How Do You Defend a B2B Sales Cycle Against a Pre-COVID Benchmark?

Question and scope

An investor shows you a B2B sales benchmark from before 2020 and concludes that your sales cycle is too long, your conversion too low, or your team too small for the revenue you plan to book. The question of this article is simple: how do you answer without getting defensive and without inventing anything, defending your sales cycle with data you can actually show?

The scope is deliberately narrow. The article covers B2B sales with a medium or long cycle, where several people on the customer side take part in the decision and an opportunity stays open for weeks or months. It does not cover self-serve sales, pricing, or the valuation of your company.

One honesty limit needs to be stated up front. You often hear that "before COVID, B2B selling relied on brute force, and everything has changed since". None of the sources opened for this article measures that change on either side of 2020. The strongest studies we read cover 2021 to 2025. They show that the indicators move a great deal from one year to the next, which is already a useful argument, but they do not prove that 2019 was better or worse. So we do not present the "before" and "after" opposition as a demonstrated fact. We treat it as a question to put to the investor: where does your figure come from, on which sample, with which definition?

An illustrative example to fix ideas. A team of three salespeople sells management software to industrial small businesses. An investor tells them that "the normal cycle is 60 days" while theirs shows 110. Before discussing the team's performance, the right reflex is to ask what the 60-day figure measures: from first meeting to signature, or from contact creation to cash collected? On which contract size? For which type of customer? Without those answers, the two numbers cannot be compared.

This guide therefore gives you a four-step method: check where the benchmark comes from, measure your own cycle, compare it with recent studies whose sample you know, then present the gap with an action plan. It also recalls what tools can contribute, without promising anything about your results. To place this decision in context, the Knowledge guides on finance gather the in-depth analyses of the same field.

Dataset

We kept sources whose original document we could open and whose sample is described. Each one is used for what it really says, with its limits.

First source: the 2023 Ebsta and Pavilion B2B sales benchmark report. It analyses 3.2 million opportunities from 364 companies, representing over $37 billion of pipeline, and mainly covers 2022 compared with 2021. The authors state that they have analysed hundreds of billions of dollars of pipeline since 2020, which places the start of their series: the report has no earlier point of comparison.

Second source: the 2025 report from the same publishers, covering 655,000 opportunities worth $48 billion, with a survey of more than 2,000 CROs and sales leaders. The report specifies that all percentage figures are relative: a rise in win rate from 20% to 30% reads as +50%, not +10 points. That precision prevents many wrong readings.

Third source: Salesforce's State of Sales report, seventh edition. It rests on an anonymous survey run from August to September 2025 among 4,050 sales professionals in 22 countries, all recruited as third-party panelists. It documents perceptions and self-reported time allocation, not cycle measurements.

Fourth source: an analysis published by Gong in July 2025 and updated in 2026, based on more than 28 million cold emails. It is used here to size the prospecting effort needed to get a meeting, with the caveat that the publisher itself sells sales analytics tools.

Fifth source: a Gradient Works blog article gathering 2025 benchmarks. It cites an average sales cycle of 6.5 months against 4.9 months in 2019 and attributes it to the 2025 Ebsta and Pavilion publication. We searched the original report for both values and did not find them, and the report itself gives 91 days for new business. We therefore do not adopt these figures. They mainly illustrate a danger: a relayed figure changes meaning as it travels from site to site.

Finally, Highspot's article on sales motions offers a useful definition: a sales motion is the repeatable go-to-market play a B2B company runs to reach a specific buyer segment. That vocabulary helps frame what you are defending in front of an investor.

This dataset has three known gaps. It contains no pre-2020 measurement verified at the source. It comes almost entirely from commercial software vendors who benefit when teams track their indicators closely. And it mixes data observed in customer relationship systems with survey answers. These three limits come back in the limitations section.

Methodology

The method has four steps, to be run in order. It is designed for a small team preparing a funding round or a review with a shareholder, without a dedicated data department.

