Lead Generation

Lead Qualification and Scoring: The MQL Debate and What the Evidence Supports

Lead qualification is where most service businesses quietly burn their pipeline. The dominant framework, score individual leads and pass marketing qualified leads (MQLs) to sales, has been under sustained attack from the analysts who once promoted it. Forrester's waterfall benchmarks show lead-centric MQL processes converting from inquiry to closed-won at under 1%, a failure rate above 99% (Forrester, 2022). Its research also finds over 80% of buying decisions are made by groups of more than three people, which individual-lead scoring structurally ignores. Meanwhile, Harvard Business Review's response timing research shows most firms lose qualified leads to slowness, not bad scoring (Oldroyd et al., 2011). This deep dive examines the MQL debate's evidence and builds a qualification approach sized for small teams.

Joshua Agonya Pi'Rwot

By Joshua Agonya Pi'Rwot

Founder, Business Growth Accelerator

Executive summary

Forrester benchmarks show lead-centric MQL processes convert under 1% from inquiry to closed-won. This research deep dive examines scoring accuracy, the buying group shift, and practical qualification for small teams.

Section 1

The five challenges at a glance

Qualification failures rarely look like failures; they look like busy pipelines. Leads are scored, thresholds are hit, handoffs occur, dashboards fill with MQLs, and twelve months later almost none of it became revenue. The research record explains why with unusual precision: scoring models are built on arbitrary point values nobody ever validated against actual purchases; the models are applied to individuals when committees of three to ten people make the decision; the resulting leads are processed too slowly to matter, with the average firm taking 42 hours to respond to a hand-raise; and the whole apparatus is disconnected from the conversations where sales actually qualifies and disqualifies. Forrester, Gartner, MarketingSherpa, and the Harvard Business Review response-time research each illuminate a different fracture in the system, and together they explain the sub-1% inquiry-to-revenue conversion Forrester documents for lead-centric processes. For small service teams the encouraging news is that every one of these fractures is cheaper to fix at low volume than at enterprise scale. The table below summarizes the five challenges examined in this deep dive, with root causes, the firms most exposed to each, and the strongest evidence base behind every row.

Section 2

Challenge 1: The sub-1% problem with MQL funnels

The sharpest evidence in the MQL debate comes from Forrester's own waterfall benchmarks. Terry Flaherty, the Forrester principal analyst behind its revenue process research, reports that the typical conversion rate from inquiry to closed-won in a lead-centric process built on MQLs is less than 1%: fewer than one won deal for every hundred people who raise a hand (Forrester, 2022). Forrester's conclusion is blunt: the cross-functional process converting early interest to revenue fails more than 99% of the time. The firm has run a multi-year campaign, the Goodbye MQL series, arguing for replacing lead-centric processes with opportunity-centric ones organized around buying groups (Forrester, 2022-2023). Two readings of the sub-1% figure matter for service firms. The pessimistic reading: most of what marketing calls a lead was never a potential buyer, and the MQL machinery manufactures false confidence. The constructive reading: the failure is concentrated in definitions and handoffs, not in demand itself, which is why Forrester's prescription is alignment around opportunities rather than abandoning qualification. The corroborating misalignment data is older but consistent: MarketingSherpa's benchmark research found 61% of B2B marketers send all leads directly to sales while only 27% of those leads turn out to be qualified (MarketingSherpa, 2011). For a 5-7 figure service business, the lesson is that adopting enterprise MQL tooling imports enterprise failure rates without enterprise volume to absorb them.

