Sales facts for 2026: what the data actually says.

Most "sales statistics" pages are copies of other sales statistics pages. A number gets quoted, then quoted again, and after a few rounds nobody can tell you where it came from. Some of the most shared ones have no source at all.

So I did this the slow way. The figures on this page were read at the publisher's own report, press release or PDF, and where one couldn't be, I say so.

Each comes with who produced it and how, because that changes what it means. A Gartner survey of buyers, Gong's analysis of calls on its own platform, and an Ofcom survey of the public are three different kinds of evidence. All three are useful.

How to read the labels:

Buyers: fewer want a rep, but most still check with one

Gartner's survey of 646 B2B buyers, published in March 2026, found that 67% prefer a rep-free experience. Gartner has asked versions of this for years. The wording changes, so treat it as a trend rather than a precise line, but the direction is clear: 33% in 2020, 43% in 2021, 61% in 2025, 67% now. Independent research.

Gartner's next release complicates that. From the same August to September 2025 survey period, it found 69% of buyers prefer to check AI-generated insights with a sales rep, and buyers were 39 percentage points more likely to say a rep, rather than GenAI, understood their needs. They don't fully trust either source: 51% say they're more likely to meet misleading information from AI, 49% from a rep. Independent research.

So buyers want to do the reading themselves, and they want a person when it's time to check what they've read and decide.

Some more on who you're dealing with:

• 45% of B2B buyers used AI during a recent purchase (Gartner, 2026).
• 73% actively avoid suppliers who send irrelevant outreach (Gartner, survey of 632 buyers, 2025).
• Buying groups range from five to 16 people across up to four functions, and 74% of buyer teams show "unhealthy conflict" during the decision (Gartner, 2025).
• Forrester puts the typical decision at 13 internal stakeholders and nine external influencers, from a survey of nearly 18,000 buyers. Procurement is a decision-maker in 53% of buying cycles. Independent research.

The often quoted "buyers spend 17% of their time with suppliers" is real, but it comes from Gartner research published in 2020 or earlier. It's still useful, as long as you know it's at least six years old.

On the phone, that means the call has to be relevant to them. A generic pitch gets you avoided. A call about something specific to their business, from someone who sounds like they understand it, is the part buyers still value.

Sellers: 40% of the week spent selling

Salesforce's State of Sales 2026 surveyed 4,050 sales professionals in 22 countries, including the UK. Vendor survey.

• The average seller spends 40% of their time selling. Gen Z reps spend 35%. The widely quoted 28% is usually credited to older editions.
• 57% say customers are taking longer to decide.
• 42% of reps feel overwhelmed by too many tools. Salesforce sells an all-in-one platform, so that finding suits its pitch, but the complaint is a familiar one.
• Reps spend nearly a full day a week prospecting, and 48% still say they lack the bandwidth for adequate cold outreach.
• 47% say cold outreach is one of the worst parts of the job. The press release rounds this up to "the worst part". The report itself says "one of".

On quota, the sources point in different directions. Ebsta and Pavilion's 2025 benchmark, built on Ebsta's platform data and a survey of sales leaders, found 78% of sellers missed quota in the year to early 2025. Vendor platform data and survey. HubSpot's 2025 survey found 60% of sales pros felt on track. That's self-reported confidence rather than attainment, so the two contradict each other less than they look.

Cold calling: the real numbers

The largest cold call dataset I found is Gong's analysis of more than 300 million cold calls recorded on its platform, published in 2024 with 30 Minutes to President's Club. Vendor platform data. It describes teams that use Gong.

Gong doesn't define "top rep" precisely, so read the right-hand column as "what good looks like" rather than a target.

Other findings from the same guide:

• The median successful cold call lasts 4.8 minutes. Gong reads that as handling at least two objections before a meeting is booked.
• Reaching a gatekeeper cuts the chance of success by 39%.
• After five dials to the same person, the chance of connecting starts to drop significantly.
• Leaving a voicemail cuts the connect rate on later dials, but nearly doubles the reply rate to your emails.
• Calls lift email replies even when nobody answers: 1.81% without calls, 3.44% with them.
• On openers, "Did I catch you at a bad time?" booked meetings on 2.15% of calls. A permission-based opener scored 11.18%.

You'll often see "'How have you been?' is 6.6 times more successful" quoted alongside the 300 million figure. That stat comes from a different, smaller Gong study of 90,380 calls in 2018. Both are real. They're different datasets and shouldn't be blended.

Cognism's 2026 report, from over 200,000 calls in WHAM's data, gives an industry average success rate of 2.7%, up from 2.3% the year before. It doesn't define success, and the figure has moved a lot between editions (4.82%, then 2.3%, then 2.7%), so treat it as directional. Vendor platform data.

On timing, the data disagrees. Gong's 2022 study of 100,000 calls found 11am to noon and 4pm to 5pm best. Cognism's 2026 data puts 10am to 11am first and 2pm to 3pm second. Test on your own list before you trust either.

And on activity: The Bridge Group's 2025 survey of 351 companies found phone-centric SDR teams average 56 dials and 4.6 quality conversations a day. Independent research, mostly North American SaaS companies.

Objections and conversations

In Gong's 300 million call dataset, the top five objections account for 74% of all objections, with "not interested" and product fit the most common. Gong describes most objections as brush-offs against the interruption itself. Vendor platform data.

