35.2 Touches Per Qualified Opportunity: The 2026 Benchmark That Breaks Your Pipeline Plan

Your team now spends 35.2 touches to create one qualified opportunity

That is the single most consequential number in Salesloft's 2026 U.S. Revenue Benchmark Report, published on 1 September. Five hundred U.S. sales and revenue decision-makers — CROs, CSOs, VPs of Sales, and RevOps leaders at companies with 200+ employees — reported that their teams average 35.2 touches to create a qualified opportunity. In the same survey, 68.4% said their pipeline quotas have increased.

Read those two findings next to each other and you have the defining operational problem of B2B sales right now. The cost of manufacturing one opportunity went up. The number of opportunities each rep is expected to manufacture also went up. Nobody added headcount to absorb the difference, and touches per qualified opportunity is the metric where that squeeze shows up first.

Most sales leaders have never calculated this number for their own team. They track activity (dials, emails, connects) and they track outcomes (meetings, opportunities, pipeline dollars). They rarely divide one by the other. Once you do, your capacity model stops being a spreadsheet exercise and starts being a hard constraint — and in most orgs, the constraint is already binding.

What the Salesloft benchmark actually found

The study was run with Censuswide, with fieldwork between 28 May and 10 June 2026. The headline findings sit across five areas, and they are worth reading together rather than in isolation:

• Touch inflation. Teams average 35.2 touches per qualified opportunity, and an estimated 19.7% of pipeline is affected by stalled deals, slipped close dates, and other execution breakdowns.
• Quota pressure. Pipeline quotas rose for 68.4% of respondents.
• Performance concentration. Average quota attainment is roughly 62%, while the top 10% of sellers generate 47.4% of closed-won revenue.
• AI maturity gap. Every single respondent reports using AI somewhere in the revenue process, but only 20.6% describe their deployments as production-ready with measurable outcomes. Another 28.2% are still experimenting.
• Stack instability. 26% of organisations are actively consolidating their revenue technology, 32.6% are evaluating where consolidation makes sense, and 26.4% still prefer specialised point solutions.

There is also a diagnostic finding that deserves more attention than it will get: 89% of leaders believe their managers assess seller performance objectively, but only about 32% can instantly diagnose why a deal has stalled. That gap between confidence and evidence is the same gap that lets touch counts inflate for eighteen months without anyone noticing.

Why 35.2 is not just a big number

Thirty-five touches is not inherently alarming. A well-run multi-channel sequence across email, LinkedIn, phone, and a couple of social interactions can easily total thirty-plus touchpoints across a buying group of four or five people, and that is fine — if those touches land on the right accounts at the right moment.

The problem is what 35.2 means as an average across the whole motion, including accounts that never had a reason to buy. It is a blended figure that mixes a handful of warm, signal-triggered accounts that converted in five touches with a very long tail of cold-list accounts that absorbed sixty touches and produced nothing. The average is a symptom. The distribution is the disease.

The capacity math most board decks skip

Here is why touches per qualified opportunity should sit on the same slide as pipeline coverage.

Take a fully-ramped SDR who can realistically execute 120 quality touches a week — meaning genuinely personalised, multi-channel, human-reviewed touches, not 400 merge-field emails. At 35.2 touches per opportunity, that rep generates roughly 3.4 qualified opportunities per week, or about 14 a month, before you subtract holidays, ramp, sick days, admin, and the meetings that no-show.

Now change one variable at a time:

(Arithmetic illustration, not survey data — plug in your own touch capacity.)

The point is the shape of the curve, not the specific rows. Moving from 35 touches to 27 touches per opportunity is a 23% reduction in the effort required per opportunity, which is equivalent to adding a full head to a four-person team without adding a full head to a four-person team. Moving the other way — from 35 to 45, which is what happens when you widen your list to hit a bigger quota — costs you a rep's worth of output while your quota goes up.

That is the trap. When pipeline quota rises 20%, the reflex is to raise activity targets 20%. But raising activity targets almost always means loosening the ICP filter to find more names to touch, which raises touches per opportunity, which eats the gain. Teams end up running harder to stay in the same place, and the benchmark data shows it: average quota attainment is 62%, and Orum's 2026 State of Sales Development report found 61.3% of SDR teams landing below 70% of target.

Four reasons touch counts inflated

Touches per opportunity did not drift upward by accident. Four structural changes pushed it there, and none of them are reversing.

