Harvest Season for Budgets: What to Lock In Before January

A lead generation budget should start from the cost of a signed job, not from last year's figure plus a percentage. Marketing budgets now sit at 9.0% of company revenues, so the pot is finite. Decide what a signed job is worth, then work backwards to what you can spend to win one.

November is when most of next year's commitments get made, and it is the last month where changing your mind is free. Contracts signed in December are harder to unwind, and January spending starts against whatever assumptions survived the autumn. That makes this the right moment to check whether the number you are about to commit was built on anything measured.

Why should your lead generation budget start with cost per signed job?

Most lead budgets are set per lead. That is the wrong unit, because a lead is an input and a signed job is the outcome you are buying. Two sources can deliver leads at the same price while one of them converts at twice the rate, and a budget expressed in cost per lead cannot tell them apart. The year is then priced on a number that does not predict revenue.

Cost per signed job is the figure that survives contact with reality. Take what a source cost you over a full season, divide it by the jobs that source actually produced, and you have a number you can multiply by next year's job target. It absorbs close rate, lead quality, response time and seasonality in one figure, which is precisely why it is harder to game than cost per lead.

The gap between the two units matters more when budgets are tight, and they are. Overall marketing spending grew just 1.7% over the prior 12 months, the weakest rate since 2021, which means very few teams will solve a mispriced budget by adding money to it mid-year. The allocation you set now is close to the allocation you will live with.

There is a structural bias worth naming before you allocate. Acquisition budgets currently run 26% larger than retention budgets, even though retention delivers stronger performance outcomes. If your plan repeats that split without testing it, you are inheriting someone else's assumption rather than making your own.

How much are companies actually spending on marketing?

Benchmarks are useful as a sanity check on the shape of a budget, not as a target to copy. A contractor with a short sales cycle and a software firm with an eighteen-month one will land in very different places, and both can be right. Use these figures to ask why you differ, not to converge.

Two numbers in that table deserve to be read together. Total spending is close to flat while digital spending grew 8.2%, which means most teams are not growing their budget, they are moving it. If your plan shows new digital money without a corresponding cut somewhere, the cut will happen anyway, later, and someone else will choose it.

Which channels are getting more expensive to buy?

Media prices do not rise uniformly, and a budget built on an average rise will be wrong in both directions at once. The published revenue figures are a reasonable proxy for where demand and therefore cost pressure is concentrated.

The detail that matters for a lead budget is search. It grew 11% year over year, slowing from 15.9% the year before, while social grew three times faster. Slower category growth does not mean cheaper clicks in a high-intent vertical, where a handful of buyers compete for the same narrow set of queries. Plan search on your own auction data and treat the category figure as background only.

A practical rule: build next year's model with a cost increase per channel rather than one blended figure, and make each increase a separate line you can revisit. A single blended assumption hides which channel broke the plan.

What should you commit for the full year, and what should stay flexible?

Split the budget into three tranches rather than twelve equal months. The split is what lets you respond to a bad quarter without reopening the whole plan.

• Committed. The spend that must run all year for the channel to work at all: the baseline search presence, the content that compounds, the tracking and attribution that tell you anything. Commit this early and do not touch it, because switching it off costs more than it saves. It is also the part worth negotiating on an annual basis, since volume commitments usually buy better terms than month-to-month buying.
• Variable. Spend tied directly to demand, which rises and falls with the season. This is the tranche you size from last year's measured demand curve, not from a flat monthly average.
• Reserve. An unallocated block, held back deliberately. Teams that spend their reserve in January did not have one. Its job is to fund the source that is quietly outperforming by March, and to absorb the cost increase you did not forecast.

How large should the reserve be? Large enough that losing it would not stop the year, which in practice means a share you can state out loud and defend. The reason to hold it is not caution for its own sake: when profits fall short of expectations, 53.1% of companies focus on cutting expenses rather than growing revenue. A visible, unspent reserve is the cheapest thing to surrender in that conversation, and surrendering it protects the committed tranche that actually produces jobs.

How do you plan for seasonal swings in demand?

Most trades do not sell evenly across twelve months, and a budget that ignores that pattern overpays in the quiet months and runs out in the busy ones. Roofing, landscaping and pool work concentrate in warm months. Heating and water damage concentrate in cold ones. Insurance and education follow enrolment and renewal dates rather than weather.

