Skip to content

Chapter 10 · Stage 04

The MQL Prediction Model

Your CFO asks how many MQLs this content will bring. Answer before a single page goes live, with a forecast built from demand, rankings and your own funnel.

11 min readUpdated By Baba HausmannOn the CMO and Head of Growth paths

CMOs, Heads of Growth and SEO leads Walk away with: a pipeline forecast you can defend to your CFO
  • The model connects content investment to a specific number of MQLs per month before you publish a single page
  • Three scenarios (conservative, baseline, optimistic) produce a range, not a point estimate
  • Every assumption is visible, debatable, and adjustable
  • Run it across your candidate segments and build the one with the most MQLs per page first
  • Forecasting changes the organic program from reporting on weather to speaking in capital allocation language

Most organic programs can’t answer the simplest question a CFO will ask: “If we invest in this content, how many leads will it produce?”

They can show traffic projections. They can wave at keyword volumes. But connecting a proposed content investment to a specific number of marketing-qualified leads per month? Most SEO teams go quiet. The conversation stalls, the budget gets cut, and organic stays an experiment instead of becoming infrastructure. SEO as a Revenue Channel covers why most programs end up reporting traffic instead of pipeline.

The MQL Prediction Model exists to answer that question before you publish a single page.

It’s not a machine learning black box. It’s not a guarantee of rankings. It’s a structured forecast that chains together search demand, realistic ranking assumptions, and your actual funnel conversion data to produce a time-based lead projection with scenario ranges so you can plan honestly.

When you walk into a budget meeting and say “organic traffic grew 30% last quarter,” you’re reporting on weather. Interesting, not actionable.

When you walk in and say “if we build the healthcare segment cluster, 9 pages, the model projects 35 to 55 MQLs per month from that segment within 8 months, with a cost per lead that drops below your current paid CAC of €280 within the first year,” you’re speaking in capital allocation language. That’s a business case, not an SEO report.

Forecasting changes the organic program in three ways.

It justifies investment before results arrive. Content takes 4 to 8 months to mature. Without a forecast, you’re asking leadership to fund a channel on faith. With one, you’re asking them to fund a channel with explicit assumptions they can scrutinize and approve.

It forces honest prioritization. When you model MQL output per segment, you quickly see that some segments produce 5x the leads of others at the same production cost. That changes what you build first.

It creates accountability in both directions. If the forecast says 40 MQLs by month 6 and you’re at 15, you have a specific conversation. Did traffic underperform the assumption? Did conversion underperform? Did we publish late? That’s a calibration discussion, not a blame conversation.

The model works in six steps. Each takes the result of the step above and applies one assumption. Every assumption is visible, debatable, and adjustable. That’s the point.

Search Demand Volume Click-Through CTR Sessions Traffic Odds Win × Ramp Conversion MQL % Scenarios MQLs
The six-layer MQL prediction model
  1. Search demand: monthly search volume for each planned page, adjusted for seasonality.
  2. Click-through rate: the share of searchers who click at your target position, lower where AI Overviews take the click.
  3. Sessions: volume times click-through rate, plus traffic from related queries.
  4. Odds: your win rate (how often pages like this actually reach their target position, measured from your own Search Console) times the maturity ramp (new pages bring 5% of their eventual traffic in month 1, 25% in month 3, all of it from month 6).
  5. Conversion: the share of sessions that become MQLs, per funnel stage.
  6. Scenarios: conservative, baseline and optimistic win rates, so you present a range instead of a falsely precise number.

The odds step is adapted from Dagan Price’s BrightonSEO talk “The SEO ROI Formula” (October 2026). Most forecasts quietly assume every page reaches its target. This one prices the chance that it doesn’t.

The output is a monthly MQL forecast per scenario, for the whole program or per segment. How to fill in each step with your own data is in The Forecast Spreadsheet.

Say you’re a B2B SaaS company selling procurement software, and you’ve identified the construction industry as a priority ABM segment.

Here’s how the model works for one page in that cluster.

Your pillar page targets “construction procurement software” with 480 monthly searches. You’re forecasting a position 4 ranking within 6 months (realistic for a focused page in a niche with limited competition). At position 4, your calibrated CTR is roughly 8%. That gives you about 38 sessions per month at steady state.

But the page won’t hit steady state immediately. The maturity ramp says month 1 delivers 5% of that, month 3 delivers 25%, and month 6 reaches full run-rate. So month 3 looks like roughly 10 sessions, not 38.

Your MOFU MQL session rate is 4.5% (calibrated from existing content performance). A secondary keyword multiplier of 1.4x accounts for related queries. At month 6, that’s approximately 2.4 MQLs per month (38 sessions x 1.4 secondary keywords x 4.5% MQL rate) from this single page.

That sounds small. But multiply it across the full cluster.