Section 3

Challenge 2: Lead scoring accuracy and the buying group blind spot

Forrester's critique goes beyond outcomes to the mechanics of scoring itself. In its revenue process research, point values assigned to profile attributes and engagement behaviors, and the thresholds that trigger MQL status, are described as typically based on guesses and random estimates without any true analysis of propensity to buy; a single individual downloading four whitepapers shows activity, not purchase intent (Forrester, 2022). The deeper structural problem is unit of analysis. Forrester's 2021 B2B Buying Survey found over 80% of buying decisions are made by a buying group of more than three people, and Gartner's parallel research puts complex-purchase committees at 6 to 10 decision makers, recently ranging up to 16 across four functions (Forrester, 2021; Gartner, 2024). Individual lead scoring is blind to the signal that actually correlates with purchase: multiple people from the same account engaging around the same problem. As Forrester notes, most scoring models respond with excitement to one binge-downloader while completely missing coordinated engagement across a group (Forrester, 2022). Forrester is equally skeptical of patching the model with AI-based lead scoring, which it has characterized as emphasis in the wrong place when the underlying unit, the individual lead, is wrong. The accuracy problem, in short, is not insufficient math; it is that the model scores the wrong thing, with weights nobody validated, against a threshold nobody tested.

Section 4

Challenge 3: Speed and capacity, the qualification variables small teams actually control

While analysts debate scoring architecture, the most actionable qualification research concerns timing. The Harvard Business Review study by Oldroyd, McElheran, and Elkington audited 2,241 US companies responding to web-generated leads and found that firms attempting contact within an hour were nearly seven times as likely to have a meaningful conversation with a decision maker, in qualification terms, as firms waiting even an hour longer, and more than sixty times as likely as those waiting 24 hours or more. Yet the average response time was 42 hours, and roughly a quarter of companies never responded at all (Oldroyd et al., HBR, 2011). Related lead response research associated with the same authors found that responding within five minutes versus thirty minutes raised contact rates by roughly two orders of magnitude. For a small service team, this reframes qualification entirely: before optimizing who to talk to, fix how fast anyone gets talked to. A founder-led firm with no coverage system is structurally a 42-hour responder, which means it is disqualifying its best leads by silence. The capacity insight follows: small teams do not need scoring to prioritize thousands of leads they do not have; they need explicit fit criteria applied in minutes, plus a response system, automated scheduling, instant alerts, and templated first-touches, that compresses time-to-conversation toward the research-backed window.

Section 5

Innovative solutions

Teams rebuilding qualification on the evidence converge on several patterns. The first is buying-group detection over lead scoring: instead of summing one person's points, the CRM flags accounts where two or more contacts engage within a window, the multi-signal pattern Forrester identifies as dramatically more predictive of purchase than individual activity (Forrester, 2022). For small firms this can be as simple as matching email domains across subscribers, webinar attendees, and site visitors. Second is explicit-criteria qualification: a short, written fit definition, size, problem, budget authority, timeline, applied by a human in minutes, which replaces invented point thresholds with the shared sales-marketing definition whose absence MarketingSherpa's misalignment data documented (MarketingSherpa, 2011). Third is the speed-to-lead stack: instant routing, calendar-first replies, and founder-signed templates designed to start conversations inside the one-hour window the HBR research rewards (Oldroyd et al., 2011). Fourth, conversation-based qualification moves discovery questions into the booking flow itself, two or three qualifying fields on the scheduling form, so every booked call arrives pre-triaged. Finally, firms adopt opportunity-centric record keeping in the spirit of Forrester's revenue waterfall: pipeline is counted in qualified conversations about a defined problem, not in MQLs, which keeps reporting honest about the sub-1% trap (Forrester, 2022).

Section 6

Solution framework

A qualification system sized for a small service team has four components. Component one is the fit definition: a one-page ideal client profile with three to five disqualifying criteria, agreed by whoever markets and whoever sells, directly addressing the definitional misalignment documented in MarketingSherpa's benchmarks (MarketingSherpa, 2011). Component two is signal tiers in place of scores: Tier A is a direct hand-raise, a booked call or pricing inquiry; Tier B is multi-person engagement from one account, the buying group signal Forrester's research validates (Forrester, 2022); Tier C is individual content engagement, which feeds nurture rather than outreach. This three-tier model preserves prioritization while abandoning arbitrary point arithmetic. Component three is the response standard: Tier A contacted within one hour during business hours, consistent with the sevenfold qualification advantage in the HBR timing research (Oldroyd et al., 2011), with automation covering nights and weekends via instant scheduling links. Component four is the review loop: monthly, trace every closed-won and closed-lost engagement back to its original tier and source, adjusting the fit definition rather than the weights, the validation step Forrester notes most scoring systems never perform (Forrester, 2022). Inside LeverageOS installations, this is the LeadOS qualification loop: define, tier, respond, review. It deliberately contains no point scores: at small-team volume, explicit criteria plus speed beats modeled propensity.

