Read a Benchmark as a Question, Not a Target
Every benchmark is somebody else's average, measured their way, on their sample, over their dates. Before a number from a report goes anywhere near your plan, it should survive four questions:
- Who is in the sample? A survey of companies with over $1 billion in revenue describes a different business from a 40-person software firm. Neither is wrong. Only one of them is you.
- What does the metric count? A "lead" in one dataset is any form fill or call. In your CRM it may be a contact sales has accepted. Two numbers with the same name can differ by a factor of ten and both be accurate.
- Is it a mean or a median? A handful of huge accounts drags a mean upward. Some publishers report medians and call them averages, which is sensible, but it changes what "above average" means.
- When was it measured? A cost-per-click figure from two years ago describes a different auction. Media prices, privacy changes and buying behaviour all move faster than most decks get updated.
If a benchmark cannot answer those four, it is an anecdote with a decimal point. If it can, it becomes useful in a narrower way than people expect: it tells you whether a gap is worth investigating, not what the right number is.
What B2B Marketing Benchmarks Say About Budget
The most quoted number in marketing is what share of revenue a company spends on it. Gartner's 2026 CMO Spend Survey puts marketing budgets at 7.8% of company revenue, up only slightly from 7.7% in 2025. In the same survey, 56% of CMOs said their marketing organisation lacks the budget to deliver its 2026 strategy.
Read the sample before you put that in a budget request. Gartner surveyed 401 CMOs and marketing leaders in North America, the UK and Europe between January and March 2026, and the vast majority of respondents reported annual revenue above $1 billion. That is a big-company figure. A smaller firm trying to grow into a market usually spends a larger share of a smaller revenue line, because the fixed cost of being visible at all does not shrink with the company.
Two details from Gartner's releases are more useful than the headline:
- Paid media is the biggest single line. In the 2025 survey it accounted for 30.6% of marketing budgets, or 2.4% of company revenue. The rest goes to people, technology, agencies and everything else, which is why "we spend X% on marketing" says little about how much reaches a buyer.
- Readiness changes the number. In 2026, the organisations reporting mature AI readiness averaged marketing budgets of 8.9% of revenue, above the 7.8% average. Budget follows a company's confidence that it can spend well, not the other way round.
So use the budget benchmark as a sense check, not a formula. If you are at 3% of revenue and trying to open a new segment, the gap is worth a conversation. If you are at 12% and pipeline is flat, the benchmark is not your problem. Our post on how to create a marketing budget works the number from the other end: from what the business needs to produce.
Paid Search: The Numbers That Travel Worst
Search advertising has the most detailed public benchmarks of any channel, and they are the easiest to misread. WordStream by LocaliQ's 2026 report looked at over 13,000 search campaigns across 23 industries, running from April 2025 to March 2026, and reports medians in US dollars. For its Business Services category:
| Metric | Business Services | What it does not tell you |
|---|---|---|
| Click-through rate | 6.10% | Whether the people clicking can buy |
| Cost per click | $5.87 | What the click was worth |
| Conversion rate | 4.85% | What counted as a conversion |
| Cost per lead | $93.69 | How many of those leads became opportunities |
The all-industry median cost per lead in the same report was $66.69. A B2B advertiser looking at those two figures might conclude that their channel is expensive. The more important point is that "Business Services" mixes a bookkeeping firm chasing local calls with a software company selling six-figure contracts, and a lead means something different to each.
That is why we report cost per opportunity, not cost per lead, on our demand generation work. A $90 lead that sales rejects is more expensive than a $400 lead that becomes a qualified opportunity, and no public benchmark can see that difference, because it lives in your CRM. Use the search benchmarks to check whether your click-through rate or cost per click is wildly off the market. Judge the channel on what happens after the click.
For the click-to-form step specifically, our post on what a good conversion rate is goes further into why the same rate can be excellent on one page and a warning on another.
Email: The Open Rate Stopped Being a Benchmark
For years the first email benchmark anybody quoted was open rate. It no longer measures what its name says. Apple's Mail Privacy Protection, a setting in the Mail app, prevents senders from seeing whether a message was opened and hides the reader's IP address. Every Apple Mail reader who turns it on stops sending a reliable open signal, so open rates across a list become partly a measure of how many recipients use Apple Mail.
Any benchmark built on opens inherits that distortion, and so does anything triggered by them: re-engagement rules, subject line tests, "send to people who opened" follow-ups. Apple's own documentation notes the protection does not extend to links, which is why the more useful email benchmarks are now further down the funnel:
- Click rate, measured on unique human clicks rather than security-scanner traffic.
