The go-to-market frameworks worth knowing between €0 and €1M ARR
What a framework is actually for
A framework is a question generator, not an answer. Its value is that it forces you to write down something you have been carrying around vaguely — who buys fastest, what they would do instead, what you will not do this quarter — and then gives you a way to check whether you were right.
- A framework without a number is a vocabulary. If applying it does not change a target, a price or a list, you have relabelled the problem rather than solved it.
- Sequence matters more than coverage. Most of these depend on the output of another. Positioning needs your best-fit customers; channel choice needs your price; hiring needs a repeatable pitch. Running them out of order produces confident nonsense.
- Frameworks describe what worked somewhere else. When one conflicts with what you are seeing in your own numbers, your numbers win — say which assumption broke, and carry on.
Positioning and narrative
Obviously Awesome
What it says. Positioning has five components worked out in order: the competitive alternatives a buyer would actually use, the attributes only you have, the value those attributes create, the segment that cares most about that value, and the market category that makes all of it obvious. The alternative is what the buyer would do without you — a spreadsheet, someone’s time, a feature inside a tool they already pay for, or nothing at all — not your competitor list.
When to use it. When prospects say "interesting" and never buy, when you keep losing to the status quo, or when you cannot explain in one sentence what you replace.
The test. Say your positioning once to a prospect and ask them to repeat it back. If they cannot, it is not finished. Our free positioning stress test runs the five components and returns a Clear, Muddy or Broken verdict.
Strategic narrative
What it says. A five-part arc for decks and pitches: name an undeniable shift in the world, show that the shift creates winners and losers, tease the promised land the winners reach, introduce your capabilities as the way to get there, then show proof you have taken others there.
When to use it. Building a sales deck, a fundraising deck or a keynote — anywhere you need someone to care before they evaluate.
The test. Slide one is a change in the world, not your company. If the deck opens with "About us", it is broken.
Category design
What it says. You can compete in an existing category or create a new one. Creating means naming a problem the market has not named and funding years of education before the product sells itself.
When to use it. Rarely at this stage. The default below about $2M ARR is to compete in — or subdivide — a category that already has a budget line.
The test. Can a buyer say which budget this comes out of? If not, you are asking them to invent a line item, which is the expensive path.
Jobs to be done messaging
What it says. Customers do not buy products, they hire them to make progress in a situation — and they fire whatever was doing the job before, however crude. Describe the progress, not the product: get from X to Y without Z.
When to use it. When your homepage is a feature grid, or when you cannot name what gets fired when you win.
The test. Name the single job, who hires for it, and what gets fired. One product, one primary job — a homepage listing six jobs gets hired for none.
ICP and customer discovery
Best-fit segmentation
What it says. Derive your ideal customer profile from the customers who closed fastest, retained best and expanded most — not from imagination. Whatever those customers share (industry, size, stack, team structure, a triggering moment) is the segment.
When to use it. Before choosing a channel, writing outreach, or setting a price. It is upstream of almost everything else here.
The test. Can you build the list from your ICP in a single search? If not, it is still a description rather than a target. The free ICP clarity scorecard estimates how many companies yours actually describes.
Lean Customer Development
What it says. Interview for past behaviour, never future intent. People are honest about the past and polite about the future, so "would you buy this?" proves nothing. Listen for workarounds and money already spent — a maintained spreadsheet, an exported report, a hired person, a competing tool.
When to use it. Pre-product, pre-pricing, and any time you are unsure whether a pain is real.
The test. Five interviews with the same profile surfacing the same pain is signal. Fifteen across scattered profiles is noise — tighten the profile first. No workaround means no felt pain, which usually means not your ICP.
Continuous discovery
What it says. Discovery is a habit, not a phase: at least one customer conversation a week, forever. The Opportunity Solution Tree starts from a desired outcome, maps the opportunities under it in the customer’s words, then solutions, then experiments.
When to use it. When you are drowning in feature requests and every one sounds reasonable.
The test. For any feature in flight, name the opportunity it maps to and the evidence that put that opportunity on the tree. No evidence, no build.