Build the seasonal shape from your own booking history rather than from intuition. Count signed jobs by month across two or three full years, express each month as a share of the annual total, and let the variable tranche follow that curve. Two or three years matters because one unusual winter will otherwise set your assumptions for the whole plan.

Two adjustments are worth making to that curve. First, shift spending earlier than the demand peak, because the gap between an inquiry and a signed job is rarely zero: if jobs close three weeks after the inquiry, the money has to be in market three weeks before the month you want to fill. Second, do not zero out the quiet months entirely. Competitors withdraw in the off-season, which is often when the same inquiry is cheapest, and a channel that has been dark for a quarter does not restart at full strength.

The off-season is also when the committed tranche earns its keep. Content, organic visibility and anything that compounds should keep running while paid volume drops, because that is the work that lowers next year's cost per signed job. A plan for a full year of lead generation that treats winter as an off switch pays for the restart twice.

What does a budget built on signed jobs look like?

The arithmetic is deliberately simple, and that is the point. Start with the number of signed jobs you need. Divide your target revenue by average job value to get it, or take it straight from the operations plan if capacity is the constraint. Then apply your measured cost per signed job, by source, to get the spend each source has to carry.

Where the method earns its keep is in what it refuses. A source whose cost per signed job exceeds what a job is worth to you does not get a budget line, however cheap its leads look. A source you cannot measure to the signed job gets a test budget rather than an annual commitment, and the test has a date on it. A source that worked last year but whose volume has already flattened gets flagged now, not in June.

Then stress the plan in both directions before you sign anything. Ask what happens if cost per signed job rises by a tenth and nothing else changes, and whether the reserve covers it. Ask what happens if close rate falls because of slower response times rather than worse leads, which is the more common cause and the cheaper one to fix. Ask which single source, if it stopped delivering, would take the year with it. That last question is usually the most useful thing a budget review produces, and concentration is easier to correct in November than in May.

Which budgeting mistakes cost the most?

• Last year plus a percentage. It carries forward every mispriced source in the previous plan and compounds the error. Rebuild from signed jobs instead.
• One blended cost assumption. Channel costs moved very differently last year, from 9.8% growth in display to 32.6% in social. An average hides which line will break.
• No reserve. A fully allocated budget cannot respond to anything, good or bad, and is the first to be cut when results slip.
• Budgeting leads, reporting jobs. If the plan is in cost per lead and the board asks about signed jobs, every review becomes a translation exercise and the numbers never reconcile.
• Treating measurement as overhead. Attribution and tracking sit in the committed tranche for a reason. Without them, next year's budget is guesswork again, and the year after that too.
• Ignoring response time. A faster reply raises close rate without raising media cost, which lowers cost per signed job at no extra spend. Teams comparing digital lead generation channels often find the cheapest improvement is operational rather than budgetary.

The common thread is that each mistake substitutes a convenient number for a measured one. The budget is a forecast, and a forecast built on inputs you have never verified is a guess with a spreadsheet around it.

Frequently asked questions

What percentage of revenue should go to lead generation?

There is no single correct share, and the benchmark is a sanity check rather than a target. Marketing budgets overall sit at 9.0% of company revenues, with lead generation taking a portion of that. Work from what a signed job is worth and what it costs to win one, then compare the result against the benchmark to check it is not wildly out.

When should next year's lead generation budget be finalised?

Finalise the committed tranche before the calendar year starts, because annual commitments buy better terms than month-to-month buying and the baseline has to be running in January. Leave the variable tranche reviewable each quarter and the reserve unallocated until something earns it.

Should the budget be split evenly across twelve months?

No. An even split overpays in your quiet months and underfunds the peak. Size the variable tranche from your own signed-job history by month, and shift spending earlier than the demand peak by roughly the length of your sales cycle.

How do you budget for a channel with no track record?

Give it a test budget with a fixed end date and a single success measure, stated as cost per signed job rather than cost per lead. Fund the test from the reserve rather than from the committed tranche, so a failed test costs you an experiment rather than your baseline.

What should change in the plan if media costs rise mid-year?

Adjust the variable tranche and draw on the reserve before touching the committed baseline, since switching off the baseline costs more to restart than it saves. If the rise is structural rather than seasonal, rebuild the cost per signed job for that channel and decide whether it still clears the value of a job.