Step 1. Ask where the benchmark comes from. Do not contest the number, question it. Five questions are enough: which year, which sample, which contract size, how the start and end of the cycle are defined, and which source. Write the answers down. If the investor cannot give them, the comparison gap becomes a subject for discussion rather than a verdict. If the benchmark is sourced, you at least have a shared starting point.

Step 2. Measure your own cycle. Export from your sales tracking tool every closed-won and closed-lost opportunity of the last twelve to twenty-four months. For each, compute the time between the creation of the opportunity and the final decision. Use the median rather than the mean, because a few very large deals pull the mean upward. Split by contract size and by type of customer. Write the definition you used at the top of the table.

Step 3. Look at what lengthens or shortens your deals. The 2023 Ebsta and Pavilion report proposes three observable levers you can measure yourself: the number of relationships engaged on the customer side, the quality of qualification, and the time elapsed since opening. On the first, for enterprise deals, the best average win rate (42%) appears with 10 to 12 relationships engaged, and the win rate falls sharply beyond 16 relationships. On qualification, when the MEDDPICC method is fully completed, the win rate is 311% higher, yet only 15% of opportunities are fully qualified. On time, opportunities open for more than twice the average cycle length had only a 3% chance of closing. You can rebuild these three views with a simple spreadsheet.

Step 4. Present an explained gap and a plan. In front of the investor, do not show an isolated figure. Show your median, its definition, its spread, the deals that stretch it, and what you are doing to shorten them. Add the number of deals behind each figure. A figure computed on twelve deals does not weigh like one computed on three hundred, and the investor knows it.

One method point is often forgotten: each quarter, freeze a copy of your pipeline with the close dates forecast at that moment, then compare them with the actual dates. The gap between the two, deal by deal, measures the quality of your forecasts. It answers directly the question an investor asks when facing a long cycle: does the team know when it will sign? The 2023 report notes that 37% of deals had slipped and that time itself works against a deal that stays open. A team that shows three quarters of forecasts compared with outcomes brings stronger proof than any benchmark.

To go deeper on this step, Validating your market with Ember's Fund Your Growth describes a use case directly related to preparing this evidence.

A word on tools. Ember is the publisher of Lead Intelligence and Fund Your Growth, and this guide is not a comparison. Both can help on specific steps: Lead Intelligence ranks opportunities from your commercial context, and Fund Your Growth structures the evidence of a funding file. The rest of the method works perfectly well without them.

Analysis

What does the analysis of the data we read say? Four readings stand out, and none of them claims that the past was better or worse.

First reading: sales indicators move a lot from one year to the next. In the 2023 Ebsta and Pavilion report, between 2021 and 2022, average deal value fell 32%, win rate fell 15%, and sales cycles lengthened by 32%. In the 2025 report from the same publishers, compared with 2024, the new-business win rate is 19% (down 10% in relative terms), deal values rose 54% and cycles shortened by 9%. In two years, one publisher therefore measures a cycle that lengthens and then a cycle that shortens. If a benchmark can vary by more than 30% in a single year, a pre-2020 benchmark cannot serve as a norm unless it is dated, described and set against a recent measurement.

Second reading: the definition matters as much as the figure. The 2025 report gives an average cycle of 91 days for new business, against 52 days for sales to existing customers. Both numbers live in the same report and differ by about 75%. An "average" cycle that blends new customers and expansions no longer means anything. Likewise, the 2025 report indicates an average of eight stakeholders on the customer side for a new deal and five for an expansion. When the investor compares your cycle with a single figure, ask which category of deal they are talking about.

Third reading: the cycle is partly the result of your method. In the 2023 report, the highest win rates match a reasonable number of relationships and complete qualification, and deals that drag lose their chances of closing. The 2025 report adds that involving decision makers in the first two stages raises the win rate by 55% in relative terms, and that 36% of deals slipped from their planned date. These elements belong to the sales team, not to the market. They support an honest defence: your cycle is long because this type of deal has six stakeholders, and here is the part you control.

