Here is how the B2B SaaS marketing budget gets set at most post-seed companies I meet. The board asks for a growth number. The founder works out how much marketing feels affordable. Someone divides that across channels in proportions that seem reasonable. Twelve months later nobody can say what the money bought, and the same conversation happens again with a bigger number.
Capital-efficient planning runs the other way. Start from the revenue the company has to produce, work back to the pipeline that revenue needs, and only then ask what that pipeline should cost and how fast the cash should come back. This post is the framework I use, the external benchmarks worth anchoring on, and the two calculations that tell you whether a budget will pay back before the runway does.
What capital-efficient actually means
A capital-efficient marketing budget is one where every euro spent on acquisition comes back, from the gross margin of the customers it produced, inside a period the company can fund. It is not the smallest budget. Small budgets can be wildly inefficient. It is the budget whose payback is short enough that growth does not depend on the next round closing.
So three things: what you spend, what it produces, and how fast it returns. Most budgets I see only manage the first.
What B2B SaaS companies actually spend
Benchmarks are a sanity check, not a target, but you should know them. The most useful public source I have found is SaaS Capital’s 2026 spending survey, which covers more than a thousand private B2B SaaS companies. The medians: marketing at 8% of ARR, unchanged from the year before; selling costs at 15%, up from 13%. For companies between $3M and $5M ARR specifically, 8% on marketing and 12% on selling costs. Equity-backed companies spend roughly twice what bootstrapped ones do on marketing and about 70% more on sales, and grow at a median of 25% against 20%.
Two things I take from that. Marketing is usually the smaller half of go-to-market spend, so a marketing budget planned without looking at sales headcount is half a plan. And the gap between funded and bootstrapped spend is large while the gap in growth is small, which is exactly why investors now ask about efficiency and not just growth.
Work backwards from revenue, not forwards from a percentage
Do not start with a percentage. Start with the number the company has to hit and derive the budget from the funnel.
Fix the new ARR target for the period, net of whatever existing customers will expand. Turn it into deals using your real average contract value, by segment if you have more than one. Turn deals into opportunities using your actual win rate, not the one in the deck. Turn opportunities into the sourced pipeline you need, and split it across channels using last year’s real mix rather than the mix you hope for. Then price each channel’s share using its cost per qualified opportunity from the last two quarters. The sum is the budget.
An illustration, with round numbers you should replace with your own. Say a company needs €1,500,000 in new ARR. At a €30,000 average contract that is 50 deals. At a 25% win rate, 200 opportunities. If the two channels that produced opportunities last year cost €1,800 and €3,200 per qualified opportunity and split 60/40, the budget comes out at roughly €216,000 plus €256,000, call it €472,000. That number is either fundable or it is not, and if it is not, the plan changes before the spend rather than after.
What the exercise exposes is that the budget is a function of four rates and two unit costs. Improving any one of them shrinks the budget more reliably than negotiating with an agency ever will. I go deeper on the channel side of the maths in the B2B lead generation post.
The payback test
A budget that produces pipeline can still be a bad budget if the cash comes back too slowly. The test is CAC payback: how many months of gross margin from a new customer it takes to recover what that customer cost to acquire. The formula and benchmarks are in the metrics post; the planning rule is simpler than the formula.
Take the budget from the exercise above, add the sales cost attributable to new business, and divide by the number of new customers. That is your planned fully loaded CAC. Divide it by monthly gross margin per customer and you have planned payback in months. If that number is longer than the runway you are willing to fund it with, the budget is not capital-efficient, however good the channel mix looks on paper.
Planned payback also tells you how to pace the spend. Nine months of payback can be spent evenly across the year. Twenty months should go out in tranches with a checkpoint after each, because you cannot afford to be wrong for a full year at that speed.
Two more checks the board will ask about
Burn multiple is net cash burned divided by net new ARR added over the same period. Burn €1,500,000 to add €1,000,000 of net new ARR and your burn multiple is 1.5. Lower is better. It moves with every line in the budget, which makes it the fastest way to test a proposed increase in marketing spend: model the extra spend, model the extra ARR it should produce, and see whether the multiple improves or gets worse.
