B2B marketing budget allocation calculator: Split demand gen spend by pipeline per dollar

To allocate a B2B marketing budget with a calculator, rank every channel by pipeline per dollar, fund the leaders until their marginal returns fall, and set aside a demand creation reserve before the ranking runs. Pipeline per dollar equals (1 ÷ cost per lead) × lead-to-opportunity rate × average deal size. Gartner put budgets at 7.8% of revenue in 2026, up 1.3% since 2025, so we think growth has to come from moving existing dollars.

How to allocate a B2B marketing budget across channels from your own numbers

Rank every channel by pipeline per dollar, fund the leaders until their returns start to fall, and hold a fixed reserve for demand creation that you judge on its own terms. That is the whole method. The rest of this piece shows the math.

The context makes the method necessary. Gartner put the average marketing budget at 7.8% of company revenue in 2026, with budgets up 1.3% since 2025. Its 2025 survey found digital channels took 61.1% of total marketing spend. With little new money arriving, we think most pipeline growth has to come from moving existing dollars. Gartner's own advice for 2026 was to shift from incremental optimization to "deliberate reallocation".

The four steps

1. Pull spend and sourced pipeline by channel for the last two to four quarters. Use one attribution rule across all channels so the ratios compare.
2. Divide pipeline by spend. This is pipeline per dollar, the ranking key.
3. Find each channel's cap. Split recent spend into increments and measure what the last increment returned. When marginal pipeline per dollar drops below the average of the next channel down, stop adding money.
4. Set the reserve first. Take a fixed share off the top for demand creation before ranking, because those programs feed the demand capture channels later and will always lose a same-quarter ranking.

Worked example

A team with $1,000,000 for the quarter sets aside 15% ($150,000) as the creation reserve, an example share, then ranks the rest using last quarter's results.

Paid search leads, but its last $50,000 returned $175,000, a marginal 3.5. That still beats syndication's 3.0, so search can grow a little before it hits its cap. Events sit last at 2.0, which makes them the first source of dollars to move.

The ranking tells you where to look, not what is causal. Confirm the top channels with incrementality tests before you shift large sums. Then work the pipeline target backward with the reverse funnel calculator to check the new mix covers it.

What pipeline per dollar is and how the calculator computes it

Pipeline per dollar = (1 ÷ cost per lead) × lead-to-opportunity rate × average deal size

The first term turns a dollar into leads. The second turns leads into opportunities. The third turns opportunities into pipeline value. The output is pipeline value, not revenue. To get closed revenue, apply your win rate.

The inputs, defined

Use cohorts rather than calendar months where your sales cycle is long. If your stage definitions are unclear, fix the marketing qualified lead (MQL) handoff first. For outside reference rates, see funnel conversion benchmarks.

A worked example

These figures are illustrative.

Paid search: (1 ÷ 200) × 0.10 × 30,000 = $15. Syndication has the cheaper lead and still returns half the pipeline per dollar. Ranked by CPL alone, the split would run the wrong way.

How this differs from the two common alternatives

Allocation by revenue percentage starts with a total and divides it. Calculations of that kind often rely on "arbitrary percentages of revenue (typically 5-15%)", according to Catch Digital. That number tells you how much to spend. It tells you nothing about where.

The unit economics ceiling works from the customer. Maximum customer acquisition cost (CAC) = customer lifetime value × margin. In Catch Digital's example, a $50,000 lifetime value at a 30% margin gives a maximum CAC of $15,000. That figure is a break-even limit, not an allocation rule.

We think the two methods answer different questions. The ceiling says how much one customer can cost. Pipeline per dollar ranks channels against each other. The calculator uses both.

What the calculator returns

For each channel you enter spend, CPL, lead-to-opportunity rate, average deal size and win rate. For the account as a whole you enter lifetime value and margin.

It returns four things:

• Pipeline per dollar for each channel.
• Pipeline at your proposed spend.
• A split weighted by pipeline per dollar.
• Implied CAC per channel, calculated as CPL ÷ (lead-to-opp rate × win rate).

A channel is flagged when its implied CAC exceeds the ceiling, however good its pipeline per dollar looks. Pair the result with CAC payback and the reverse funnel calculator to check the volume you need.

How to compare channels with different lead quality

In our reading, cost per lead (CPL) is the number most channel reports lead with, and it says the least about pipeline. The same audience can produce very different CPLs depending on the form, the offer and the seniority you target. Each of those choices also changes how many leads become opportunities.

What moves CPL before quality enters the picture

• Form format. On LinkedIn, Lead Gen Forms typically deliver a CPL of $45-$95 across most B2B industries, while external landing pages typically run at $65-$140 CPL. The same source says the forms' cost advantage comes "often at the expense of lead quality."
• Offer type. Content downloads run $50-$100, webinar registrations $75-$150, demo requests $150-$300, and contact sales forms $100-$200.
• Audience seniority. Stackmatix benchmarks assume targeting of director-level and above in North America. Targeting individual contributors reduces CPL by 30-40%, and targeting the C-suite raises it by 40-60%.