Pricing and packaging
Monetizing Innovation
What it says. Have the price conversation during discovery, not at launch. Three questions give you the willingness-to-pay corridor: what would you expect to pay, at what price is it expensive but still worth buying, and at what price would you never consider it. Four failure types: feature shock (too much in one tier, fix by unbundling), minivation (right product priced too timidly, fix by raising), hidden gem (valuable capability never monetised, fix by packaging it), and undead (nobody would pay, fix by killing it).
When to use it. Before building something new, and any time close rates look suspiciously good.
The test. A 95%+ close rate is the minivation signal — the product is right and the price is too low. The free willingness-to-pay test designs a commitment test you can run in seven days.
Value metrics
What it says. A good value metric passes three tests: it aligns with the value the customer receives, it grows as the customer succeeds, and it is predictable enough to budget for. Pricing is a process, not a project — revisit quarterly, act every six to twelve months.
When to use it. When revenue does not grow as customers get more successful, or when buyers cannot predict their bill.
The test. Does a customer who doubles their success pay meaningfully more? If not, the metric is wrong.
Early-stage packaging
What it says. Good/better/best is the default at 0–$2M ARR, and a single plan is fine pre-PMF. Anchor the top tier high to make the middle look reasonable. Most SaaS at this stage underprices by two to four times, and a first price rise rarely raises churn measurably.
When to use it. Setting up a pricing page, or deciding whether you dare raise prices.
The test. If nobody complains about your price, it is too low. Quote the new price to new buyers and watch the close rate.
Channels and growth
The four fits
What it says. Growth needs four fits working together, not just product/market: market/product (the market pulls the product out of you), product/channel (the product shapes to the channel, never the reverse), channel/model (the channel must afford your price — low contract values cannot fund outbound, high ones waste on pure self-serve), and model/market (enough buyers at that price to hit the goal).
When to use it. When growth stalls and the instinct is to add another channel.
The test. Name which of the four broke before you change anything. Changing one element — raising prices, say — can silently break a fit two steps away. The free channel recommender runs the channel/model check against your own numbers.
Growth loops over funnels
What it says. Funnels consume inputs; loops compound them, because the output of one cycle feeds the next. Content brings users who create content; users invite collaborators who become users.
When to use it. When every new customer costs the same as the last one and nothing is compounding.
The test. Ask what in your product’s normal use creates the next user. If the answer is "nothing", growth will always be paid or pushed — which is a valid choice, but price it accordingly.
Demand creation vs demand capture
What it says. Capture harvests intent that already exists (search ads, review sites, intent data) and is finite. Creation builds intent where none existed (education, founder point of view, community) and compounds. Most early budgets over-invest in capture because it is measurable, then hit a ceiling.
When to use it. When paid cost per customer climbs while volume stays flat — the capture pool is drained and more budget will not fix it.
The test. Add a free-text "how did you hear about us?" field to signup and demo forms, and believe the humans over the attribution software.
The law of shitty clickthroughs
What it says. Every paid and owned channel decays as competitors discover it. A channel that works today is a depreciating asset, and paid acquisition is a treadmill: stop paying and growth stops.
When to use it. When deciding how much to invest in a channel that is currently working.
The test. Plan the next channel before the current one decays, and know which of your channels is a loop and which is a treadmill.
Activation, retention and product-market fit
Retention curve diagnosis
What it says. Read the cohort curve before prescribing anything. A curve that flattens at any level means fit exists for the segment that stuck — find them and reorient around them. A curve declining to zero means no fit for anyone yet, and no growth tactic fixes a product problem. A curve that flattens then decays means value was delivered but is not durable.
When to use it. Before any acquisition spend, and any time someone calls it a "growth problem".
The test. Which of the three shapes is yours? The answer changes the entire plan. The free PMF scorecard weights this above every other signal.
The Sean Ellis test
What it says. Ask users who have actually experienced the core value how they would feel if they could no longer use the product. 40% or more answering "very disappointed" is the widely used product-market fit threshold.
When to use it. Once enough customers have genuinely reached value — not at signup.
The test. Below 40%, segment rather than despair: the subgroup that would be very disappointed is your real ICP.
The Superhuman PMF engine
What it says. What to do with a score under 40%: ignore the "not disappointed" group entirely, build for the "very disappointed" group, and win over the "somewhat disappointed" only where their reasons overlap with what the loyal group already loves. Ask two follow-ups — what is the main benefit you get, and how could we improve it — then split the roadmap between deepening the benefit and removing the top blocker.