Fourth reading: planned close dates are rarely reliable, and investors know it. The 2023 report finds that 68% of opportunities had a close date set earlier than the observed "golden period", that 89% of close dates fell on the last day of a calendar month, and that 17% of opportunities had their date changed more than three times by more than a week. The authors place that golden period between 31 and 60 days for small deals, between 61 and 90 days for medium ones, and between 150 and 180 days for the largest, with a win rate 165% higher when the opportunity closes within it. For your defence, this means a long cycle is not what worries an experienced investor most. What worries them is a pipeline whose dates are optimistic and keep moving. A long but stable cycle, with close dates revised on written criteria, is easier to defend than a short cycle whose forecasts slip at every meeting.

Salesforce's survey completes the picture on human time: according to respondents, sellers spend almost one full day of their workweek on prospecting, 47% consider cold outreach one of the worst parts of the job, and 47% say their team lacks the bandwidth to do it. Respondents also report spending on average 16% of their time on preparation and planning. These are self-reported perceptions, not stopwatch measurements. They still indicate where sales time is lost.

Finally, Gong's analysis sets an order of magnitude on outbound prospecting effort: the average rep has to send 344 cold emails to land one meeting, and top performers book 8.1 times more meetings than average performers. The most useful message of that study is not the 344 figure but the gap between the best and the rest, which shows that targeting and message quality weigh more than volume. Gong also says that pitching the product in the email body can reduce reply rates by as much as 57%, and that the best emails run 100 words or fewer.

Findings

Here is what you can defend, worded carefully, and what you cannot defend with the data we read.

What the data allows. A sales cycle has no single reference value. Figures vary with deal size, type of customer (new or existing), number of stakeholders and year. The 2025 Ebsta and Pavilion report, for instance, distinguishes 91 days for new business from 52 days for expansions. An investor who compares you with a figure lacking these distinctions is comparing badly.

What the data also allows. Qualification, decision-maker involvement and control of deal age are associated with better results in the observed samples. According to the 2023 report, outside the ideal period the probability of closing drops markedly: opportunities running one extra month beyond the optimum can lose up to 60% of their chances, and up to 90% after two months. That gives a concrete argument for a weekly pipeline review.

What the data does not allow. It does not let you say that pre-2020 B2B selling was slower, faster, more profitable or more "brute force" than today's. Nor does it let you say that a given tool shortens your cycle. Any sentence of that kind, in a document meant for an investor, should be removed or reworded as a hypothesis to test.

A three-sentence summary for an investor. First, our median cycle is X days over Y deals, measured from opportunity creation to decision. Second, it ranges from A to B depending on contract size, and here are the two factors that lengthen it. Third, here is what we are changing this quarter and the indicator that will show whether it works. This structure replaces a discussion about a 2019 figure with a discussion about your data.

A point of caution on relayed figures. The Gradient Works blog gives a win rate of 19% in 2025, "down from 29% in 2024". The original report gives 19% with a relative decline of 10%, which implies roughly 21% the previous year, by our calculation. Between the two versions the gap is eight points. That kind of divergence is enough to justify always going back to the source before putting a number behind an argument.

To extend the reflection, What an SME leader should prioritise in 2026 when capital is tight and growth has stalled offers a decision frame for the next choice.

Limitations

This analysis has limits that you should announce to the investor before they find them on their own.

Vendor sources. Ebsta sells a revenue intelligence platform, Salesforce sells customer relationship management products, Gong sells a conversation analytics platform and Highspot sells sales enablement tools. Their studies are useful and detailed, but their conclusions also serve them. They are not independent academic studies and do not claim to be.

Unrepresentative samples. The 2023 Ebsta and Pavilion report relies on data from 364 companies, described by its authors as among the best performing. The 2025 report mixes pipeline data and survey answers. A small company that sells to ten customers a year will not necessarily recognise itself in these averages.

Relative percentages. Almost all the percentages quoted are relative changes. A 10% drop from a 20% win rate gives 18%, not 10%. The 2025 report says so explicitly, and we have said so too, but it is the first source of error when a figure is copied into a presentation.

Perceptions rather than measurements. The Salesforce survey reports what 4,050 sellers declare, not what they do. Time spent prospecting in particular is not measured with a stopwatch. Gong's figures come from data belonging to the publisher's customers, with no detail on industries or on the exact sending period.

