Discovery call frameworks: BANT vs MEDDIC vs SPIN
Three discovery frameworks compared, and which one to use when
BANT, MEDDIC, and SPIN are the three discovery frameworks every B2B sales team eventually adopts, modifies, or resents. Each was built for a different era of selling — BANT for IBM's product-sale 1960s, SPIN for Huthwaite's research-driven 1980s, MEDDIC for the enterprise-software 1990s — and each captures a different theory of what discovery is for. Picking the right one for your deal is less a religious choice than a function of three things: how complex the buy is, how many stakeholders sign, and how disposable the prospect's pain is. This page compares the three honestly, names the situations where each one wins, and shows the specific question shapes that produce dramatically different answers from the same prospect.
BANT — built for transactional, single-decision-maker B2B
BANT (Budget, Authority, Need, Timeline) was IBM's framework for qualifying prospects in the era when buying software meant calling IBM. The four questions are sharp, transactional, and assume a single decision-maker who can speak for the buying decision. Modern critique: BANT-by-the-book skips past the harder qualification work — does the prospect actually have the pain, or did they just say yes to your meeting request? Where BANT still wins: SMB deals under $25k ARR with a single buyer (usually a department head). Where BANT fails: any deal with 4+ stakeholders, any deal where the buyer doesn't yet know they have a problem, any deal where 'budget' is decided after value is proven (most modern enterprise SaaS).
MEDDIC / MEDDPICC — built for complex enterprise
MEDDIC (Metrics, Economic buyer, Decision criteria, Decision process, Identify pain, Champion) was developed at PTC in the 1990s for enterprise-software deals where 6+ stakeholders signed. MEDDPICC adds Paper process and Competition. The framework's edge: it forces the rep to map the buying organization's structure — who signs, what the legal review looks like, how the decision actually gets made — instead of just qualifying the lead. The cost: MEDDIC discovery takes 3–5 calls instead of one; reps who try to compress it sound interrogative ('who's your economic buyer?' is a terrible first-call question). Where MEDDIC wins: ACV >$50k with 4+ stakeholders, regulated industries (finance, healthcare, government), procurement-heavy buying. Where MEDDIC fails: SMB deals where the framework's scaffolding becomes friction.
SPIN — built for question-driven discovery
SPIN (Situation, Problem, Implication, Need-payoff) came out of Neil Rackham's 35,000-call research at Huthwaite in the 1980s. The framework's insight: top performers ask more Implication and Need-payoff questions, not more Situation questions. Implication questions force the prospect to articulate the cost of their problem; Need-payoff questions force them to articulate the value of solving it. The prospect convinces themselves. Where SPIN wins: any sale where the prospect knows they have a problem but hasn't quantified it (most modern B2B). Where SPIN fails: transactional commodity sales where Situation questions are wasted time; pure cold outbound where the prospect doesn't yet acknowledge the problem and Implication questions land flat.
When to mix, and how
Most senior AEs end up running a hybrid: SPIN for the discovery conversation itself (because the question shape produces better answers than BANT-style direct qualification), MEDDIC for the deal-stage tracking and stakeholder mapping (because the framework forces you to articulate what you don't yet know), and BANT as a sanity check at the end of discovery (because if Budget / Authority / Timeline aren't viable, the deal is dead regardless of how well-discovered the pain is). The mistake to avoid: picking one framework and applying it dogmatically. Each framework's specific question shape produces specific answer types; the senior-AE skill is recognizing which question type the moment calls for and switching live.
The specific question shapes that matter most
If you only memorize three discovery questions, make them these. SPIN Implication: 'What's the cost of [the problem] continuing for another quarter?' (forces quantified pain — answers like '$200k/quarter in lost productivity' do most of the discovery work for you). MEDDIC Decision Process: 'Walk me through how a decision like this gets made at your company — who's involved, what's the timeline, what's the legal review look like?' (surfaces stakeholder map and procurement complexity in one question). The Mom Test pivot: 'Tell me about the last time you tried to solve [problem]' (replaces 'would you buy this' with a memory question that produces specific answers). These three questions compress most of what BANT/MEDDIC/SPIN are trying to do; everything else is scaffolding around the answers they generate.