LTV:CAC compares the gross-margin lifetime value of a customer to what they cost to acquire. The floor most people use is three to one. For budgeting it sets a ceiling on what you can afford to pay per customer: if lifetime value is €90,000, a plan whose fully loaded CAC lands above €30,000 is buying growth the company will never recover. Both are defined properly in the metrics guide.
How to split it across channels
Fund what already produces opportunities before you fund anything new. A channel with a known cost per qualified opportunity is the lowest-risk home for the next euro. A new channel is an experiment and should be budgeted like one: fixed amount, fixed period, written kill criterion.
Ours, for the record. A LinkedIn message test where loss framing beat capex and ROI framing on every cut, nearly double the click-through rate. We kept spending because the test was winning. Five months later there were still zero conversions. A test can win and the channel can still lose. The kill criterion should have been opportunities by week four, not CTR by month five.
Two channels run properly beat five run partially. Splitting a post-seed budget five ways guarantees that none of the channels gets enough spend to produce a signal you can read, and you end the year knowing nothing. Pick the two that match how your buyer buys and what your deal size can support, and fund them for two quarters.
Reserve for the foundation first. Positioning, the ideal customer profile, the measurement layer and the sales handoff cost more time than money, and they raise the conversion rate of every euro spent afterwards. It is the least glamorous line in the budget and the highest-return one.
A note on post-seed specifically
Post-seed is where percentages mislead the most. Revenue is small, so 8% of ARR is not a budget, it is a rounding error, and the real spend is a multiple of it funded by the round. That is normal. What matters is that the spend is tied to a payback the next round will respect.
In practice that means keeping the experiment budget small and time-boxed, judging channels on opportunities and never on leads or impressions, and instrumenting the funnel before scaling spend. The difference between a 20% and a 30% opportunity rate is worth more than any channel optimisation you will ever run, and you cannot see it without the numbers. The post-seed marketing strategy post is mostly about that sequence.
Where budgets go wrong
Percent-of-revenue budgeting at a stage where revenue is not yet a meaningful base. Marketing and sales budgeted separately, so pipeline and capacity never match and one of them is always idle. No cost per qualified opportunity by channel, which turns every allocation into a guess. No kill criteria, so failed experiments run for a year out of politeness. And scaling spend before the funnel converts, which is paying to push more water through the same hole.
One account, Q1 2026. Visits came in at half the forecast and leads at under 40% of it. The whole marketing budget for the quarter was tools, nothing else. So the reallocation was hours, not euros. We stopped the paid test at zero opportunities and put the time into the webinar series and the existing CRM database. August was the best month of the year for closed deals, with no ads running. Whether one caused the other is exactly the question the attribution should have answered, and could not.
Frequently asked questions
What percentage of revenue should a B2B SaaS company spend on marketing?
SaaS Capital’s 2026 survey puts the median at 8% of ARR for private B2B SaaS companies, with equity-backed companies at roughly double. Use it as a sanity check. At post-seed, derive the budget from your revenue target and funnel rates instead, then test it against CAC payback.
What is a good CAC payback period for planning?
Shorter than the runway you are willing to fund it with. A lot of operators aim for under twelve months and treat anything much longer as a signal to spend in tranches with checkpoints. Benchmarks and the calculation are in the metrics guide.
How much of the budget should go to experiments?
A minority, fixed in advance, with a written kill criterion. The majority belongs behind channels with a known cost per qualified opportunity. If nothing has a known cost yet, the first budget is mostly about establishing one.
Annual or quarterly?
Plan the year, release by quarter. A quarterly checkpoint against opportunities sourced and planned payback lets you move money between channels while it still matters.
Build the plan before you fund it
If your marketing budget was set as a percentage or an educated guess, run the backwards maths before next quarter starts. It takes an afternoon and it usually changes the number. Piquiyo builds capital-efficient GTM plans for post-seed B2B SaaS teams: the funnel model, the channel allocation and the measurement to check it against. Book a free growth audit and we will show you what your budget should be and how fast it should pay back.