No comparison with the pre-COVID period. This is the main limit. We have no verified series on either side of 2020. If the investor puts forward a 2019 benchmark, we can neither confirm nor refute it. We can only ask where it comes from and suggest confronting it with your own deals.

Tool limits. Lead Intelligence guarantees no commercial result. Sending is manual by design: the cadence plans and proposes, and the person validates and sends. Searching profiles on LinkedIn or Sales Navigator depends on a connected LinkedIn account and on its activation. Fund Your Growth prepares a file without guaranteeing funding, and access is opened progressively depending on the account.

Conclusions

Faced with a pre-COVID benchmark, the best defence is not to argue that the world has changed, it is to prove that your figures are solid. Three ideas to keep.

The first: a figure without a definition cannot be compared. Ask for the year, the sample, the contract size and the boundaries of the cycle. Most disagreements between a founder and an investor disappear when both measure the same thing.

The second: benchmarks move fast, even within one publisher. Between 2022 and 2025, Ebsta and Pavilion successively measure a cycle 32% longer and then 9% shorter and, in each of the two reports, a declining win rate. A reference that is not dated is only an opinion.

The third: a good defence is made of your own data, its definition, the factors you control and a plan. It calls for neither record prospecting volume nor a tool promise. It calls for rigour in measurement and clarity in presentation.

What should you say on the day of the meeting? First acknowledge that the investor's benchmark is a legitimate reference without accepting it as is. Ask where it comes from. Then present your figures with their definition and sample size. Finally, acknowledge what you do not yet know, for example the effect of a recent change on your cycle, and propose a date by which you will come back with the measurement. That attitude is more convincing than a defence that looks perfect, because it shows a founder who can tell what is established from what remains to be shown.

Tools can support this rigour without replacing it. Ember is the publisher of Lead Intelligence and Fund Your Growth. Lead Intelligence starts from a prospecting or fundraising mission, reuses the context already present in Ember, looks for accounts and people according to your ideal customer profile, checks sources and ranks opportunities with a proposed next action. Fund Your Growth guides the founder through nine building blocks of the project, with human validation at each step, and stores the documents in a Data Room linked to the file. Neither replaces the calculation of your median cycle.

In practice, What Angel Investors and Pre-Seed VCs Screen in the First 10 Minutes complements this frame with another angle on the same subject.

Recommendations

Here are eight concrete actions, from the most urgent to the least urgent.

1. Write down your definition of the cycle. One sentence is enough: "from the day the opportunity is created to the day the contract is signed". Place it above every cycle figure you show.
2. Compute the median on your recent deals. Twelve to twenty-four months, won and lost, split by contract size. State the number of deals behind each figure.
3. Separate new customers from expansions. The Ebsta and Pavilion data show that the two behave very differently, with 91 days against 52 days in their 2025 sample. Do not mix them.
4. Count the stakeholders per won deal. The 2025 report cites an average of eight stakeholders for a new deal. See where you stand and whether the lost deals had fewer or more.
5. Set up a weekly review of ageing deals. Decide in advance a duration beyond which a deal is requalified or dropped, for example twice your median. The 2023 report observes that deals open for more than twice the average length closed in only 3% of cases, an order of magnitude to compare with your own.
6. Prepare the answer to the provenance question. Before the meeting, write the five questions to ask about the investor's benchmark, and how you will present your figures if the answers are missing.
7. Improve targeting before volume. Gong's data show that the gap between top sellers and the average is larger than volume alone would explain. Sharpen your ideal customer profile and test a short message, with no product pitch in the first email.
8. Use tools where they save time. Lead Intelligence can help rank accounts from your context, including from an imported Excel or CSV file, and Fund Your Growth can structure the documents of your file. To present the strategy to a board, Creation offers three templates with a preview, asks the useful questions, then asks for confirmation before creating the document. Check which functions are activated on your account.

To go further on preparing a funding round, Angel Investor Screening Criteria for Pre-Seed in 2026 helps connect this method to investors' priorities.

When to use this analysis

Use this frame in four specific situations.