A one-click form fill on a report and a demo request from a vice president are different units. Comparing their CPLs compares two different products.

Divide by the rate that matters

Stackmatix gives a worked case. A $50 CPL that converts at 5% to SQL costs $1,000 per SQL, and a $120 CPL that converts at 20% costs $600. A sales qualified lead (SQL) is the stage where sales accepts the lead; see our SQL definition.

Ranked on CPL, campaign A wins. Ranked on cost per SQL, campaign B does.

The measurement steps

1. Tag every lead at creation with channel, format, offer and seniority band, so each combination can be reported on its own.
2. Measure the lead-to-opportunity rate for each combination on cohorts old enough to have progressed. Compare against funnel conversion benchmarks only as a sanity check.
3. Compute cost per opportunity as spend divided by opportunities, then pipeline per dollar from that figure.
4. Re-measure the rate whenever the offer or form changes. The old rate no longer applies.

Step 4 matters because the rate moves with the offer. In one Stackmatix case, a cybersecurity company started at an initial CPL: $165 (Lead Gen Form) with a lead-to-SQL rate: 18%. After it switched to an interactive assessment and made other changes, it reached a final CPL: $98 (41% reduction) and a lead-to-SQL rate: 22%. In our reading, a calculator that kept the old 18% would have understated that channel. If you score leads, keep lead scoring consistent across channels so the rates stay comparable.

What LinkedIn and Google Ads leads cost in 2026 benchmarks

Published benchmarks give you a starting row for each channel before your own data builds up. Its figures are spend-weighted aggregates, not averages of advertiser averages, so large spenders pull each number toward their results.

Across the Metadata sample, cost per lead ran from $138 on Instagram to $524 on Google Ads.

All rows come from Metadata's 2026 benchmark reference. CTR is click-through rate, CPC is cost per click and CPM is cost per thousand impressions.

Why CPM and CTR do not travel across channel types

CPM follows from the other two columns. CPM equals CPC times CTR times 1,000. For Google Ads: $9.76 × 0.0633 × 1,000 comes to about $618. For LinkedIn: $9.39 × 0.0067 × 1,000 comes to about $63. In our reading, the gap reflects what an impression is on each channel, not how efficient the channel is.

A search impression appears after someone types a query. A feed impression interrupts a scroll. Metadata's guidance is to compare CPM within a channel type, and compare cost per lead across them. On CTR it says to judge a channel on cost per lead, and judge creative on CTR within its own channel. In the calculator, that means CTR and CPM stay on the creative report and never feed the allocation.

Google Ads click prices outside B2B samples

Cross-industry search data runs lower per click. LocaliQ reported that the average CPC for search advertising across all industries in 2026 is $5.42, with Business Services | $5.87.

In our reading, the gap between $5.42 and the B2B sample's $9.76 comes from different advertiser mixes. Pick one dataset for a planning row and keep it. Mixing a cross-industry CPC with a B2B click-to-lead rate produces a cost per lead that no real account would report.

How to use these rows in the calculator

1. Enter the benchmark cost per lead only for channels where you have under one quarter of your own data.
2. Apply your own lead-to-opportunity rates from funnel conversion benchmarks or your CRM, since benchmark leads carry no quality signal.
3. Replace each benchmark row with your account's numbers once they exist.

In our reading, Google Ads' high cost per lead reflects demand capture from people who are already searching. The calculator judges that cost against the pipeline those leads produce, not against the lead price on social channels.

Where returns fall and how to set a capacity limit per channel

Every dashboard reports average return: total pipeline divided by total spend. The budget decision needs marginal return, the pipeline the next dollar would add. Bell Statistics calls marginal return the only number relevant to a budget decision, and notes that on a saturating curve the marginal is always below the average.

The gap can be large. In the same glossary, a channel showing a 4.0 average return can easily have a marginal return under 1.0. Its worked example uses paid social, which had a half-saturation point of £180k a week at £320k of weekly spend. The dashboard showed a 3.4 average return. The last pound earned 1.15, falling to 0.82 once £80k was added. The same money elsewhere earns about £2.10.

The allocation rule follows from this. The optimum equalises marginal returns across channels. It does not put everything into the top-ranked channel.

Setting a capacity limit, step by step

1. Estimate pipeline per dollar for each tranche, not for the channel as a whole.
2. Fund the single highest tranche across all channels, then the next highest, until the budget runs out.
3. A channel's capacity limit is the last tranche funded before its next one falls below another channel's next tranche.

