Phone Call vs. Form Fill Conversion Value: Why Equal Worth Breaks Smart Bidding for Local Businesses
The Hidden Assumption Wrecking Your Bids
Most local business Google Ads accounts are set up with a simple, logical-sounding default: count every conversion. A form fill is a conversion. A phone call is a conversion. Done.
The problem? Smart Bidding — whether you're running Target CPA, Target ROAS, or Maximize Conversion Value — doesn't just count conversions. It weights them. When you assign the same value to a 45-second phone call from a ready-to-book customer and a contact form submitted by someone price-shopping three competitors, you're telling Google's algorithm they're worth the same. They almost certainly aren't.
The result is a bidding model that is mathematically precise and strategically wrong.
Why Calls and Form Fills Close at Different Rates
Before we touch bid strategy, let's anchor this in close-rate reality.
For most local service businesses — plumbers, dentists, HVAC contractors, law firms, med spas — inbound phone calls close at a materially higher rate than form fills. The reasons are structural:
- Intent signal: A person who picks up the phone and dials has already made a micro-commitment. A form fill has zero friction.
- Qualification speed: A 3-minute call can confirm budget, timing, and fit. A form fill starts a follow-up sequence that may take days.
- Drop-off risk: Form leads go cold fast. Industry experience consistently shows that response-time decay is steep — leads contacted after even a few hours convert at a fraction of the rate of immediate follow-up.
A rough rule of thumb we use with local service clients: assume phone calls close at 2–4× the rate of raw form fills until your CRM tells you otherwise. For high-ticket services (think legal, roofing, cosmetic procedures), the gap is often wider.
This isn't a citation — it's a starting model you should calibrate with your own data. But it's a far more honest starting point than assuming parity.
What Miscalibrated Values Do to Smart Bidding (A Modeled Example)
Let's make this concrete with a clearly-labeled illustrative model.
Scenario: A residential HVAC company running Target ROAS
| Conversion Type | Actual Close Rate (estimate) | Avg Job Value | True Revenue Value | |---|---|---|---| | Phone Call | 40% | $1,800 | $720 | | Form Fill | 12% | $1,800 | $216 |
If you assign both a value of $1 (or both the same flat dollar amount), Smart Bidding treats them as identical signals. It will happily trade expensive clicks that generate phone calls for cheaper clicks that generate form fills — because on paper, both look like equal wins.
The modeled CPA impact: Imagine your account generates 60 conversions/month — 20 calls and 40 form fills. At parity values, Smart Bidding sees 60 equal conversions. In reality, those 20 calls are generating roughly $14,400 in pipeline (20 × $720) while the 40 form fills are generating roughly $8,640 (40 × $216). Total true pipeline: ~$23,040.
Now imagine the algorithm, optimizing for the wrong signal, flips that ratio over 90 days to 10 calls and 50 form fills. Same 60 conversions on paper. True pipeline drops to approximately $18,000 — a roughly 22% revenue decline with no change in spend. Your cost-per-conversion looks identical. Your revenue quietly craters.
This is the shape of the problem. The exact numbers will differ for your business — but the direction of the error is consistent.
The Three-Step Framework to Fix Conversion Values
Step 1: Pull your actual close rates by source
This requires a CRM, not just Google Ads data. Tag leads by source (call vs. form) and track them through to closed revenue. Even 60–90 days of data gives you a working ratio. If you don't have a CRM yet, start with a conservative assumption (calls close at 3× forms) and revisit quarterly.
Step 2: Assign revenue-based conversion values, not arbitrary scores
Use this formula:
`Conversion Value = Close Rate × Average Job/Transaction Value`
For the HVAC example above: Call value = 0.40 × $1,800 = $720. Form value = 0.12 × $1,800 = $216. Enter these as your conversion values in Google Ads. Now Smart Bidding is optimizing toward revenue, not conversion volume.
This connects directly to the ROAS framing covered in our article [Offer Page vs. Service Page: Which Cuts CPA for Local Ads?] — bid strategy and landing page strategy have to be calibrated to the same revenue signal.
Step 3: Audit your conversion action setup for double-counting
A common error: businesses import Google Analytics goals AND set up native Google Ads call tracking, meaning every phone call fires twice. Double-counted calls at the wrong value will destroy Smart Bidding's model. Audit your conversion actions and mark duplicates as 'Secondary' (count for reporting only, not bidding).
For more on how conversion action structure affects algorithm performance, see our article [Google Ads: Consolidate or Segment for Smart Bidding?].
Account Type Nuances: Not Every Local Business Has the Same Gap
The call/form value gap isn't uniform. Here's how it tends to vary by account type:
- Emergency services (plumbing, locksmith, HVAC repair): Gap is largest. Calls are almost always immediate-intent. Form fills are often comparison-shoppers. Calls may be worth 4–6× more.
- Elective/scheduled services (med spa, landscaping, remodeling): Gap still exists but is narrower. Form fills with a strong offer can convert reasonably well if follow-up is fast. A 2–3× call premium is a reasonable starting model.
- Professional services (legal, financial, therapy): Forms may actually be preferred by cautious buyers doing research. The gap may be smaller — but calls that do come in tend to be very high intent. Worth modeling rather than assuming.
- E-commerce-adjacent local (boutique retail, specialty food): Form fills (newsletter, quote request) may have near-zero close rate. Calls may not be your primary conversion at all. Map your actual funnel before assigning any values.
The point: there is no universal correct ratio. The framework is universal. The numbers aren't.
The Payback Period Problem
Miscalibrated conversion values don't just hurt this month's performance — they compound. Smart Bidding models learn over 2–4 weeks of data. A model trained on wrong values builds the wrong intuitions about which audiences, devices, times of day, and keywords produce 'good' conversions.
This is why fixing conversion values mid-flight often causes a temporary performance dip before it improves — the model is unlearning bad patterns. Expect 3–4 weeks of recalibration. Budget for it.
This payback dynamic is exactly why channel-level CAC tracking matters from day one — a theme we break down in [CAC Payback Period by Channel: Local Business Guide]. The sooner your conversion values reflect real revenue, the shorter your payback window.
Bottom Line: Smart Bidding Is Only as Smart as Your Inputs
Google's algorithm is genuinely powerful. It processes thousands of signals per auction that no human can replicate manually. But it is not psychic. It optimizes toward the target you set, using the values you define.
Garbage values in, garbage bids out.
The fix isn't complicated — it's disciplined: 1. Pull real close rates from your CRM by lead type. 2. Calculate revenue-based conversion values. 3. Enter them in Google Ads and audit for double-counting. 4. Revisit every quarter as your data improves.
This is the kind of account hygiene that separates businesses paying for noise from businesses paying for pipeline.
---
Want a second set of eyes on your conversion value setup? At Nika Spark, our account audits start with exactly this analysis — because no amount of creative or budget optimization fixes a bidding model that's optimizing toward the wrong signal. Book a free strategy call and we'll show you what your current setup is actually telling Google to do.
Sources
- 1.Google Ads Help (2024) — Official documentation confirming that Smart Bidding (Target ROAS, Maximize Conversion Value) uses assigned conversion values to weight auctions — not just conversion counts. link
- 2.Harvard Business Review / InsideSales research (widely cited) — Lead response time decay: odds of qualifying a lead drop dramatically after the first hour of non-response. Often cited as 10x drop after 5 minutes vs. 30 minutes — use directionally, not as a precise figure. link