How much should you actually spend on marketing?

There is no credible published research that breaks this out by specific revenue tier, and any article that hands you a precise percentage for a $10M company versus a $50M one is presenting more precision than the data supports. What does hold up directionally: a functioning marketing engine has a real fixed cost, one strategic owner, a working website, a base level of content and program spend, and that fixed cost doesn't scale down linearly. A $10M company often has to run above the 7-to-10-percent band just to field a credible function, while a $100M company nearer the same percentage is funding meaningfully more absolute dollars and can do more with it. Expect the percentage to compress somewhat as revenue grows, not because marketing matters less, but because the fixed costs are a smaller share of a bigger number.

Yes, more than it should depend on a single industry benchmark. A company targeting flat or modest growth can run near the low end of the range and lean on renewal, referral, and existing-customer expansion. A company with an aggressive new-logo growth target is asking marketing to generate a specific volume of pipeline it doesn't yet have, and that requires budget above maintenance-level spend, sized to the gap between current pipeline and the target, not to a percentage pulled from a benchmark article.

Yes, and it's one of the more reliably documented splits: The CMO Survey's 2026 data puts B2B product companies at roughly 7.0 percent of revenue and B2B services companies at roughly 10.1 percent. The gap makes sense structurally. A services firm is selling trust in a team's judgment more than a fixed feature set, which takes more thought leadership, case studies, and brand-building content to establish, while a product company can lean more heavily on efficient, repeatable demand generation and a self-serve or sales-assisted motion that doesn't require rebuilding credibility with every deal.

When it can't fund one accountable senior owner plus enough spend on at least one channel to reach real frequency, that is the practical floor, well before any percentage calculation matters. A budget spread thin across five channels at token spend on each rarely outperforms a smaller budget concentrated on one or two channels run well, because most channels have a minimum threshold of spend and consistency below which they simply don't compound. If the honest answer to "what would we cut to fund one more channel properly" is nothing, the budget is too small for the ambition attached to it, and the fix is narrowing the ambition, not spreading the same dollars thinner.

Budget against next year's target whenever you're pursuing growth, not against this year's actual revenue. Marketing spend against a target this year produces pipeline that closes months from now, so a budget set against where the company already is will always be a step behind an ambitious growth plan. A steady-state business without an aggressive target can reasonably budget against current revenue, since it isn't trying to outrun its own pipeline math.

Fully loaded compensation for marketing-function headcount, agency and freelance fees, paid media, marketing technology and software licenses, content production, events and sponsorships, and any research or brand work the function commissions. What doesn't belong: sales commissions, sales-specific tools, and general company overhead that isn't specifically a marketing asset. Keeping that line consistent year to year is what makes a percentage-of-revenue figure meaningful at all, a budget that quietly absorbs or sheds categories from one year to the next makes any trend line unreliable.

Yes, for anyone whose role sits inside the marketing function, count fully loaded compensation, salary, benefits, and overhead, not just base pay. No, for sales-side roles, even ones that touch demand generation day to day, like an SDR who follows up on inbound leads. The test isn't who touches marketing-adjacent work, it's which budget and which leader that person's performance rolls up to.

Content that marketing builds, case studies, competitive comparisons, one-pagers, counts as marketing spend even though sales is the one using it in a deal. The CRM, sales engagement platform, and other day-to-day sales tooling belongs on the sales budget, even in the common case where marketing is the one populating and maintaining the content inside it. Draw the line at who owns the asset and its production cost, not at who benefits from using it.

The marketing website's CMS, hosting, and front-end development costs typically belong in the marketing budget, since the site is a marketing asset even when engineering builds it. General company IT infrastructure, internal tools, email systems, security, doesn't. Where it gets genuinely ambiguous is a shared platform like a customer portal, and the right call there is whichever team owns the roadmap and content decisions for it, applied consistently rather than re-litigated each budget cycle.

Often, yes, but there's no single right formula, some companies split it by seat count, others assign the full cost to whichever team owns the platform and cross-charge the other side internally. What matters more than which method you pick is picking one and applying it the same way every year, so a marketing-budget-to-revenue comparison isn't quietly distorted by a shared cost that moved between budgets.

Both are marketing budget regardless of employment status, but it's worth tracking them as separate lines internally, personnel versus outsourced program delivery, because the ratio between the two says something a single total doesn't. Two companies spending the same total percentage of revenue can have very different capacity and risk profiles if one runs mostly in-house and the other runs mostly through agencies, and that ratio matters more when benchmarking against a company with a different insourced-versus-outsourced mix than the raw percentage does.

Most mid-market B2B marketing budgets run more toward demand generation than practitioners say they'd ideally prefer, largely because demand gen produces attributable, short-cycle results and brand doesn't show up in a pipeline report the same quarter it's spent. The healthier default for a company chasing durable, compounding growth is closer to an even split between the two than an all-in demand-gen posture, with the balance shifting toward brand once a category is established and toward demand gen when a specific, near-term pipeline gap needs to close.

Less than most companies are currently spending, if the honest answer to "how much of our stack do we actually use" is less than most of it. Martech is consistently one of the largest single line items in a B2B marketing budget and, across the industry, one of the most underutilized, tools bought for a use case that never fully materialized, or that duplicate a capability another tool already covers. Before adding a martech line, audit what's already licensed and unused, consolidation is usually a faster path to more budget for programs than asking for a bigger number next year.

This depends more on sales cycle length and category maturity than on a fixed ratio. A long, considered B2B sales cycle rewards organic and content investment because its value compounds, an asset built this quarter keeps generating pipeline for years, while paid media stops producing the moment spend stops. A company in a fast-decision category, or one that needs pipeline in the next quarter rather than the next year, should lean more heavily on paid media even knowing it doesn't compound the same way. Neither approach is wrong in isolation; the mismatch is running a long-cycle business on an all-paid-media budget, or a fast-decision one on an all-organic bet.