Before a funding round or a shareholder review. An investor quotes an old benchmark to challenge your cycle, your sales efficiency or your hiring plan. You have ten days to prepare the answer. The frame gives you the questions to ask and the structure of the reply.

When preparing a sales budget. You need to justify a headcount or a prospecting budget from a revenue target. Computing the median and the spread of your cycle avoids building the budget on a market figure you cannot verify.

When your cycle changes. You notice deals lengthening or shortening over two or three quarters. The levers observed in the studies (number of stakeholders, qualification, deal age) offer a diagnostic grid.

When you doubt a figure found online. You want to use it in a presentation. The provenance rule in the method, with the 19% versus 29% divergence noted above, shows how to verify before quoting.

A last precaution on vocabulary. In documents addressed to an investor, avoid phrases like "the market has changed since COVID" without supporting data. Prefer dated and quantified sentences, such as "on our deals closed between January and December, the median is X days". The investor can argue with a general statement; they can hardly argue with a measurement whose definition and deal count you provide.

This frame does not suit you if you mostly sell self-serve, if you have fewer than ten closed deals over the period (the median then has no statistical meaning), or if the debate with the investor is about something other than the cycle, such as market size or margin. In those cases, start by clarifying the subject of the disagreement.

A few examples of questions worth asking an investor, to adapt to your situation: "Does this benchmark include expansions at existing customers?", "On which contract band was it measured?", "Does the cycle start at contact creation or at the first meeting?". These questions are not defensive. They show that you know what needs to be measured.

A worked example to redo with your data

This section is an illustrative example built for this guide. The numbers below are fictional and have nothing to do with a real company. Replace them with your own.

Imagine a small team that closed forty deals over eighteen months, twenty-two of them won. It exports its deals and computes the time between opportunity creation and decision. Here is the summary it prepares for the meeting.

• Small contracts: 18 deals, median duration of 45 days, a single stakeholder in most cases.
• Medium contracts: 15 deals, median duration of 82 days, two to three stakeholders on the customer side.
• Large accounts: 7 deals, median duration of 150 days, procurement and management involved, seven to ten relationships per deal.

Three lines of reading go with this summary. First, the overall median (for example 68 days) means nothing if the three segments are mixed: the team shows the three lines. Second, the seven large accounts are few: the team says so and draws no strong conclusion. Finally, it states which deals went past twice the median of their segment and what happened in each.

From there, the answer to the investor becomes concrete. It does not say "our cycle is normal". It says: "On our small contracts, our median is 45 days. On our large accounts it is 150 days, but we have only seven deals. Here are the two factors that lengthen cycles and here is what we are testing next quarter."

To build this summary faster, an Excel or CSV file imported into Lead Intelligence can serve as a starting point to rank accounts, and Fund Your Growth can store the summary in the file's Data Room with the other documents. These functions do not compute your median for you. They help organise the material. A spreadsheet is enough for the exercise itself.

Sources

The following sources were opened to prepare this version of the article, on 29 September 2026. Every figure in the text comes from one of them, in its original version, except the illustrative summary in the previous section, whose numbers are fictional.

• Ebsta and Pavilion, 2023 B2B Sales Benchmark Report: 3.2 million opportunities, 364 companies, 2022 trends compared with 2021, relationships, qualification and deal age.
• Ebsta and Pavilion, 2025 GTM Benchmarks Report: 655,000 opportunities, $48 billion, survey of more than 2,000 leaders, cycles, win rates, stakeholders and slipped deals.
• Salesforce, State of Sales, seventh edition: survey of 4,050 professionals in 22 countries, August and September 2025, prospecting time and perceptions.
• Gong, Does cold email even work anymore?: more than 28 million emails, 344 emails per meeting, gap between top reps and the average, published on 24 July 2025.
• Gradient Works, 2025 B2B sales performance benchmarks: secondary source, used only to show that a relayed figure can diverge from its original report.
• Highspot, Rethinking your enterprise sales motion in the age of AI: definition of a sales motion, updated on 11 August 2026.

The Ember functions described in this text match what the product does today. Access depends on your account. See Ember's pricing page for the conditions of use.