When a lead-by-lead hand-off stops fitting
The guide to MQLs, SQLs and lead stages names the three situations in which the marketing-qualified leadA lead that marketing judges ready for sales follow-up against agreed criteria. stops helping. In account-based programmesA B2B approach that targets a chosen list of companies with coordinated marketing and sales. an organisation buys, so one person's score says little. In product-led businesses, usage says more than form behaviour. And where buyers are mostly created rather than caught, they arrive asking for a conversation, so a stage that decides whether to call them only adds delay. The guide keeps the stages and changes the hand-off: it should fire on whichever signal really foretells a purchase, and marketing should stop being reported on a stage count that has lost its meaning. This page walks through that change for the first case, a business whose sellers own named accounts. The product-led case has its own stage, set out in the guide to product-qualified leads.
The comparison of MQLs and SQLs gives the second reason to change. The comparison warns that no funnel figure is easier to inflate than a marketing-qualified count, since relaxing the criteria lifts it straight away, and it puts marketing's target on what sales accepts or on the pipeline that follows. Goodhart's law describes the mechanism: once a measure becomes a target, people chase whatever shifts it at the lowest cost, and lowering a score threshold shifts a lead count very cheaply indeed.
The signs that the person-level hand-off no longer fits usually show up in the CRM before anyone names them.
| What you see | Where it shows | What it suggests |
|---|---|---|
| Sellers ignore marketing-qualified leads from accounts they already own and work the accounts directly | Acceptance rate by whether the account had an owner | The hand-off duplicates work sellers do anyway |
| Opportunities open at accounts where nobody ever became marketing-qualified | Opportunities by whether any contact crossed the threshold first | The stage misses much of the buying |
| Qualified leads are mostly practitioners and researchers, while the people who sign never fill in a form | Job roles of qualified leads against the roles on won opportunities | The score rewards reading, not deciding |
| Several people from one organisation qualify separately and reach different sellers | Qualified leads per account, and owners per account | Contacts are not attached to their accounts |
| Demo requests wait for a score while the buyer compares other suppliers | Time from request to first contact by lead type | The hand-off delays declared intent |
If most of the table describes your pipeline, the stage has outgrown its job. If none of it does, and sellers still call whoever is ready from a large inbound flow, the marketing-qualified lead may be doing what it was built for, and the comparison page's point stands: the stage still pays its way where sellers cannot call everyone and a threshold decides who is called first.
Before you start
- The sales leader's agreement to change the hand-off, given before any configuration. The account criterion is a definition both teams will be measured against, and the sales and marketing alignment guide treats stage definitions as part of the agreement between them, not as a marketing setting.
- The current stage definitions in writing, from the lead stages guide and the CRM lead stages and pipeline guide, including the picklist of return reasons. You will change the first stage and keep the rest.
- A named-account list with owners, built from the ideal customer profile and tiered as the account-based marketing guide describes. Accounts with no owner need a routing rule of their own.
- Contacts matched to company records, or a plan to match them, following the guide to matching contacts to company records. The account criterion adds up the activity of everyone attached to an account, so it is only as good as the matching.
- The current scoring sheet, ideally kept in the lead scoring model template, so the fit and intent lists the new criterion reads are written down.
- A cohort reading of the current stage rates: leads that became marketing-qualified in a period, followed to acceptance, opportunity and outcome, by source. This is the baseline the parallel run is judged against.
- A place to record measurement changes, such as the measurement change log, because this change moves every report that counts qualified leads.
The steps
Check what the marketing-qualified stage still predicts
Take the last few complete cohorts of marketing-qualified leads and follow each to acceptance, opportunity and outcome, by source and by whether the account already had an owner. Beside them, list the opportunities opened in the same period at accounts where no contact ever crossed the threshold, and note who on each won opportunity was involved. The conversion rate calculator turns each pair of counts into a rate; the guide to reading small numbers explains why each rate should carry its count and a range. You are looking for the pattern in the table above: a stage that accepts a small share of what it sends, misses many of the opportunities that open, and is populated by roles that rarely decide. Write the finding down with its counts; it is the baseline for every later comparison. If the stage still predicts well, stop here and tune it instead.
Attach every contact to its account
An account-level rule adds together what everyone from one organisation has done, so a contact attached to the wrong company, or to none, is engagement the rule cannot see. Apply the matching rules the matching guide sets out: the email domain where it is reliable, a review route for personal and shared addresses, every known domain held on the right record, and parents, subsidiaries and regional offices arranged so that the account a seller owns is the one whose activity is counted. Re-matching a backlog moves historic pipeline between accounts, so record the date and the rule in the change log before the reports move. Do this first; a criterion built on fragmented records will appear to fail for reasons that have nothing to do with it.