MQL session rates by funnel stage (applied to all pages in that stage):

  • BOFU: 7.2%
  • MOFU: 4.5%
  • TOFU: 1.8%

Most companies don’t have per-page MQL rates, and you don’t need them to start. Instead, segment your organic conversion data by funnel stage and apply those rates across all pages in that stage. As your tracking matures, you can refine rates per page or per content type. Calibrate quarterly as actuals come in.

PageFunnelVolumeTarget pos.SessionsMQLs/mo
Construction procurement softwareMOFU4804532.4
Construction procurement solutionsBOFU3203423.0
[Product] vs [Competitor] constructionBOFU2102382.7
Construction procurement ROI calculatorBOFU1703241.7
How to streamline construction purchasingMOFU3905381.7
Construction material cost tracking guideMOFU2605251.1
Procurement challenges in construction 2026TOFU7206541.0
Construction supply chain management trendsTOFU5807360.6
What is e-procurement in constructionTOFU4405430.8
Total (construction segment)—3,570—35315.0

At steady state (month 6+), if every page reaches its target position. Sessions include the secondary keyword multiplier. MQL rates are per funnel stage, not per page. The scenarios then multiply the gain by your win rate, so the baseline forecast you present is lower than this table.

Nine pages across the construction segment, each contributing 1 to 4 MQLs depending on funnel position and volume, can produce about 15 MQLs per month from one segment. At an average deal size of €45K and a 20% close rate, that’s about €135K of new business every month from nine pages.

No deal size to work with, or finance disputes it? Borrow a lead value from paid: what you pay per paid lead × the share of those leads sales closes. One question to sales (“of the leads we send you, how many close?”) is enough. The number comes from your own ad spend, so nobody can dismiss it as an SEO invention. It’s the fallback: a real deal size and close rate always win.

Now model three segments. Organic isn’t a cost center. It’s a growth engine with a quantifiable return.

The same model answers a second question: you have three segments, budget for one cluster at a time, and need to pick. Gut feel, the loudest stakeholder or whichever segment sales complained about this quarter all produce an answer. None of them survive “why healthcare first and not logistics?”

Run the model for each candidate segment with the same inputs: a keyword map for the planned cluster (8 to 10 pages, volume per page), a realistic target position per page, and your funnel-stage conversion rates. Then compare.

SegmentPagesTotal volumeProjected MQLs (baseline, month 8)MQLs per page
Healthcare94,200182.0
Logistics92,800222.4
Construction93,570151.7

Illustration, not client data. Healthcare and logistics are invented for comparison; construction is the example above.

The gut-feel answer is healthcare, because it has the most search volume. The model says logistics, because its keywords are more commercial (higher CTR and conversion) and competition is thinner (more realistic position targets).

Search volume is a poor proxy for pipeline. The model forces you to look at the full chain from impressions to MQLs, and the ranking often surprises you.

MQLs per page at steady state is the number to rank on. It’s pipeline per unit of effort. A high figure means commercial intent is concentrated, ranking targets are realistic and conversion is proven. A low figure means informational keywords, a crowded results page, or a funnel for that audience that hasn’t been proven yet.

The maturity curve is why the order matters. Build the highest-yield segment first and each cluster ramps while the next is being built:

  • Months 1 to 2: publish cluster 1 (logistics in this illustration)
  • Months 3 to 4: publish cluster 2; cluster 1 is at a quarter to half of its steady state and producing its first leads
  • Months 5 to 6: publish cluster 3; cluster 1 is at full run-rate, cluster 2 is ramping
  • Month 8 onwards: all three at or near steady state

Publishing all three at once gets you the same end result, with a longer stretch of low output while everything ramps in parallel.

Being clear about boundaries builds more trust than overselling.

It doesn’t promise rankings. The target position is a planning target. The win rate prices the chance of reaching it, from your own history, not from a promise about Google’s behavior.

It doesn’t model individual CRO changes. Adding a trust badge, improving page speed, or rewriting a CTA might improve conversion. But attributing a specific MQL increment to each tactical change creates false precision. The model holds your baseline conversion rates. Tactical improvements show up as actual performance exceeding the forecast.

It doesn’t replace CRM data. The forecast tells you what organic should produce. Your CRM tells you what it did produce. Comparing the two is how you calibrate the model over time and get sharper with each quarter.

It isn’t precise, and doesn’t need to be. If you’ve never run organic content for a segment, your conversion rate is an informed guess. The value is comparability: the same method across three segments gives you a defensible ranking even with rough inputs. By quarter 2 the assumptions are calibrated; by quarter 3 the forecast is genuinely predictive. The teams that never build a model aren’t more honest. They make the same guesses without writing them down.

It doesn’t run on autopilot. The assumptions need human judgment: which position to target, which conversion rates to use, how aggressively to set scenarios. The model is a thinking tool, not a magic spreadsheet.