Where the tranche estimates come from

The curve is only trustworthy over spend you have actually run. Per the glossary, a curve can only be identified where spend actually varied. If a channel has sat at one spend level for a year, its higher tranches are guesses. Vary spend on purpose, with flighting or a geo holdout, before you raise a cap. Read the result as incrementality, not as attributed pipeline. We set caps at the edge of observed spend and move them only after a test.

How much budget to reserve for demand creation

Pipeline per dollar ranks channels on what they produced inside the measurement window. Demand creation does most of its work before that window opens, so a ranking run on in-quarter pipeline will keep cutting it. The fix is to set the demand creation money aside first and run the ranking only on what is left. The difference is laid out in demand creation vs demand capture.

The 70/20/10 split as a starting structure

Digital Applied describes the 70/20/10 framework as a split of budget "into core performance, emerging growth, and experimentation." Its core bucket holds channels "that have returned positive CAC over at least two consecutive quarters with enough volume to read signal." CAC here means customer acquisition cost. The same guide calls the experimentation bucket the one to protect when the chief financial officer asks for mid-year cuts.

Here is the split applied to an illustrative $600,000 quarter.

Only the $420,000 goes into the calculator's ranking.

The reserve-fund pattern

Improvado reported that it analyzed $480 million in marketing spend across 150 enterprises in 2024. Top-quartile return on investment (ROI) performers in that set allocated 18% of budgets to reserve funds for mid-year reallocation, against 3% for the bottom quartile. That reserve is for reallocation, not a demand creation line. We read it as the same discipline: money held outside the locked plan. The figures show a pattern, not proof that the reserve caused the returns.

One of Improvado's anonymized failure cases, a startup in a new category, had over-allocated to demand capture when no one was searching for the solution yet. In our reading, a quarterly pipeline ranking would not have shown that growth in the first quarter.

Our view: fund the reserve before the ranking runs

1. Start from the quarter total.
2. Remove the growth and bet lines, plus any mid-year reallocation reserve you hold. Write a review date next to each.
3. Run the pipeline per dollar ranking on the core remainder only.
4. Judge the reserved lines at their review date on their own test, such as branded search lift or a holdout, not on in-quarter demand capture numbers.

If the reserve enters the ranking, it competes against channels with a measurement window built for them, and it will lose every quarter. Setting it first keeps the ranking honest about what it can see.

Worked example: a $300,000 quarter allocated by pipeline per dollar

Here is the calculator run on a $300,000 quarter with five channels. The inputs are illustrative, so the logic matters more than the figures. Swap in your own and the steps stay the same.

The chart shows how the worked example splits a $300,000 quarter across the five channels and the 20% demand creation reserve.

Step 1: take the reserve off the top. The demand creation reserve is 20%, so $60,000 is set aside before any ranking. That leaves $240,000 to allocate by pipeline per dollar.

Step 2: rank channels by pipeline per dollar. Pipeline per dollar is lead-to-opportunity rate times average deal size, divided by cost per lead (CPL). Every channel here uses a $30,000 deal. Review sites come first at $15.00, then paid search at $14.40 and paid social at $10.00. Events and content syndication tie at $7.50.

Step 3: fill each channel to its cap, in rank order. Review sites cap at 100 leads, which costs $30,000. Paid search caps at 400 leads, which costs $100,000. That uses $130,000 of the $240,000. The remaining $110,000 goes to paid social. Its cap of 800 leads would cost $144,000, so paid social is not capped and keeps $34,000 of headroom. Events and content syndication get nothing because the money runs out first.

The reserve is not counted here because its return shows up later, outside this quarter's lead math.

Two checks before you commit. If events or content syndication sit just below paid social, a small change in CPL could reorder the list, so test those inputs first. If paid social's real cap is lower than 800 leads, the leftover money moves down to the $7.50 channels.

Input template

Copy this, replace the values with your trailing-quarter numbers, and keep one row per channel.

channel,cost_per_lead,lead_to_opp_rate,avg_deal_size,lead_cap
Review sites,300,0.15,30000,100
Paid search,250,0.12,30000,400
Paid social,180,0.06,30000,800
Events,400,0.10,30000,150
Content syndication,80,0.02,30000,1000

Rerun it every quarter. Gartner's 2026 guidance on CMO spend was to move to always-on, portfolio-based funding that shifts with performance. We think a quarterly rerun from fresh inputs is a practical way to do that.

What this page does not cover

This page does not cover how to choose an attribution model, how to design and read an incrementality test in detail, or how to set the total size of the marketing budget. Its figures are illustrative or drawn from published benchmarks, and it is not financial advice.

Related on Take the Market: Pipeline coverage calculator: the ratio you need from your own win rate and slippage; Demand generation calculator: reverse funnel math from a revenue target to visitors, leads, MQLs, SQLs and opportuniti; MQL to SQL conversion rate benchmarks and B2B funnel conversion rates by stage.

Where the defaults come from