Blogs / How Long Should a Call Last to Count as a Lead?

How Long Should a Call Last to Count as a Lead?

Nov 15, 20267 min read
Pulkit Khurana

Founder, SproutMe

A line drawing of a funnel filtering out short calls, illustrating call length conversion in Google Ads.

Your Google Ads account reports more call conversions than your front desk remembers taking. The calls they do remember include wrong numbers and people who hung up at the greeting. A minimum call length setting decided which of those counted, and if it's set wrong, your bidding is learning from noise.

No single call length makes a phone call a lead. Google Ads doesn't set or recommend one: you choose the minimum, and every call lasting at least that long counts as a conversion. The right threshold is the shortest call in which your team can genuinely qualify a caller, and you read it from your own call data. Where calls close sales, imported outcomes should replace duration altogether.

How does Google decide a call counts?

The threshold is one decision inside the wider job of tracking calls from Google Ads. It matters more than most, because it decides what your bidding learns.

Google Ads offers five ways to count phone calls as conversions. Only two use a duration you control: calls from ads, and calls to the phone number on your website. For both, Google's documentation on call conversion tracking states the mechanism plainly. You set a minimum call length, every call that lasts at least that long is counted, and shorter calls aren't. Both methods depend on Google forwarding numbers with call reporting switched on.

What the documentation doesn't do is suggest a number. It gives no recommended duration and no range, only an instruction to pick a length appropriate for your business needs. If the field arrived pre-filled in your account, treat that value as a placeholder, not as advice.

Two other methods ignore your threshold entirely. Clicks on call ads and call assets count based on Google's estimate of whether a meaningful phone call happened after the click, and Google doesn't disclose a duration or explain how the estimate works. The import method skips the stopwatch. You track calls in another system, such as a CRM, and send back only the calls that mattered, with sale values if you have them.

Calls to your website number raise setup questions of their own, covered in how to track calls from your website, not just your ads. For both duration-based methods, though, the decision is the same: where to draw the line.

Why isn't there a standard call length?

Duration is a proxy. What you actually want to count is a qualified conversation, and how long qualification takes depends on what you sell and who answers the phone.

Take an emergency plumber. Their dispatcher can qualify a caller in moments: location, problem, available slot, done. A firm selling a complex B2B service needs a proper discovery conversation before it knows whether the caller is a prospect. One threshold applied to both would overcount the second business and undercount the first.

The businesses with the most riding on this question don't agree on a figure either. In pay-per-call, a qualified call is the unit a buyer pays for, yet every campaign defines its own payable event. Duration sits alongside other tests there: a phone menu that screens for intent, geography, staffed hours, and the quality of any transfer to a live agent. If buyers who pay by the call can't settle on one number, no general rule will hold for your account.

What the threshold does decide is which direction your errors run. Each direction has its own cost.

  • Too low, and wrong numbers, misdials and instant hang-ups count as leads. Conversion-based bidding treats them as wins and goes looking for more auctions that produce them. Your cost per conversion looks healthy while your pipeline doesn't move.
  • Too high, and genuine short calls disappear. A returning customer booking a slot or a caller who already knows what they want can both finish fast. Your bidding gets less signal, and your reports understate what Google Ads is contributing.

Of the two, counting too much is usually the more expensive mistake. The bidding doesn't just report the noise, it actively optimizes toward it.

How do you find your own threshold?

The answer is already in your account. You just have to put it next to what happened after each call.

  1. Pull call durations. The Call details report in Google Ads' Report Editor lists how long each tracked call lasted.
  2. Match calls to outcomes. Take a recent stretch of calls and mark each one against what actually happened: booked, quoted, existing customer, spam, wrong number. Your front desk log, call recordings or CRM will tell you.
  3. Find the break. Sort the calls by length. In most accounts there's a band of short calls that are almost all noise, then a point where real enquiries start to dominate. Set the threshold just above that noise band, not at the length of your typical good call. A threshold set at the typical good call would throw away every genuine call shorter than it.
  4. Recheck when anything changes. A new recorded greeting or phone menu adds time to every call without making any of them more qualified. A new offer, a different team on the phones, or a seasonal shift in who calls can all move the break.

If there's no clean break, that tells you something too. When good calls and junk calls overlap heavily in length, duration is a weak proxy for your business, and tuning it further won't fix the signal.

When should duration stop being the test?

Duration measures how long someone stayed on the line, not whether they became a customer. If the phone is where your sale happens, that gap is the whole problem. Measuring the length of a call becomes a poor substitute for knowing what the call produced.

The import method closes it. Google describes it as giving you more control over which calls you count, for example only the calls that include a sale, with the sale value attached. The cost is operational. Your CRM or call system has to record calls reliably and match them back to Google Ads. That's worth doing when call volume is meaningful and the value of a call varies widely, because a two-minute repair enquiry and a two-minute contract enquiry should never teach your bidding the same lesson.

The wider call-tracking market is moving the same way. Qualification has historically leaned on call duration, but call tracking increasingly judges calls on intent, transcribed conversation and what happened afterwards.

You don't have to switch overnight. Keep a duration-based action counting while the import beds in. Once outcomes are flowing reliably, make the imported conversion the one your bidding optimizes toward. Checking whether your threshold ever tracked real leads means putting ad data and CRM data side by side. SproutMe Companion answers questions grounded in Google Ads and HubSpot data held in one place, so that comparison becomes something you ask rather than a spreadsheet you rebuild every month.

Conclusion

The minimum call length is yours to set, and Google deliberately leaves it that way. Borrowed figures, whether from a forum, a pay-per-call network or a pre-filled field, describe someone else's calls. Set the threshold just above the noise in your own call data, recheck it whenever your call handling changes, and move to imported outcomes once the phone is where revenue is decided.

See how SproutMe Execute adjusts bids and budgets continuously from the conversions you define, within spend and scope guardrails.

Frequently Asked Questions

Google Ads shows call durations in the Call details report, found in the Report Editor. It only covers calls tracked through Google forwarding numbers with call reporting enabled. If you haven't switched those on, Google has no duration data to report. In that case, you'll need to enable them before you can calibrate a minimum call length against your own calls.

No. Pay-per-call thresholds decide when a buyer owes a network money. Each campaign sets them for itself, usually alongside other tests like intent screening and staffed hours. No published evidence shows that any of those durations predicts sales. Copying one into Google Ads gives you false precision. A threshold calibrated on your own matched calls is more defensible.

Not reliably. Raising the minimum filters out more wrong numbers and hang-ups, but it also discards genuine short calls such as quick bookings from returning customers. None of the available guidance links a longer minimum to better sales outcomes. Only your own call data, matched against what each caller went on to do, can show where quality actually starts.

Grow smarter with AI marketing tips

Join our newsletter to get practical insights, automation ideas, and performance tips straight to your inbox.

Get a complimentary audit to uncover AI opportunities hidden in your data.

Put these strategies to work