Checklist
- ☐Identify the deal complexity tier first: SMB single-buyer (<$25k ACV) → BANT works; mid-market 2-3 stakeholders → SPIN-led; enterprise 4+ stakeholders → MEDDIC scaffolding
- ☐On the first discovery call, lead with SPIN Implication and Need-payoff questions — produces specific quantified pain instead of vague positivity
- ☐Use the MEDDIC 'walk me through your decision process' question on every deal >$25k ACV — surfaces hidden stakeholders before they kill the deal
- ☐Apply the Mom Test pivot when the prospect's answer is too positive: 'tell me about the last time you tried to solve this'
- ☐Sanity-check at end-of-discovery with BANT — Budget viable? Authority on the call or named? Timeline within your sales cycle? Need quantified?
- ☐Document the answers immediately post-call (the per-call session summary auto-captures this) — discovery details decay fast when retold from memory in CRM-update batch mode
How HearQA Helps
- Upload The Mom Test, SPIN Selling, MEDDPICC, and your specific discovery template into the document library — HearQA surfaces the right question shape for the conversation moment
- Practice → Mock Interview lets you rehearse the harder discovery patterns (Implication-question recall, MEDDIC stakeholder-process probing) before a high-stakes call
- Per-call session summary auto-captures the BANT/MEDDIC/SPIN signal you heard — quantified pain, stakeholder names, decision-process specifics — so the post-call CRM update is the customer's actual quotes, not your approximate recollection
- Multi-language support means an AE running international territories can ask the same SPIN Implication question fluently in Portuguese / Spanish / French / German / Hindi / Japanese / Korean without losing depth
FAQ
Which framework should I learn first as a new AE?
SPIN — because the question-shape difference (Implication and Need-payoff vs Situation and Problem) is the deepest skill, and once you internalize asking better questions, BANT and MEDDIC become checklists you apply at the end rather than scripts you read live. Most new AEs make the same mistake: they ask Situation questions because those feel safer, then run out of time before getting to the Implication questions that produce the actual qualified-pipeline signal. Reading Rackham's SPIN Selling cover-to-cover (1988, but the research holds up) is one of the highest-ROI 8 hours an AE can invest.
Is BANT really obsolete?
Not obsolete — narrower in fit. BANT works fine for transactional, single-decision-maker B2B sales where the prospect already knows they have the pain and is shopping for vendors. Where it fails is anywhere 'budget' is decided after value is proven (most modern enterprise SaaS), or where the prospect doesn't yet acknowledge the problem (cold outbound, category-creation sales). HubSpot's GPCT framework (Goals, Plans, Challenges, Timeline) is essentially BANT updated for inbound-led modern sales. If your prospect found you, GPCT works. If you found them, SPIN's deeper question shapes produce better signal.
What's the biggest mistake reps make with MEDDIC?
Treating it as a checklist to read off in the discovery call rather than a discipline for what to learn ACROSS the call sequence. 'Who's your economic buyer?' is a terrible first-call question — it sounds like an interrogation. The MEDDIC question is something the rep should be able to answer about the prospect after 3 calls, derived from how the prospect talks about their organization. Asking the prospect to fill in the MEDDIC template directly produces shallow answers; using it as the rep's mental model for what to extract produces deep ones.
How do I avoid sounding like I'm running a script?
Internalize the framework's logic, not its words. SPIN's Implication question can be phrased a hundred ways — the underlying question is 'how much does this cost you?' MEDDIC's Decision Criteria question is 'what would they need to see to say yes?' Most reps who sound scripted are reading specific framework phrasings; reps who sound like consultants are extracting the same answers using their own conversational style. Practice → Mock Interview is the fastest way to build this — run 5 sessions phrased differently each time, the scripted-ness drops out by the third run.