Once the forecast is running, you compare it with what actually happened. If a segment falls behind, the gap is either in traffic (rankings, internal links, authority) or in conversion (landing pages, intent match). The model doesn’t tell you what to fix. It tells you where to look first. The Quarterly Review covers how to run that comparison and present it to your board.

Most organic programs work in audit mode: here’s what’s broken. A forecast puts you in planning mode: here’s where we’re going, where we are, and the work that closes the gap.

The most powerful use of the MQL Prediction Model isn’t the number itself. It’s the conversation the number enables.

When you can show a CFO that investing €30K in content production for two segments is projected to produce 40 to 70 MQLs per month within 8 months, with explicit assumptions they can challenge, organic moves from “nice to have” to “funded growth channel”.

Go one step further and price each position. Your click-through curve isn’t flat, so each step up page one is worth a different amount: “8 to 7 adds X MQLs a month, 5 to 4 adds Y, worth €Z in pipeline a year.” A CFO can’t buy “visibility”. They can buy the next position, at a stated price. It prices the downside too: when a cluster slips a position, the loss has a number, and a budget cut reads as written-off pipeline rather than a saving. In Dagan Price’s example, the step from 5 to 4 was worth nearly four times the step from 8 to 7, which is why pushing pages already near the top usually beats pulling up the tail.

And when the first quarter of actual data comes in and you can show forecast vs. actuals, calibrate the assumptions, and present an updated projection, you’ve built something most organic programs never achieve: a credible, evolving business case that compounds trust with every review cycle.

This is how organic earns a seat at the planning table. Not by reporting traffic, but by forecasting revenue.

Once the forecast model is running for your first segment, three things need to happen next.

Compare forecast to actuals every quarter. The model’s value compounds when you calibrate it against real data. The quarterly review cadence covers the board-level presentation: what to show your CFO, how to explain gaps, and how to use the review to earn expanded budget.

Understand the channel economics. The forecast gives you projected MQLs. The channel economics comparison puts those MQLs in context: how organic CAC compares to paid CAC per segment, when the crossover happens, and why the portfolio allocation should shift over time.

Decide whether to restructure at all. If your current program is organized by topic, ABM SEO vs. Traditional SEO covers when moving to segments is worth it and when it isn’t. The full CMO sequence is on the CMO overview.

Hand the spreadsheet to your SEO lead. The mechanics of building and maintaining the model are in the SEO Lead version below.

Other perspective MQL Forecasting: The Spreadsheet How to pull GSC data, calibrate the CTR curve, wire the calculations, and recalibrate quarterly. Written for SEO Leads.
Take this away: 3 asks for your AI, your CFO and your sales lead

Your AI

Your AI can run the six steps with your numbers.

See the prompt
Build a lead forecast for one content cluster, using this six-step method. Inputs I'll give you: a list of planned pages with their main keyword, monthly search volume, funnel stage (decision, consideration or awareness) and target ranking position; my click-through rate per position if I have it; my conversion rate from visit to lead per funnel stage; my win rate (the share of pages that reach their target position within a year). Steps: (1) monthly searches per page; (2) click-through rate at the target position; (3) visits = searches × click-through rate, × 1.0 to 1.3 for decision pages, 1.3 to 1.8 for consideration pages, 2.0 to 3.0 for awareness pages, to count related searches; (4) apply the ramp, new pages reach 5%, 10%, 25%, 45%, 70% of full traffic in months 1 to 5 and 100% from month 6, and multiply by the win rate; (5) leads = visits × conversion rate for that stage; (6) three scenarios: win rate minus 20 points with conversion × 0.8, my win rate, and win rate plus 20 points with conversion × 1.2. Give me a table per page at month 6, a month-by-month total for months 1 to 12 per scenario, and leads per page. Ask me for each input before you start. If I say I don't have one, use these starting values and mark them "assumed": win rate 25%; conversion 7.2% decision, 4.5% consideration, 1.8% awareness.

Ask your CFO

You need: average deal size per segment, and what a paid lead costs today.

Hi [name], I'm putting together a lead forecast for our organic search before I ask for any budget, and I want it built on our numbers rather than industry averages. Could you share two figures: our average first-year deal value for [segment], and what we currently pay per lead from paid campaigns? Rough is completely fine. I'll show you every assumption in the model before it goes anywhere, so you can challenge it.

Ask your sales lead

You need: how many leads become customers.

Hi [name], quick one for a forecast I'm building: of the leads marketing sends you, roughly how many end up closing? If you can split it by segment, even better, but one overall number already helps a lot. Last year's figure is fine. Thanks, this saves me from guessing.

Get notified when new chapters publish

You're on the CMO path. Step 3 of 5.

0 of 5 read

  1. Next: SEO as a Revenue Channel
  2. Not read: ABM SEO vs. Traditional SEO
  3. Reading now: Forecasting Pipeline
  4. Not read: The Quarterly Review
  5. Not read: Channel Economics

Switch path: Head of Growth or SEO lead