Write the account criteria with sales
Sit down with the sales leader and the sellers who own accounts and write, in plain words, when an account becomes theirs to act on. As general practice, a workable criterion combines three parts. Fit: the account matches the ideal customer profile and none of its disqualifiers applies. Breadth: people in more than one role of the committee have engaged, mapped against the roles the account-based marketing guide describes, so one enthusiastic reader cannot qualify an organisation alone. And a buying action: something only a would-be purchaser does, for example a pricing visit, a security question or a request to talk to someone. How each person's activity is scored, weighted and decayed stays with the lead scoring guide and its separate fit and intent lists; this step writes the line at which the account is handed over. Record the criterion, who agreed it and the date. Goodhart's law adds one rule worth writing into the agreement: the team measured by the number must never be the sole owner of its definition, so any change to the criterion goes through a recorded decision that sales also signs.
Keep a fast lane for declared intent
A demo request, a quote request or a direct question about buying is intent at its clearest, and it must never wait for the rest of the account to engage. The lead scoring guide already routes a declared action at once, regardless of score, and the product-qualified leads guide keeps the same path for free users who ask to talk to sales. Keep that route outside the account criterion: the request goes to the account's owner if it has one, to the owner of the relationship if the organisation is already a customer or in an open opportunity, and to the rule for unowned accounts otherwise. Tag these hand-offs so the pipeline that arrives by declaration can be counted apart from the pipeline the criterion produces.
Rewrite the hand-off around the account owner
The lead stages guide gives every hand-off four parts, and each needs rewriting for accounts. The trigger is the account meeting the criterion, recorded as a status and a date on the account. The route is the account's owner, with a rule for accounts nobody owns; write each route one line at a time in the lead routing rules template. The service level is how quickly the owner accepts or returns the account, measured from CRM timestamps and reported where both teams meet, as the alignment guide insists. And the return path is a reason from the picklist, now applied to the account: not ready, wrong fit, already in conversation, or a missing role. Send the owner the context that makes the account worth a call: which roles engaged, what each did and which buying action tipped it over. The guide to sales alerts from marketing activity shows how account-level engagement can reach its owner as a digest rather than as one alert per click. Before anything goes live, run the lead routing QA checklist with test records for every route, including the unowned account and the fast lane.
Run the old and new hand-offs side by side
Do not switch everything on one day. The one-way and two-way doors framework separates decisions that can be undone cheaply from those that cannot, and this one stays a two-way door only as long as the old stage is still calculated and the old reports still run. Begin in shadow: the CRM marks the accounts that meet the new criterion, while routing still follows the old threshold, and for each cohort you compare which accounts the new rule would have sent with what the old rule sent and what became opportunities. Then move one segment, or the territories of sellers who volunteer, to the live rule while the rest continue as before. The groups were not chosen at random, so read any difference as an estimate, not a proof. Note the start date of each phase in the change log. Agree the end date of the parallel runRunning the old and new versions of a process, platform or tag set side by side for an agreed period, compared on one measure, before the old one is retired. at the start as well, so the comparison ends in a decision rather than drifting on.
Move targets from lead counts to accounts and pipeline
Once the new rule is live anywhere, marketing's target cannot stay a count of qualified people. The MQL and SQL comparison puts marketing's goal on what sales accepts or on the pipeline that follows, because neither can be met by loosening marketing's own criteria. Set the new target on accounts accepted by their owners and on the opportunities and pipeline opened from them, derived from your own stage rates rather than from a figure borrowed from elsewhere. Kaushik's measurement model, as the ABM diagnosis linked below applies it, derives every measure from the business objective, and a target on accepted accounts keeps that chain intact where a count of handed-over accounts would not. Pair it with a guardrail metricA metric an experiment must not worsen beyond an agreed bound, watched beside the primary metric that decides the test.: the share of handed-over accounts that owners accept, so that marketing cannot hit the volume by sending accounts sellers reject.
Change the agreement, the reports and the training
Rewrite the parts of the service-level agreementA published promise between teams, or with a supplier, about what each type of request needs and how quickly it will be handled, agreed rather than imposed. that mention marketing-qualified leads: what marketing hands over, with what context, and what the owner owes in return. Keep the later stages as they are; a sales-acceptedA lead sales has reviewed and agreed is worth working, recorded as a stage between marketing-qualified and sales-qualified with a reason when rejected. account still becomes an opportunity under the same criteria, so pipeline reports keep their meaning. Update the shared dashboards so both teams read the same fields, change the weekly lead review in the inbound lead handling playbook to look at returned accounts and their reasons, and add committee roles engaged to the data pack of the weekly pipeline review. Then brief sellers on what arrives and what they owe, and marketers on nurture and plays that now run for accounts below the line.
Retire the lead count from marketing's scorecard
When the live rule covers the whole business, take the marketing-qualified count off the target list. It can stay as a diagnostic with no target attached, useful for seeing whether content still reaches people, but it no longer judges anyone. Never add qualified people and qualified accounts together in one figure; the comparison page warns that adding stages counts the same buyer twice. Marketing-sourced pipelineValue of sales opportunities whose original source was a marketing activity. keeps the definition the business-to-business measurement guide gives it, first touch on the account, so the outcome reports run on through the change. Mark the date in every trend chart that shows qualified volumes, so a step in the line is read as the rule changing rather than the market.
