A quarterly lead generation plan

Most quarterly lead generation plans are twelve slides long and get opened twice, once at the kickoff and once during the postmortem. The version that survives contact with a real quarter fits on a single page, holds six blocks, and can be read out loud in four minutes.

That constraint is doing real work. A one page plan forces you to name a number per source, name the money behind it, and name the date you stop. Twelve slides let you avoid all three.

Here is the structure, the arithmetic behind each block, and the review cadence that makes the page mean something after week two.

What goes on the one page

Six blocks, in this order: goals by source, budget by source, three tests, kill criteria, dependencies, risks. No narrative, no market context, no restatement of the company strategy. Those live in the B2B SaaS go to market plan and nobody needs them twice.

The goals block carries two numbers per source, leads and opportunities, because a lead goal alone can be hit by a source that produces nothing a salesperson will accept. The budget block carries one number per source plus a total that ties to the finance plan. The tests block carries three hypotheses written as falsifiable sentences.

Kill criteria get their own block rather than a footnote inside the tests. Dependencies list work owned by people who do not report to you, with names against them. Risks list the two or three things that would break the quarter, written before they happen so nobody can claim afterwards that they were unforeseeable.

How to size each source without using ambition as an input

Start from each source’s own measured conversion rate over the last two quarters, not from the number you need. This is the difference between a plan and a wish, and it is the step most teams skip because the honest arithmetic usually says the target is short.

The order is fixed. Decide how many opportunities the quarter needs, allocate that opportunity count across sources based on what each has actually delivered, then divide each allocation by that source’s lead-to-opportunity rate to get its lead goal. Only then check whether the lead volume is achievable at the budget you have.

A worked example at 40,000 dollars ACV, needing 150 opportunities in the quarter:

Two things jump out of a table like this and neither is visible in a blended plan. Partner and referral is the cheapest opportunity source by a factor of six, and it is only being asked for 19 opportunities because supply is constrained rather than because economics are poor. Paid social costs nearly seven times what content costs per opportunity, which is defensible only if it reaches accounts the other sources cannot.

Now apply the haircut. Take the conversion rates you just used and cut them by 15 percent before you commit, because last quarter’s rates were measured on last quarter’s list quality, creative freshness and rep capacity, all of which decay. If the plan still works at the haircut rates, commit it. If it only works at last quarter’s best rates, you have written a forecast, not a plan.

For the upstream arithmetic connecting an ARR target to an opportunity count, the lead goal calculator runs the chain, and the broader SaaS lead generation hub covers what feeds each of these sources.

What total spend should be, and the 70 20 10 split

Private B2B SaaS companies spend a median near 8 percent of ARR on all of marketing, and recently funded growth stage companies commonly run 12 to 20 percent. That is the anchor. Lead generation programs, meaning the spend that is not headcount or brand, usually take 40 to 60 percent of the marketing total.

Funding status moves the number more than company size does. A bootstrapped 12 million dollar ARR company running at 6 percent has roughly 180,000 dollars of quarterly marketing spend. A Series B company at the same ARR running at 16 percent has 480,000 dollars. Same revenue, different plans, and both are defensible depending on what the board asked for.

Inside whatever total you land on, split the program spend three ways:

• 70 percent to proven sources. Anything with at least two quarters of measured cost per opportunity inside your acceptable band. This is the money that has to deliver the number.
• 20 percent to scaling sources. Sources that work at small volume and have not yet been pushed to the point where cost per opportunity breaks. This is where growth comes from next quarter.
• 10 percent to experiments. No proof required, and no expectation of contribution to this quarter’s goal.

The reasoning is about failure tolerance rather than optimism. If the entire experimental bucket returns zero, you lose 10 percent of program spend and still make the number, because the proven bucket was sized to carry it. If you instead run 40 percent experimental, one bad quarter of tests takes the whole plan down and the organisation stops letting you test at all.

The other honest limitation: this split assumes you have proven sources. A company in its first two quarters of lead generation has no proven bucket at all, and the correct allocation there is closer to 50 percent on the single most likely source and 50 percent spread across three tests. The demand generation budget calculator handles both cases, and SaaS lead generation strategies, ranked covers which sources tend to prove out fastest by motion.

Writing three tests that can actually fail

A test needs a hypothesis, a threshold, a date and a spend cap. Write it as one sentence with all four in it, and refuse to add a fourth test no matter how good the idea is.

Bad version: we will try LinkedIn document ads this quarter. Good version: we believe LinkedIn document ads targeted at the 400 accounts in our tier one list will produce opportunities under 1,200 dollars each; we will spend 18,000 dollars by 20 November, and if cost per opportunity exceeds 1,800 dollars the test stops.

Three is the limit for a reason. Each test consumes creative time, analytics setup, a landing page and roughly two hours a week of attention. Teams that run seven tests run all seven badly, read none of them cleanly, and finish the quarter unable to say which one worked.

One tradeoff worth stating plainly: disciplined testing is slower than it feels like it should be. Three tests a quarter is twelve a year, and most of them fail. That is the honest cost of not burning 30 percent of your budget on channels nobody ever formally evaluated.

Kill criteria, and why nothing ever dies without them

Kill criteria are a threshold, a date, an owner and a destination for the money. All four, written before the quarter starts. Plans without them never kill anything, which is precisely why underperforming channels survive for years inside otherwise competent marketing teams.

The mechanism is ordinary loss aversion plus the absence of a forcing function. In any given week, a weak channel looks marginal rather than terrible. Nobody wants to be the person who stops something that might have been about to turn around. Six weeks of marginal compounds into a quarter, four quarters compound into a line item that has been running since 2024 and that nobody can defend.