1. Reply rates fell while send volume exploded

Cold outbound reply rates have fallen from roughly 6.8% in 2023 to about 3.43% in 2026. If your reply rate halves and your reply-to-opportunity conversion stays flat, your touches per opportunity roughly doubles. That is the whole mechanism. Most of the inflation in the 35.2 figure is simply arithmetic working against a channel that got noisier.

2. Buyers finish most of the journey before you exist to them

Gartner's research found 67% of B2B buyers prefer a rep-free buying experience, and 45% used AI during a recent purchase. Consensus's 2026 B2B Buyer Behavior Report points the same direction: buyers arrive with requirements already defined and a shortlist already forming.

When the buyer has done 60%+ of the journey privately, your first twenty touches are not "nurturing" anything. They are landing on someone who has not started, or someone who already finished. Both are wasted, and both inflate the average.

3. AI made sending cheap and made attention expensive

Generative tooling collapsed the cost of producing a personalised-looking message to approximately zero. Predictably, volume went up across the entire market. The result is that the marginal touch is worth dramatically less than it was three years ago, because your prospect's inbox absorbed everyone else's marginal touch at the same time.

This is the part where the Salesloft AI maturity finding becomes painful. Every respondent uses AI somewhere. Only 20.6% have deployments that are production-ready with measurable outcomes. The overwhelming majority are using AI to produce more of the thing that stopped working, which is the fastest possible way to drive touches per opportunity up rather than down.

4. Single-threading against a committee

Modern B2B purchases are decided by groups, not individuals. If your sequence targets one contact per account and that account needs five people to agree, you either fail or you go back and start a second sequence against a second contact — doubling touches while the opportunity count stays at one. Multi-threading from the start looks like more touches on paper but usually produces fewer touches per qualified opportunity, because a committee-aware sequence converts at a materially higher rate than five sequential single-threaded attempts.

Touches per opportunity is a targeting metric, not an effort metric

This is the reframe that matters. When teams see 35.2 and decide the answer is "work harder", they add touches. When they see 35.2 and understand it as a targeting number, they remove accounts.

The ICONIQ State of Go-to-Market 2026 research makes this concrete. Companies with heavy AI integration in their funnel show:

• New Lead to MQL: 38% vs 27% for light-AI peers (+11 points)
• MQL to SQL: 37% vs 29% (+8 points)
• SQL to Closed Won: 29% vs 28% (+1 point)
• Demo to Closed Won: 40% vs 37% (+3 points)

Almost the entire measurable AI advantage lands at the top of the funnel. Not in negotiation, not in closing — in who you contact and why. That is precisely the variable that drives touches per qualified opportunity, and it is the only lever in this data that produces double-digit improvement.

What "better targeting" concretely means

Better targeting is not a tighter firmographic filter. Firmographics tell you who could buy. They tell you nothing about who is moving right now. What actually compresses touch counts is timing evidence — an observable behaviour that says this account is in motion this week:

• Someone from the account viewed your profile or your company page
• A decision-maker engaged with a competitor's post or a category thought-leader's post
• The company posted a role that only exists because they have the problem you solve
• A prospect complained about an incumbent tool on LinkedIn, Reddit, or X
• A funding round, a leadership change, or a new office closed in the last 30 days
• Someone in the buying group changed jobs into a role where your product is their problem

A cold ICP-matched account and a warm signal-matched account can look identical in your CRM. They convert nothing alike. The first needs thirty-plus touches and usually produces nothing. The second frequently produces a reply inside five, because the message can reference a real, recent, verifiable thing rather than a merge field.

This is the entire premise behind Updately — capture the warm signals (profile views, post engagers, competitor mentions, hiring signals, pain-point posts across Reddit, LinkedIn and X), score them against ICP, research the prospect properly, and only then write. Fewer accounts, better reasons, far fewer touches per opportunity. The goal is not to send more. It is to earn the right to send at all.

The concentration problem hiding behind the average

The Salesloft finding that the top 10% of sellers generate 47.4% of closed-won revenue is usually read as a talent story: hire better reps, coach the middle up. That reading is incomplete.

Top performers do not generally execute more touches. They execute better-aimed touches. They pick accounts more carefully, they read a signal and act on it the same day, they multi-thread instinctively, and they abandon dead accounts faster than average reps do. In touches-per-opportunity terms, your top decile is probably running at half the blended number, and your bottom half is running at well above it.

Which means the fastest path to improving the blended metric is not coaching your bottom half to work harder. It is systematising what your top decile does in their account selection so the rest of the team inherits it. That is a process and tooling problem, not a motivation problem.