Yes, and it should be a real, protected budget line with its own review cadence, not whatever happens to be left over at the end of a quarter. Without a deliberate experimentation slice, an existing channel mix tends to calcify, the same channels get the same budget every year because they're the ones with a track record, while genuinely new opportunities never get funded long enough to prove out. The exact size matters less than treating it as non-negotiable and reviewing it on its own merits rather than folding it into whichever channel had the best last quarter.

There's no universal ratio, but there are two warning signs worth watching for. More people than program dollars to deploy means execution capacity is outrunning fuel, a well-staffed team with nothing meaningful to spend on will feel busy without producing much. A large budget with too little accountable staff means the reverse, dollars going out without anyone close enough to the work to know whether they're being spent well. A healthy mid-market marketing organization keeps enough program budget that every hire has real dollars to execute with, rather than headcount alone standing in for a strategy.

Buyers increasingly research vendors through AI tools before ever visiting a website, which means a slice of the content and website budget that used to be aimed purely at search ranking now needs to be aimed at being accurately cited by AI systems, structured, specific, and easy for a model to extract and quote correctly. For most mid-market companies today, this is incremental effort layered onto existing content, SEO, and website spend rather than a large new standalone budget line, but it's worth explicitly naming as a line item in the plan rather than assuming existing SEO spend automatically covers it.

Expect the percentage of revenue to compress somewhat even as the function grows, because the fixed cost of running marketing, one strategic owner, a working website, a base level of content, becomes a smaller share of a larger number. A company at $10M often has to run above the general 7-to-10-percent band just to field a credible function; by $50M to $100M, the same or even a slightly lower percentage funds meaningfully more absolute dollars. Whether a company is B2B product or services explains more of the variance at any given size than raw revenue does on its own.

Reflexive across-the-board cuts are common and usually expensive later, they show up as a pipeline gap and a brand-visibility gap several quarters after the cut, right when the company needs pipeline the most. A better approach is to identify and cut the weakest-performing programs first, the spend with no demonstrated return, while protecting the channels with a track record of producing pipeline. A slowdown is a reason to get more disciplined about what's working, not a reason to cut everything by the same percentage.

Treat it as a distinct project budget layered on top of the ongoing operating budget, not carved out of it, since the day-to-day demand generation engine still needs to run while the repositioning work happens. Size it to the actual scope, messaging and positioning work, visual identity, website rebuild, updated sales collateral, and expect it to run as a one-time or multi-quarter initiative rather than something absorbed into the normal annual run-rate.

Yes. A new vertical needs its own budget for positioning, initial content, and pilot campaigns, tracked separately from the core budget rather than absorbed into it. Without a separate line, it's nearly impossible to tell whether the new vertical is actually earning its keep or simply free-riding on the results and reputation the existing budget already built, which makes the eventual go or no-go decision on that vertical a guess rather than a measured call.

Increase headcount when the constraint is strategic ownership or oversight, a channel with no one accountable for running it well, or a leadership bottleneck where decisions are waiting on someone's time. Increase program spend when the team already has the capacity and judgment to execute but doesn't have enough budget, reach, or frequency behind a channel to see results. Hiring to solve a budget-insufficiency problem, or spending more to solve an ownership-and-judgment problem, is one of the more common and avoidable mismatches in a mid-market marketing budget.

Set the annual number as a planning baseline, then revisit allocation, not necessarily the total, on a quarterly cadence as real performance data comes in. A budget locked in January and never touched again tends to keep funding whatever seemed reasonable at the time even after six months of evidence about what's actually working. The total figure can reasonably hold for the year; which channels and programs get that money should move more often than that.

Anchoring the number to a single industry benchmark percentage instead of to the company's own growth target and B2B product-versus-services dynamics. A close second: treating the number as fixed for the year rather than revisiting how it's allocated once real performance data starts coming in. Both mistakes come from the same instinct, wanting a defensible number without doing the work to defend it.

It's a useful sanity check and a dangerous primary input. Competitor spend and, more importantly, competitor results are rarely visible in any reliable way, so matching a competitor's estimated budget says nothing about whether either company is spending it well. Use competitive context to notice if you're wildly out of step with a category, not as the actual basis for the number.

Often it looks like savings on the current quarter's numbers while quietly building a larger cost later, a pipeline gap that shows up two or three quarters out, lost visibility that takes time to rebuild, and the cost of re-hiring or re-learning once spend resumes. The programs cut fastest are usually the ones easiest to cut, not the ones actually underperforming, which is exactly backward from where the cuts should land.

For anything but a small, deliberately reserved experimentation line, yes. Reactive year-end spending, driven by a fear of losing next year's budget rather than a real opportunity, rarely buys anything as valuable as the same dollars spent deliberately earlier in the year against a plan. If a team routinely finds itself scrambling to spend down a budget in the fourth quarter, the actual problem is that the budget wasn't allocated against a real plan to begin with.

No, the percentage is a planning tool, not a target to hit for its own sake. A company that matches a benchmark percentage while missing its growth target has the wrong number, regardless of how closely it tracks the average. A company spending below the typical range while hitting its growth targets efficiently doesn't need to "catch up" to a benchmark it's already outperforming.

Start from the growth target, not from a benchmark. Name the pipeline and revenue the company needs marketing to produce, work backward to what program spend and headcount that requires, and only then check the resulting number against the general 7-to-10-percent range as a sanity check rather than treating that range as the target itself. A budget built this way is defensible in a way that one pulled from a benchmark article never quite is.