by either cutting spend or chasing higher conversion volume. But if the conversion definition is loose, you are optimizing a ratio where the denominator is partly fiction.

Tightening your call duration threshold is one of the highest-leverage, lowest-cost account changes you can make. It costs nothing, takes ten minutes to implement, and — when done with proper audit and monitoring — gives Google's algorithm a materially cleaner signal to work with.

The payoff is not a lower reported CPA. It is a higher revenue-per-conversion and a smart bidding engine that actually knows what a good customer looks like.

---

Want a call duration audit run against your actual account data? Nika Spark's diagnostic review looks at your full conversion definition setup — call thresholds, form attribution, and bidding signal quality — before recommending any changes. Book a strategy call and we will show you exactly where your reported numbers and real business outcomes have drifted apart.