Review the rule after a full sales cycle
Set the review date when the rule goes live, far enough ahead for the first cohorts of handed-over accounts to reach outcomes. At the review, compare those cohorts with the baseline on acceptance, opportunities, pipeline, win rate and cycle length, separately for the fast lane and the criterion. Read the return reasons: accounts returned for a missing role suggest the breadth part is too loose; accounts accepted but never progressed suggest the buying action is too weak. Adjust through the recorded decision, not quietly. The lead management audit scorecard rates the whole path from stage definitions and routing to follow-up and closed-loop reporting if a broader check is due, and the diagnosis of an account-based programme that is not working is the place to start if accounts qualify and still produce nothing.
What changes and what stays the same
The change is narrower than it sounds: it moves the first hand-off and the target, and leaves the later stages and the sourcing rule where they were.
| Part of the funnel | Person-level hand-off | Account-level hand-off |
|---|---|---|
| What is handed over | One contact whose score crossed a threshold | An organisation that meets criteria agreed with sales, with the people engaged listed |
| Trigger | The contact's fit and intent scores | Fit, engagement from more than one committee role and a buying action |
| Who receives it | The next seller by territory or rotation | The account's owner, or the rule for unowned accounts |
| Context sent | The contact's recent actions | Roles engaged, what each did and the action that tipped it over |
| Declared intent | Routed at once | Still routed at once, outside the criterion, and counted separately |
| Return path | A reason for the contact | A reason for the account, including a missing role |
| Marketing's target | Qualified leads | Accounts accepted by owners, with the opportunities and pipeline they open |
| Guardrail | Often none | The share of handed-over accounts that owners accept |
| Scoring | Decides the hand-off | Still runs, and feeds the account criterion |
| Later stages and sourced pipeline | As defined in the stage guides | Unchanged |
Two things are easy to forget. Consent and contact records belong to people, not to accounts, so an account that qualifies does not make every contact in it someone marketing may email. And the people in an account still have their own journeys: the lead scoring threshold test remains the way to test where a person-level line should sit, if the business keeps one for nurture.
A door access control supplier moves to account qualification (illustrative)
Krovaleth Access Systems replaces its lead-by-lead hand-off
Krovaleth Access Systems is an invented supplier of door access control systems to property managers that run several sites. Its 6 sellers own 420 named accounts, and marketing hands over any contact whose score crosses a threshold. Over the last four quarters the threshold produced 1,150 marketing-qualified leads. Sellers accepted 240 of them, about 21%, and 61 became opportunities. In the same period, 38 opportunities opened at owned accounts where no contact had ever qualified, and on the won deals the people who signed were operations directors and finance managers, while 70% of qualified leads were facilities coordinators reading maintenance guides.
The team first matches contacts to companies, merging 37 duplicate company records and attaching 900 contacts that had been unmatched. With the sales leader it writes the criterion: an account in the profile, at least two committee roles engaged in the last 60 days, and one buying action such as a pricing visit or a security question. Demo and quote requests still go straight to the account owner.
In a shadow quarter, the old rule sends 290 leads from 170 accounts; sellers accept 61 and open 14 opportunities. The new rule would have sent 96 accounts, and by the end of the quarter 29 of them already had opportunities, including 11 of the 14 from the lead hand-off. The next quarter it runs live in the territories of 3 sellers who volunteered: of 52 accounts handed over, owners accept 41 and open 15 opportunities, while the other territories open 7 from the old hand-off. The team reads the gap as a direction, since the groups were not chosen at random and the counts are small, and rolls the rule out. Marketing's target becomes 75 accepted accounts a quarter and the pipeline they open, with a guardrail that owners accept at least 70% of what is handed over. The lead count stays on the dashboard as a diagnostic, with the date of the change marked on every chart.
The pipeline section sets this change beside the guides on stages, alignment and measurement that it depends on. If the qualified stage is still in use and sales reject most of what it sends, the diagnosis of MQLs not converting to SQLs checks the threshold, the scoring and the follow-up before a change of model is considered. For named accounts, the ABM account plan template records the committee map that the breadth criterion reads.
Common mistakes
- Renaming the stage and keeping the rule. An account flagged because one person crossed a score is a marketing-qualified lead with a new label; the criterion has to need more than one role.
- Switching before contacts are matched. Fragmented company records hide the engagement the rule adds up, and the new hand-off appears to fail for a data reason.
- Holding demo requests until the account qualifies. Declared intent goes straight to the owner, whatever the account's state.
- Writing the criterion without sales. A line marketing set alone is a target marketing can move; agree it, record it and change it only through a signed decision.
- Changing everything on one day. Without a shadow period and a partial rollout there is no baseline, and no easy way back if the rule is wrong.
- Keeping lead volume as marketing's target. The old count invites the old behaviour; target accepted accounts and pipeline, with acceptance as the guardrail.
- Adding accounts and leads together. One buyer counted as a qualified person and as a qualified account inflates every report that uses the sum.
- Leaving the change out of the charts. Without the date marked, a drop in qualified volume looks like a collapse in demand.
