According to DOJO AI’s 2025 research, 85% of leads that marketing passes as MQLs are rejected or ignored by sales. Meanwhile, 82% of C-level executives believe their marketing and sales are aligned - compared to just 35% of practitioners who say the same. The gap is not in intentions: both teams want to close deals. The gap is in definitions: marketing and sales understand MQL to mean fundamentally different things, without ever having agreed explicitly.
The cost is real: misaligned teams lose up to 10% of annual revenue to missed leads and misdirected spend. Aligned teams show 24% faster revenue growth and 36% higher customer retention. In EU B2B with long sales cycles and multiple stakeholders in the buying committee, this gap becomes a structural problem, not an operational one.
This article covers why standard MQL/SQL definitions do not work, how to build a qualification system for EU mid-market, and how to embed it in a CRM. If you want to solve this problem systematically, see our article on RevOps: aligning marketing and sales is one of the core RevOps functions.
Why activity-based scoring does not work
The standard MQL approach: a lead accumulates points for actions - opened an email (+5), visited the site (+10), downloaded a whitepaper (+20), watched a demo (+50). Once the threshold of 100 points is reached, it becomes an MQL and is passed to sales.
The problem with this model is that activity does not equal purchase intent. A competitor, a student, and an ideal buyer can accumulate the same score by taking the same actions. Sales receives the MQL, makes two calls, realises it is not a target lead, and stops trusting marketing leads as a category.
In EU B2B the situation is more complex: buyers here are traditionally more cautious about sharing contact details, less likely to open cold emails (GDPR and general culture), and more likely to conduct independent research before first contact with a vendor. Activity in your owned channels is a poor proxy for intent with a European audience.
What MQL criteria should include for EU B2B
A working MQL definition combines two types of data: firmographic (who) and behavioral (what they do).
Firmographic signals - mandatory filters, without which a lead cannot become an MQL regardless of activity:
- Company size: fits your ICP (for example, 15-200 employees)
- Geography: country or region from your target market
- Industry: a vertical you serve
- Role: the title matches your buyer persona or influences the buying decision
High-intent behavioral signals - not any activity, but actions that correlate with genuine purchase interest:
- Requested a demo or pricing
- Visited the pricing page three or more times within two weeks
- Viewed a case study in your vertical
- Returned to the site five or more times in a month
- Submitted a form with a specific question (not a newsletter signup)
MQL = firmographic fit + high-intent behavioral signal. Without both conditions simultaneously - not an MQL.
BANT, MEDDIC, CHAMP: what to choose for EU mid-market
SQL criteria define when a lead is ready for full engagement with an AE. The qualification framework is a sales team decision, but marketing needs to understand the logic.
BANT (Budget, Authority, Need, Timeline) - the simplest framework, works for transactional deals. For EU mid-market B2B SaaS with deal sizes of 20-100k EUR per year, it is insufficient: it does not account for buying committee complexity or organisational pain.
MEDDIC (Metrics, Economic Buyer, Decision Criteria, Decision Process, Identify Pain, Champion) - an enterprise framework, usually overkill for companies under 100 people. But some elements are valuable: the presence of a Champion inside the target company is a critical signal for EU mid-market, where decisions move slowly and an internal advocate is essential.
CHAMP (Challenges, Authority, Money, Prioritization) - the practical choice for EU B2B companies of 15-100 people. It starts with Challenges (the real problem the product solves), which aligns with how EU buyers make decisions: they want to understand fit before discussing price.
Recommendation: take CHAMP as the foundation and add Champion from MEDDIC. SQL status is reached when: there is a real problem, there is a decision-maker or a champion who can reach one, a budget exists (not necessarily approved), and there is a time horizon.
SLA between marketing and sales
An SLA is a formal agreement about what marketing commits to do with leads and what sales commits to do with MQLs. Without it, both sides operate by their own rules.
Marketing’s commitments:
- MQL is passed with firmographic fields filled (company, role, size, country)
- The last 30 days of activity history is attached to the MQL
- Additional context where available: which campaign it came from, what content was read or downloaded
Sales’ commitments:
- First contact with an MQL: within 24 hours on business days (standard for EU mid-market; enterprise can stretch to 48 hours)
- MQL status is updated in the CRM: Accepted, Working, or Rejected with a reason
- Rejected MQLs: a reason is required (not ICP, wrong timing, already a customer, competitor)
Feedback loop - the most important part: rejected MQLs with reasons are returned to marketing monthly. This is the only way to iteratively improve the scoring model based on real data rather than intuition.
Lead scoring: rules vs ML
Two approaches to building a scoring model.
Rule-based scoring - you define the weights yourself: +30 for the pricing page, +20 for a demo request, -20 if company size is outside your ICP. The advantage: transparent, easy to explain to sales, easy to update. The disadvantage: weights are subjective and require manual calibration.
For most EU B2B teams of 15-100 people, this is the right starting point. The complexity of ML models is justified when you have thousands of MQLs per month; at 50-200 MQLs per month, rules outperform ML through sheer transparency.
ML-based scoring - the model trains on historical data (which MQLs converted to SQL, which did not) and predicts conversion probability. HubSpot Predictive Lead Scoring and Salesforce Einstein are available on enterprise plans. Works well when you have clean historical data for 12+ months and a volume of 200+ MQLs per month.
Switch to ML when: the rule-based model stopped improving from manual adjustments, sufficient data exists, and there is capacity to configure and monitor the model.
Setup in HubSpot and Kommo: lifecycle stages
In HubSpot: Lifecycle Stage is a built-in field (Subscriber -> Lead -> MQL -> SQL -> Opportunity -> Customer). Set up a workflow: when the scoring threshold is reached AND firmographic criteria are met, automatically move to MQL and notify sales. Create a dedicated view for sales: all MQLs sorted by score and transition date.
In Kommo (formerly amoCRM): pipeline stages are configured manually. Create a separate qualification pipeline or use tags: “MQL” and “SQL” as tags with their own automations. For scoring, use custom fields and scoring through the built-in automation or Zapier.
In both systems: rejection reasons should be a required field when changing status. These are the data points you analyse later.
Process metrics
Four metrics that reflect the health of the MQL/SQL process:
MQL-to-SQL rate - what percentage of MQLs sales accepts for active work. Normal range for B2B SaaS: 25-40%. Below 20% signals a problem with MQL quality or overly strict SQL criteria from sales. Above 50% suggests SQL criteria may be too loose.
SQL-to-Opportunity rate - what percentage of SQLs becomes a full deal. Normal range: 40-60%.
Opportunity-to-Close rate - win rate. In EU B2B mid-market: 20-30% is a strong result.
Time-to-MQL-contact - how long from MQL to first sales contact. Every additional day reduces conversion. Monitor monthly - this is an operational discipline metric for the sales team.
A quarterly review of these four metrics with both marketing and sales present is the minimum process for maintaining alignment. Without shared data, both teams will inevitably diverge in their reading of the situation.