- Reply rate on sales-assisted sequences, which no privacy setting inflates.
- Meetings or opportunities per thousand sends, which connects the channel to pipeline.
If a report still leads with open rate, read it for the click and conversion data and skip the headline.
Pipeline: The Numbers Marketing Is Really Judged On
Budget and channel benchmarks describe inputs. The ones a board cares about describe whether the inputs produced revenue, and the most useful public data here comes from sales platforms rather than marketing surveys, because they see the deals themselves.
Ebsta's 2024 B2B Sales Benchmarks mid-year update, an analysis covering $57.3 billion in revenue, found:
Source: Ebsta, 2024 B2B Sales Benchmarks, H1 Update. The same report found sales cycles grew 20% and win rates fell 7% against 2023.
The second and third numbers belong together. Fewer than one opportunity in five fit the profile the seller had defined, and the ones that did won at three times the rate. That is a marketing benchmark in disguise: it says the biggest lever on win rate is not a better sales script but sending sales the right accounts in the first place. It is the argument for an account-based marketing programme that starts with a finite, scored target list.
The first number explains why so many marketing metrics look healthy while deals stall. Eight people now have a say in a typical purchase. A campaign that reaches one enthusiastic champion produces a good lead and a stuck opportunity. The benchmarks that capture this are the ones on our own demand generation reporting list:
Win rate by source
Not one blended win rate, but win rate split by where the opportunity came from. A channel that produces fewer opportunities that close twice as often is the better channel, and a blended number hides it.
Sales acceptance rate
The share of what marketing hands over that sales agrees is worth working. It is the fastest early warning that targeting or qualification has drifted, weeks before win rate can show it.
Pipeline created and influenced
Reported separately, with the attribution window stated. Created means marketing originated it; influenced means it touched an opportunity sales already had. Adding the two together and calling it marketing's number is how pipeline reports lose credibility with finance.
Build Your Own B2B Marketing Benchmarks First
The comparison that matters most is against yourself, last quarter and the same quarter last year, measured the same way. An external benchmark tells you where the market sits. Your own baseline tells you whether you are getting better, and it is the only one built on your definitions.
The B2B marketing benchmarks worth building yourself
- Fix the definitions before the numbers. Write down what a lead, an accepted lead, an opportunity and a closed deal are, so that two people looking at the same CRM record classify it the same way.
- Pull four trailing quarters. One quarter is noise. Four shows seasonality and a trend.
- Split creation from capture. Demand capture converts this month; demand creation compounds over quarters. Reporting them as one number hides which is working.
- Label platform-reported numbers as platform-reported. An ad platform's conversion value is its own claim about its own contribution. Never add two platforms' figures together and present the total as revenue.
- State the attribution window on every view. A 7-day and a 90-day window produce different answers from the same data, and both are correct.
- Flag early data as early. A channel three weeks old has no baseline yet, and pretending otherwise invites bad decisions in either direction.
Steps four to six are the house rules on our own reporting, published on our demand generation page, and the rule does not change with the service. They exist because the fastest way to lose a finance team's trust is a number that cannot be traced back to its source.
| Benchmark | Best external reference | Use it to | Do not use it to |
|---|---|---|---|
| Marketing budget as % of revenue | Gartner CMO Spend Survey | Sense-check a budget against large companies | Set your own budget |
| Search CTR, CPC, cost per lead | WordStream by LocaliQ annual report | Spot an account that is far off the market | Judge a channel's return |
| Email open rate | None worth trusting since Mail Privacy Protection | Watch deliverability trends | Measure engagement |
| Win rate, cycle length, stakeholders | Sales-platform benchmark reports such as Ebsta’s | Explain why deals stall | Replace your own win rate by source |
Three Ways a Good Benchmark Gets Misused
Even well-sourced B2B marketing benchmarks do damage when they are used for the wrong job. Three patterns come up again and again:
- Budget by average. "Companies spend 7.8% of revenue on marketing, so should we." The figure describes mostly billion-dollar firms holding position, not a smaller company trying to win a segment. Work the budget back from the pipeline the business needs.
- Optimising the visible rate. A team told its cost per lead is above benchmark will lower it, usually by loosening targeting until cheaper, worse leads arrive. The benchmark improves and the pipeline does not.
- Comparing across definitions. Your MQL-to-opportunity rate against a report's "lead-to-opportunity" rate is two different fractions with similar names. If you cannot match the definitions, do not draw the conclusion.
The common thread is treating a benchmark as a verdict. It is a prompt to go and look at something in your own data, and the looking is where the useful answer is.