The 32% figure — only about a third of leaders can instantly diagnose why a deal has stalled — is the reason this rarely happens. You cannot copy the behaviour of your best reps if you cannot see what they did differently. Most orgs have activity data (how many touches) and outcome data (did it close), but nothing in between that explains why one sequence worked and another burned forty touches.

A playbook for the next 90 days

Five things you can do before your next planning cycle, ordered by how quickly they move the number.

1. Instrument touches per qualified opportunity, segmented

Do not settle for a blended company-wide figure. Calculate it per segment, per source, and per rep. Total touches in a period divided by qualified opportunities created in that period. Then look at the spread. In almost every team we see, cold-list sourced accounts run two to four times the touch cost of signal-sourced accounts. You cannot argue for changing the motion until that gap is visible in a chart.

2. Cut list size before you cut cadence length

The instinct when touch counts look bad is to shorten sequences. That is usually the wrong lever — shorter sequences on the same bad list just lowers conversion without lowering the cost per opportunity much. Cut the list instead. Take the bottom 40% of accounts by signal strength out of the motion entirely and watch what happens to the ratio. Most teams find total opportunity count barely moves while touch spend falls sharply.

3. Rebuild cadences around triggers, not calendars

A day-1/day-3/day-7 cadence assumes the prospect's readiness is a function of your calendar. It is not. Replace time-based steps with trigger-based entry: an account enters the sequence when a signal fires, and the first message references that signal explicitly. Time-based follow-up is fine after a signal-triggered entry. It is close to worthless as an entry condition.

4. Multi-thread from touch one

Build sequences that address three to five people in the buying group in parallel, with role-appropriate angles, rather than running one contact to exhaustion and then starting over. It costs more touches per account and fewer touches per qualified opportunity.

5. Set an explicit touch budget per account tier

Give each tier a hard ceiling: Tier A gets 25 touches across the committee, Tier B gets 12, Tier C gets 6. When an account exhausts its budget without a response, it exits and returns to the pool for re-entry only when a new signal fires. This single rule stops the long-tail waste that drags the blended average up, and it forces the uncomfortable but correct conversation about which accounts deserve effort.

What this means for your stack decision

The other Salesloft finding — 26% actively consolidating revenue tech, 32.6% evaluating consolidation, 26.4% sticking with point solutions — is not a separate story. It is the same story.

Most outbound stacks were assembled to solve a volume problem: a data provider to find more names, an enrichment layer to fill more fields, a sequencer to send more messages, a scraper to gather more contacts. Every component was optimised for the numerator of the touches-per-opportunity ratio. Almost nothing in a typical 2022-era stack was designed to improve the denominator.

If you are evaluating consolidation this quarter, that is the question worth asking of each tool: does this reduce touches per qualified opportunity, or does it just make it cheaper to send more touches? Tools that help you find timing evidence, research a prospect properly, and write something worth replying to belong in the stack. Tools whose primary value proposition is throughput probably do not, at least not at the price they were worth in 2022.

Takeaways

• 35.2 touches per qualified opportunity is the 2026 U.S. benchmark, from 500 revenue leaders surveyed by Salesloft. Calculate yours. Most teams have never done it, and the number is usually worse than they expect.
• 68.4% of teams face higher pipeline quotas. Raising activity targets to meet them widens your list, raises touches per opportunity, and cancels out the gain. The math is unforgiving.
• Touches per qualified opportunity is a targeting metric. The ICONIQ data shows AI's measurable advantage is almost entirely top-of-funnel — an 11-point lift at lead-to-MQL, a single point at SQL-to-closed-won. Better selection beats more sending, and it is not close.
• Your top decile already runs a lower number. Systematise their account selection rather than pushing your bottom half to work harder.
• Only 20.6% of teams have production-ready AI. Using AI to increase send volume actively makes this metric worse. Use it to decide who and when, then let humans approve what goes out.
• Set a touch budget per tier and enforce it. The long tail of accounts that absorb sixty touches and produce nothing is where the blended average goes to die.

If your 2027 plan assumes the same touch cost per opportunity as your 2025 plan, it is already wrong. The channel got noisier, the buyers got more self-directed, and the benchmark moved. The teams that hit quota next year will be the ones that cut the list rather than the ones that raised the target.

Signal-led outbound is not a nicer way to do the same thing. It is the only lever in the current data that reliably moves the denominator.