That makes objection handling more learnable than it feels on a bad day. On Gong's numbers, five answers, practised until they're natural, cover three quarters of what you'll hear.

On talking and listening, Gong's 2025 analysis of 326,000 sales calls of at least ten minutes found:

• The average talk-to-listen ratio is 60:40.
• Reps talked 57% of the time in won deals and 62% in lost ones.
• Sellers who won asked 15 to 16 questions per call. Those who lost asked about 20.

Cold calls are different. Gong's 2022 cold call study found successful reps talked more, about 55:45, and found no statistical difference in how many questions were asked on successful and unsuccessful calls. On a cold call you're carrying the conversation until there's a reason for them to carry it. On longer sales calls, the winners asked fewer questions.

You'll often see a "43:57 golden ratio" quoted, usually credited to an early Gong study. I haven't read that study at source, but Gong's own 2025 data shows reps talk more than that even in deals they win.

AI in sales: lots of use, little measured proof

Use is high:

• 87% of sales organisations use some form of AI, and 54% of sellers say they've used AI agents (Salesforce, 2026). Vendor survey, and Salesforce sells AI agents.
• Only 8% of reps in HubSpot's 2025 survey said they use no AI at all. Vendor survey.

What it actually changes is far less certain:

• Gartner's survey of sales leaders says AI saves sellers about 4.8 hours a week. But 72% of sales organisations report low reinvestment of that time into high-value selling. Those are leaders' estimates rather than timed measurement. Independent research.
• Returns are split. 25% of sales organisations report a return of 50% or more on AI, and 20% report a negative return of 50% or more (Gartner, 2026).
• Salesforce's widely shared "agents cut research time by 34%" is what sellers expect once agents are fully in place. It's a projection.

The best controlled study I could find comes from customer support. Brynjolfsson, Li and Raymond (NBER, 2023) measured a generative AI assistant rolled out to 5,179 customer support agents: issues resolved per hour rose 14% on average and 34% for novice and low-skilled workers, with little gain for the most experienced. It's a reasonable guide to what AI might do for a new SDR, and nothing more.

I'm a strong believer in using AI for research and first drafts, then spending the time it saves on the phone. Gartner's 72% says most teams don't.

The UK phone market

These figures come from Ofcom and the ONS, so they cover the UK rather than a vendor's customer base. Official.

• Only 26% of UK phone users say they're likely to pick up a call from an unrecognised UK mobile number. 19% would pick up a withheld number, and 9% an international one (Ofcom survey of 2,216 people, February 2025). That's consumers rather than business buyers, but it's the clearest official evidence on answering unknown numbers, and it's why a recognisable UK number matters.
• 42% of UK phone users received a suspicious call in the three months to February 2025. That's the noise your call arrives into.
• UK call volumes fell 7% in 2025, to 164 billion minutes, and 91% of call minutes now start on mobile networks (Ofcom, 2026).
• 88% of UK SMEs use mobile calls, 81% use WhatsApp for calls, and 52% still have a landline. Mobile is the most used service for business-to-business calls (Ofcom business research, 2025).
• About 35% of UK businesses with ten or more staff used at least one AI technology in June 2026, up from around 12% in late 2023. Only 10% of those use it extensively (ONS, 2026).

There's no official or independent UK dataset on B2B connect rates, meeting rates or SDR careers. Everything in the cold calling section above is international, and where the sample is described, mostly North American.

Sales careers

From The Bridge Group's 2025 survey of 351 companies, 78% of them North American and mostly SaaS. Independent research.

• SDRs take 3.0 months to ramp, the fastest since 2010.
• Average SDR tenure is 1.9 years.
• 60% of SDRs hit quota, the lowest in the study's history.
• Median annual SDR attrition is 40%.

And from Salesforce's 2026 survey: 46% of Gen Z sellers rarely get feedback on their sales conversations, and only 26% of sales teams run roleplay. If you run a team, regular call review and roleplay is the cheap fix most teams skip.

Stats to stop sharing

These turn up everywhere. I couldn't find any of them at a primary source, or the source says something different.

• "80% of sales need five follow-ups, but 44% of reps give up after one." No primary study found.
• "It takes 8 (or 18) attempts to reach a prospect." Often credited to Gartner or TOPO. No primary source found.
• "Three call attempts reach 93% of everyone who'll ever pick up." Credited to Cognism. Not on its report page.
• "Win rates fell from 29% to 19%" (Ebsta). A misreading. The report gives a 19% new-logo win rate and a 10% relative fall, which implies about 21% the year before.
• "The average B2B sales cycle is now 6.5 months." Not in the report it's credited to, which says sales cycles got 9% shorter.
• "Buying groups have 6 to 10 decision makers." An old Gartner figure. The current ones are five to 16 (Gartner) and 13 plus nine external influencers (Forrester).
• "Reps spend 28% of their time selling." Superseded. The 2026 Salesforce figure is 40%.
• "Social sellers outsell 78% of their peers." No primary LinkedIn source found.
• The number of businesses on the CTPS. Figures from about 1 million to 2.3 million circulate. The TPS homepage doesn't publish one.

I'll update this page as new primary research comes out. If you've seen a figure you think belongs here, check where it came from first. If you can't find the original, don't share it.