Write one kill criterion per test and one per proven source. Proven sources need them too, because the most expensive failures in lead generation are not bad experiments, they are good channels that quietly degraded while nobody had a stopping rule. A paid search program that ran at 900 dollars per opportunity in Q1 and 2,100 dollars in Q3 usually got there one week at a time.

An exception worth honouring: brand and community programs should not carry a cost per opportunity kill criterion at all, because the measurement lag makes the number meaningless inside a quarter. Give them a different rule, usually a qualitative review at two quarters plus a share-of-search or branded-query check, and keep them out of the 70 percent bucket.

Dependencies and risks, written down while they are still cheap

Dependencies are deliverables you need from people who do not report to you. Risks are the two or three things that would break the quarter. Both belong on the page because writing them down converts a future argument into a present conversation.

Typical dependency lines from real quarters: sales hiring two SDRs by 1 October, because the outbound opportunity target assumes four reps and you have two; product shipping the integration page template, because 340 of your content leads depend on it; RevOps fixing lead source attribution in Salesforce, because three of your six sources currently report through a single channel bucket.

Each line needs a name and a date. Not a team, a person. If the integration template slips by five weeks, you want to have written on 1 October that it was needed on 20 October, so the conversation in November is about what to do rather than about what was agreed.

Risks are shorter and more brutal. Two or three lines, each with an early warning signal you can actually observe:

• Our top outbound sequence relies on one data provider; if match rates fall below 60 percent, outbound opportunity volume drops by roughly a third within three weeks.
• Paid search cost per click rose 22 percent over the last two quarters as two competitors entered the auction; if it rises another 15 percent, paid search cost per opportunity crosses our kill line by mid quarter.
• One person writes all of our comparison pages; their planned leave in November removes four pages from the quarter.

If you outsource any part of the plan, dependencies get sharper, not softer. Hiring a SaaS lead generation agency covers the contract terms that make an external dependency observable, and evaluating SaaS lead generation companies covers the diligence that should happen before a provider appears on your plan at all.

The review cadence: four numbers weekly, six monthly

Weekly, review four numbers per source: spend to date, leads created, opportunities created, and cost per opportunity. Twenty minutes, standing, no slides. Monthly, add pipeline value, stage progression, the forecast for the quarter, and any kill decision that has come due.

The weekly four exist to catch the two failure modes early. A source producing leads but no opportunities has a quality problem, which spend will not fix. A source producing neither has a volume problem, which spend sometimes will.

Monthly adds the numbers that are too noisy to read weekly. Pipeline value by source, because opportunity counts hide a shift toward smaller deals. Stage two progression rate, because accepted opportunities that never advance are a qualification problem disguised as a volume win. And the quarter forecast, rebuilt from actuals rather than adjusted from the plan.

One thing not to do weekly: rebalance the budget. Moving money between sources every week guarantees that no source ever gets a clean read, and it converts the plan into a series of reactions. Rebalance at month boundaries, and only between the scaling and experimental buckets unless a kill criterion has fired.

What to do when a source misses by 30 percent at week four

Diagnose before you spend. A 30 percent miss is either a volume problem or a conversion problem, and the corrections point in opposite directions, so getting this wrong doubles the damage.

Volume problem: leads are down, but the leads that do arrive convert to opportunity at the usual rate. Causes are budget pacing, audience exhaustion, creative fatigue, or reduced rep capacity. This one is often fixable with money or with a creative refresh.

Conversion problem: lead volume is fine or even up, and opportunity creation has collapsed. Causes are targeting drift, an offer that attracts the wrong role, a broken routing rule, or a sales team that stopped working the source. Spending more here actively makes things worse, because you buy more of the leads that were not converting.

Symptom	Most likely cause	Correction	Do not
Leads down 30%, conversion flat	Pacing, fatigue or capacity	Refresh creative, raise bids, or add rep hours for four weeks	Change the targeting, which resets your baseline
Leads flat, opportunities down 30%	Targeting drift or wrong offer	Tighten ICP filters, change the offer, audit routing	Increase spend, which buys more unqualified leads
Both down 30%	Channel or market change	Check auction dynamics and competitor entry, then apply kill criteria	Wait another month hoping for reversion
Cost per opp up 30%, volume flat	Auction pressure or lower deal quality	Segment by campaign, cut the worst third of spend	Blanket budget cut across the whole source

Whatever the diagnosis, give the source exactly one fix cycle of four weeks with a single change. Changing three things at once means you learn nothing, and at week eight you are in the same conversation with less budget left. If the fix cycle does not restore the number, the kill criteria you wrote in week zero apply, and the remaining budget moves to whichever proven source has the lowest cost per opportunity.

The arithmetic of the correction matters too. A source that is 30 percent behind at week four needs to run roughly 20 percent above plan for the remaining eight weeks just to finish level, which is usually more than a fix cycle can deliver. Accept the partial miss on that source and decide where the shortfall gets made up, rather than pretending the recovery will be complete.

Write it this afternoon

Open the demand generation plan template, pull the last two quarters of lead-to-opportunity rates by source out of your CRM, and fill the goals block first. That is the part that takes real time, maybe two hours, because the CRM data will be messier than you expect.

Budget follows in twenty minutes once the goals are set. Tests take an hour if you write them properly with thresholds and dates. Kill criteria take ten minutes and save you a quarter of wasted spend. Dependencies and risks take fifteen minutes and one uncomfortable conversation with sales.

Then send the page to your VP of Sales and your CFO before the quarter starts, not after. If either of them disagrees with a number, you want that argument in week zero. For the strategy underneath the plan, B2B SaaS lead generation covers the source mix by company stage, and SaaS lead generation examples shows what these plans look like when they worked.