There's a metric most B2B marketing teams would rather not think about: the percentage of MQLs that go completely unworked. In revenue operations audits across growing SaaS companies, it's common to find that somewhere between 60 and 75 percent of all marketing-qualified leads generated in a given quarter receive zero meaningful follow-up from sales. They get created in the CRM, assigned a status, maybe auto-enrolled in a generic nurture track — and then nothing. They age. They go cold. Eventually someone bulk-deletes the cohort to clean up reports.
This is the MQL graveyard, and it is not primarily a sales discipline problem. It is a systems problem — a predictable failure that emerges when qualification definitions are loose, routing rules are one-size-fits-all, and the SDR team has a fixed SLA window that can't accommodate the volume of inbound leads a decent demand-gen motion produces.
Why MQLs Die: The Structural Root Causes
The graveyard fills for three compounding reasons. First, MQL definitions at most companies are threshold-based on marketing engagement score — a contact downloads a whitepaper and attends a webinar and suddenly hits 75 points and becomes an MQL. The problem is that marketing engagement and purchase intent are correlated but not identical. A junior analyst at a company that's nowhere near your ICP can hit 75 points as easily as a VP of Revenue Operations at a 200-person SaaS company. When the SDR team gets that lead queue, they quickly learn which ones convert and which ones waste call time. The non-converting ones start getting skipped, and your official worked-lead rate diverges dramatically from your actual worked-lead rate.
Second, contact-to-opportunity SLAs — typically 24 to 72 hours for inbound MQLs — are designed for peak-state lead volumes. Consider a 30-person AE team operating with a stated 24-hour SLA on inbound that's actually running 3 to 4 days behind target during content campaign pushes. By the time the SDR makes first contact, the lead's context (they were just comparing vendors, they just attended your webinar) has decayed. Response rates drop. The SDR mentally files the lead as "not ready" and moves on.
Third — and this is the one RevOps teams fix last because it requires the most cross-functional coordination — there is no systematic process for what happens to MQLs that didn't convert on first contact. They fall into a recycled status in Salesforce or HubSpot, get added to a long-running email nurture sequence that sends the same four emails to everyone, and then disappear into the contact record history. Nobody owns the question of whether that lead's situation has changed three months later.
What "Graveyard Leads" Actually Look Like
Take a concrete scenario: a 120-person B2B SaaS company running $18M in ARR generates roughly 2,800 MQLs per quarter through content, events, and paid search. Their SDR team of six works through approximately 900 of them — the ones with above-average scores, mid-market or larger firmographic fit, and intent signals the team has learned to look for manually. The remaining 1,900 get a status of "Recycle - Low Priority" and an enrollment in a 6-email nurture track that hasn't been updated in 14 months.
Of those 1,900 recycled leads: roughly 15 to 20 percent will, within the next 90 days, show renewed behavioral signals — they return to the pricing page, they search terms related to your product category on G2 or Capterra, they change job titles in a way that suggests new buying authority. Without a system watching for those signals, none of those renewed-intent leads get worked. They stay in the graveyard, even though they've essentially woken up.
The MQL Qualification Problem Is Also a Scoring Problem
The deeper issue is that legacy point-based scoring treats all engagement types as roughly equivalent inputs to a single threshold. A webinar attendance might be worth 20 points, an ebook download 10 points, a pricing page visit 15 points. The model doesn't know whether the pricing page visit happened on day one of a 3-day comparison sprint or two months ago as an idle click-through. It doesn't factor in whether the company expanded headcount recently, whether they're in a buying stage indicated by technographic changes, or whether the contact's LinkedIn activity suggests they're researching your category.
This produces two failure modes. False positives: leads scored high enough to become MQLs that have low actual purchase intent, wasting SDR time. False negatives: leads scored low because they did most of their engagement anonymously or through channels that don't feed your MAP, who are actually high-intent but never get worked. The graveyard contains significant numbers of false negatives — leads who had genuine interest, received inadequate follow-up, and moved on to a competitor or deferred the decision.
We're not saying traditional lead scoring is broken — for qualification gates and SLA triggers, a threshold model is operationally workable. We're saying it is not a sufficient signal for prioritizing reactivation efforts, because reactivation requires predicting future engagement probability, not measuring past engagement accumulation.
