Here’s an uncomfortable exercise. Open your analytics, note last month’s conversion count, and then ask your front desk how many new-customer phone calls they answered in the same period. In most call-heavy businesses, those two numbers have never been in the same room — and the second one is often bigger.
The distance between what your marketing actually produces and what your reporting shows is the attribution gap. This post is about how big it is, why it systematically corrupts budget decisions, and how to measure yours this week.
What the attribution gap is
Web analytics counts conversions that happen in a browser: form fills, purchases, sign-ups, chat starts. But in industries where customers prefer to call — because the purchase is urgent, expensive, complicated, or personal — a large share of conversions happen on a phone line that analytics cannot see.
The journey is digital right up to the last step. The customer searched, clicked your ad, compared you against two competitors, and read your reviews. Then they converted by phone, and every measurable trace of that journey terminated in what your reports call… nothing. A bounce, maybe. An “engaged session” if you’re lucky.
Multiply that by every caller, every month, and you get a reporting system that’s confidently describing a minority of your business.
The data: call share of conversions by industry
How big is the share of conversions arriving by phone? Published industry studies and platform benchmark reports converge on a consistent picture: it varies enormously by vertical, and in phone-first industries it’s not a rounding error — it’s frequently the majority of inbound conversions.
Directionally, across published benchmarks:
| Industry | Typical phone share of inbound conversions |
|---|---|
| Home services (HVAC, plumbing, roofing) | High — often a majority, driven by urgency |
| Legal | High — complex, personal, high-stakes purchases |
| Healthcare & dental | High — appointments overwhelmingly booked by phone |
| Automotive (sales & service) | Substantial — calls rival or exceed form leads |
| Real estate | Substantial — speed-to-contact culture is phone-native |
| Financial services & insurance | Moderate to substantial |
| B2B / SaaS | Lower overall, concentrated in high-value deals |
| E-commerce | Low — the one major vertical where the gap is small |
(Editor’s note: refresh with current cited figures at publication; the pattern is stable year over year, the precise percentages move.)
Two things make these numbers worse than they first appear. First, callers tend to be better leads than form-fillers — higher urgency, faster close rates, larger transactions — a pattern reported consistently across industry studies. So the invisible conversions aren’t a random sample; they’re skewed toward your best customers. Second, mobile search keeps growing, and a phone in the hand makes calling the lowest-friction conversion action there is. The gap is not shrinking on its own.
Why the gap skews budget decisions
An incomplete dataset wouldn’t matter much if the missing data were evenly distributed. It isn’t. Channels differ sharply in how call-heavy their conversions are — and that’s what turns a measurement gap into a decision gap.
Consider how the gap distorts three routine decisions:
Channel comparison. Campaigns targeting urgent, mobile, local intent (“emergency plumber near me”) convert overwhelmingly by phone. Campaigns targeting researching desktop users convert more by form. Compare them on form-fills alone and the urgent-intent campaign — likely your most profitable — looks like the loser.
Automated bidding. Smart Bidding optimizes toward the conversions it’s shown. Feed it only web conversions and it will dutifully shift your budget toward keywords, audiences, and times of day that produce form-fillers, actively bidding away from callers. Your own tooling becomes an engine for misallocation. (The fix is feeding calls in as conversions — see Smart Bidding With Call Conversions.)
Channel-level ROI verdicts. SEO and local search are chronic victims: map-pack visibility and organic rankings drive heavy call volume that never appears in the analytics used to defend the budget. The channel that makes the phone ring gets cut in favor of the channel that makes the dashboard move.
A worked example: the misallocated $10k/month
A home services company spends $10,000/month split evenly across two campaigns.
- Campaign A (“emergency repair” keywords, mobile-heavy): produces 10 form fills and 90 phone calls per month.
- Campaign B (“repair cost guide” keywords, desktop-heavy): produces 40 form fills and 15 phone calls.
What analytics-only reporting shows: A converts 10 times, B converts 40 times. B’s cost per lead looks 4× better. The rational move is to shift budget to B — and many businesses would.
Reality: A produces 100 total leads to B’s 55 — and A’s phone leads, being urgent, close at a higher rate. A is the better campaign by a wide margin, and the “data-driven” decision just defunded it.
Nothing in this example is exotic. It is the default state of any call-heavy account without call tracking: the reporting isn’t just incomplete, it’s directionally wrong, because the missing conversions cluster in specific campaigns.
How to measure your own gap
You don’t need new software to get a first estimate — you need one week and a tally sheet.
Method 1: The front-desk tally (one week, free). Have whoever answers the phone log every inbound call from a new prospect, with a quick “how did you hear about us?” (Imperfect, but fine for sizing.) At week’s end: new-prospect calls ÷ (calls + web conversions) = your rough gap percentage. Most call-heavy businesses are startled by the result.
Method 2: Click-to-call signals (free, directional). Check click-to-call interactions where your stack captures them — ad platform call clicks, GA4 tel-link events if tagged, GBP call taps. These undercount (they miss dialed calls and repeat views) but establish a floor.
Method 3: Call tracking (accurate, ongoing). Tracking numbers plus dynamic number insertion measure the gap precisely and continuously — per channel, campaign, and keyword — which is what you need to actually act on it. The mechanics are covered in the complete call tracking guide.
Whichever method: express the result as “X% of our conversions were invisible to the reports we budget with.” That sentence gets attention in a way dashboards don’t.
Closing the gap: a 3-step plan
Step 1: Capture calls as conversions. Implement call tracking across your website and key placements so every call carries its source. This alone converts the gap from unknown to measured.
Step 2: Feed calls to the systems that decide. Send call conversions into your ad platforms (offline conversion tracking) and analytics (GA4 integration), so both your reports and your bidding algorithms see the whole picture.
Step 3: Weight by quality, not just count. Not every call is a lead. Score or qualify calls — even simple duration rules to start — so the conversions you report and bid on reflect real opportunities. Then compare channels again; expect at least one budget conclusion to reverse.
Frequently asked questions
What percentage of conversions happen by phone?
It ranges from near-zero in e-commerce to a majority in urgent, local, high-consideration industries like home services, legal, and healthcare. The honest answer for your business comes from a one-week measurement, not an industry average — see the methods above.
Which industries get the most phone leads?
Industries where purchases are urgent, expensive, complex, or personal: home services, legal, healthcare and dental, automotive, real estate, and financial services lead the pack. The common thread is that customers want a human before they commit.
How do I measure my attribution gap?
Fastest version: one week of front-desk call tallies compared against the same week’s web conversions. Accurate version: call tracking with dynamic number insertion, which attributes every call to its channel and campaign continuously.
This is a sensitive number to show a CFO — bring the fix along with the problem: the complete call tracking guide and offline conversion tracking. For the modeling side of attribution, see Attribution Models Explained.