Section 7

Evidence-based action plan

Week 1: write the fit definition with disqualifiers, and audit your current response time honestly, timestamp the last twenty inquiries from arrival to first human reply, and compare against the 42-hour average and one-hour standard in the HBR research (Oldroyd et al., 2011). Weeks 2-3: implement the speed stack: instant notification for hand-raises, a calendar-first reply template, and two or three qualifying questions embedded in the booking form. Retire any inherited point-scoring automation; replace it with the three signal tiers. Month 2: add buying-group detection at whatever fidelity your tools allow, even a weekly manual scan for repeated email domains across your list, webinar attendees, and inquiries, acting on Forrester's finding that multi-person engagement is the high-propensity signal (Forrester, 2022). Month 3: align reporting to opportunities, not MQLs: count qualified conversations, proposals, and wins by source, accepting Forrester's sub-1% benchmark as the cautionary baseline for lead-centric counting (Forrester, 2022). Months 4-6: run the monthly review loop, tracing outcomes back to tiers and tightening the fit definition each cycle; expect the gains to show up first as fewer bad calls, then as faster cycles, consistent with Gartner's evidence that buying groups reward suppliers who reduce decision friction (Gartner, 2024-2025). The end state is not a smarter score; it is a shorter, faster path from hand-raise to honest yes-or-no. For adjacent evidence in this series, see [The Speed-to-Lead Crisis: What the Response-Time Research Actually Says](/blog/speed-to-lead-crisis-response-time-research-deep-dive) and [Rising Customer Acquisition Costs: The Research Behind CAC Inflation and the Owned-Audience Counter-Strategy](/blog/rising-customer-acquisition-costs-research-deep-dive).

FAQ

Direct answers for operators.

Is the MQL really dead?

As a unit of measurement for committee-based B2B buying, the evidence is damning: Forrester's waterfall benchmarks show lead-centric MQL processes converting under 1% from inquiry to closed-won, and its 2021 buying survey found over 80% of decisions made by groups of three or more, which individual lead records cannot represent. Qualification itself is not dead; the individual-lead unit of analysis is what the research discredits.

Should a small service business use lead scoring software?

Usually not. Forrester's research shows scoring weights and thresholds are typically guesses with no validated link to purchase propensity, and small firms lack the lead volume that makes even imperfect prioritization useful. A written fit definition with explicit disqualifiers, three signal tiers, and a one-hour response standard delivers the prioritization benefit without the false precision, and it can be run entirely inside a basic CRM.

What is the most evidence-backed way to improve qualification rates?

Respond faster. Harvard Business Review research auditing 2,241 companies found firms contacting web leads within an hour were nearly seven times likelier to qualify them than those waiting even an hour longer, while the average firm took 42 hours and about a quarter never replied. Before redesigning scoring, instrument your own response time; for most small teams it is the single largest fixable leak.

How do we qualify buying groups instead of individual leads?

Watch for clustered signals from one account: two or more people from the same domain subscribing, attending a webinar, or visiting key pages within a short window. Forrester's research identifies this multi-person engagement as dramatically more predictive of purchase than any individual's activity. At small-firm scale, a weekly manual scan for repeated company domains across your lists and inquiries captures most of the value without new software.

Joshua Agonya Pi'Rwot

Written by

Joshua Agonya Pi'Rwot

Founder, Business Growth Accelerator · Country Director, AVODA Group Uganda · EMBA

Joshua helps service-business operators turn scattered marketing into a clear path from first attention to booked call. He is Founder of Business Growth Accelerator and Country Director of AVODA Group Uganda.