A Systematic Approach to Reactivating the Best Ones
Fixing the MQL graveyard doesn't require redesigning your entire demand-gen motion. It requires adding one layer: a systematic process for identifying which dormant MQLs have renewed buying intent and routing them back to human follow-up before they fully disengage.
The process has four operational components. First, define the dormancy threshold with precision. A lead is dormant when: it was created as an MQL, it received at least one contact attempt, it has had no positive engagement or response within a defined window (typically 30 to 60 days post-MQL creation), and it has not been formally disqualified (no wrong persona, no active competitor deal, no do-not-contact flag). This distinguishes dormant leads from truly dead ones.
Second, build a reactivation scoring model that runs independently of your primary MQL scoring. Reactivation scoring looks at different signals: recency and frequency of return website visits, product category keyword activity on third-party intent platforms, job title or company changes in the last 90 days, and the time elapsed since the original MQL event (there are patterns in when dormant leads become re-responsive that vary by your sales cycle length). This is where machine learning approaches have clear advantages over point thresholds — gradient boosted models trained on your historical MQL-to-opportunity conversion data can identify the combination of signals that predicts re-engagement far better than a simple score.
Third, route reactivated leads with context. The sequence that reaches out to a dormant lead should not look like a generic nurture email. The SDR or the automated touch should reference what the lead previously engaged with, acknowledge that time has passed, and offer something new — a case study relevant to their vertical, a product update that addresses a gap they showed interest in, or a direct question about whether their evaluation timeline has changed. Salesloft and Outreach both support conditional personalization tokens that make this operationally feasible at scale without requiring one-off manual customization.
Fourth, set realistic expectations on response rates. Reactivated MQLs will not convert at the same rate as fresh inbound — that would be a strange outcome. But if your fresh MQL-to-opportunity rate is around 8 to 12 percent, a well-run reactivation motion can produce 3 to 5 percent conversion on the dormant cohort, with dramatically lower acquisition cost since you're not spending demand-gen budget to source those leads again.
The SLA Trap: Why More Urgency Isn't the Fix
A common response to the graveyard problem is to tighten the SLA: if SDRs called every MQL within 4 hours instead of 24, would conversion improve? The answer is: only marginally, and only for a small subset of leads. For genuinely high-intent inbound — someone who requested a demo, someone who visited pricing and then signed up for a trial — speed matters significantly. Gartner research on B2B buyer self-service behavior consistently shows that buyers completing digital research phases expect fast response when they do reach out.
But for the larger pool of engagement-scored MQLs who aren't at the end of a buying journey, calling faster doesn't fix the underlying problem. You're still calling leads who aren't ready. The SDR experience doesn't improve. The graveyard just fills faster because leads cycle through the SLA window more quickly before being marked unresponsive.
The actual opportunity is in the middle of the funnel: building the capacity to identify when a previously-dormant lead has moved into a more active research phase, and deploying a targeted, contextually-aware touch at that moment. That is fundamentally different from trying to accelerate initial contact, and it requires different tooling and different operational ownership.
Getting RevOps and Marketing Aligned on the Fix
The structural challenge with fixing the graveyard is that it requires genuine alignment between RevOps and Marketing on who owns recycled MQLs. Marketing often considers their job done once the lead is created and handed to Sales. Sales considers recycled MQLs to be Marketing's problem. In the absence of clear ownership, the recycled cohort gets no systematic attention from either side.
The most effective model we've seen is a RevOps-owned reactivation queue: RevOps defines the dormancy criteria, builds or deploys the reactivation scoring logic, and owns the handoff rules (which leads get an automated multi-channel touch, which leads go straight to an SDR with enrichment context, which leads get added to a long-cycle nurture because the timing signals suggest they're 6+ months out). Marketing feeds the queue but doesn't manage the individual tactics. Sales receives the leads when they've hit a reactivation score threshold, not when they've simply aged into a follow-up window.
That ownership model requires buy-in from a VP of Revenue Operations or CRO, because it involves changing how SDR capacity is allocated and how MQL SLA compliance is measured. But companies that get it right typically find that their dormant MQL cohort is one of the highest-ROI buckets in their entire demand generation portfolio — because the leads have already been partially qualified, there's no incremental sourcing cost, and the conversion economics, even at lower rates, are significantly better than net new outbound.