When to use it. When the score comes back mediocre and the instinct is to build for the complainers.
The test. Re-run it quarterly and track the percentage as a metric, not a one-off.
Bowling alley onboarding
What it says. Design onboarding as a straight line from signup to first value, with two rails of bumpers: in-product (checklists, progress, empty states that teach) and conversational (behaviour-triggered email, plus a human touch where the contract value justifies it).
When to use it. When you have signups but no activation.
The test. Count the steps between signup and the moment of value, then remove the removable ones. Settings, integrations and team invites all wait until after first value.
Activation as a repeatable event
What it says. Activation is not the aha moment — it is when a user has experienced the habitual value at least once and has a reason to come back. Retention decides which growth model you have earned; acquisition scales the retention you already have, good or bad.
When to use it. When defining what counts as an activated user, and before scaling any channel.
The test. Is your activation definition a repeatable event or a one-time setup step? Setup steps flatter the number and predict nothing.
Sales and hiring
Founder-led sales discipline
What it says. The founder sells first — not because it scales, but because every objection heard raw is product and positioning intelligence no rep will ever relay. Follow up until you get a yes or a no. Respond to inbound in minutes. Qualify hard and disqualify fast. Ask for the close.
When to use it. From customer one until the motion is repeatable.
The test. Most founders quit after one or two touches; most deals close after more. Count your own touches before blaming the channel.
Cold Calling 2.0 and role specialisation
What it says. Do not cold-call the target. Email above them asking who owns the relevant area; a forwarded introduction arrives with implied endorsement. Specialise roles — SDR prospects, AE closes, CSM retains — but only after founder-led sales works. Fifty hand-picked prospects with a personal opener beat five thousand generic emails.
When to use it. When you cannot get to a senior buyer, or when outreach volume is up and replies are not.
The test. Measure outbound by qualified opportunities created per month, not by activity volume.
The Sales Acceleration Formula
What it says. Treat hiring as an engineering problem. Build the ideal rep profile from your own closed-won data rather than gut feel; his top predictors were coachability first, then curiosity, prior success, intelligence and work ethic. Use the same documented ramp for every rep so you compare them on process, not luck.
When to use it. When you are ready to hire your first reps and want to be able to tell a people problem from a system problem.
The test. Test coachability in the interview: run a roleplay, give feedback, run it again, and watch whether they apply it. If rep one and rep two both miss, the problem is the system — ICP, pitch or pricing — not the people.
When founders can stop selling
What it says. Hire the first two AEs at the same time — one rep proves nothing, because you cannot tell whether a miss is the rep or the motion. Prerequisite: the founder has personally closed at least 10 customers of the same profile with a repeatable pitch. Founders step back from every deal around $1–1.5M ARR at the earliest, and a VP of Sales before about $1M is premature.
When to use it. The moment hiring a salesperson starts to feel overdue.
The test. Have you closed 10 of the same profile with a pitch you could hand over? The free first sales hire calculator runs this against your numbers.
Execution and focus
OKRs for a five-person startup
What it says. Run the minimum: one company objective per quarter, not five, with three measurable key results. Key results are outcomes ("20 customers at €500+ MRR"), never activities ("launch outbound campaign").
When to use it. At the start of every quarter, and any time everything feels equally urgent.
The test. Everyone should be able to recite the objective from memory. Around 70% attainment means you aimed right; 100% means you sandbagged.
The Mochary Method
What it says. An operating system for founder effectiveness: a weekly top goal with the first two hours of each day blocked for it, an energy audit of the last fortnight’s calendar, and significant decisions written up as issue-options-recommendation before the meeting rather than debated in it.
When to use it. When you are busy but not progressing.
The test. If you could only move one thing this quarter, which one? If you cannot answer, nothing is prioritised.
Lead bullets, not silver bullets
What it says. When you are losing deals on the product, there is no repositioning trick that substitutes for fixing the actual product and motion. Peacetime management suits margin for error; wartime — short runway, existential threat — demands speed, directness and one priority.
When to use it. When you catch yourself hunting for the clever workaround for the third week running.
The test. Name it out loud: is this a lead-bullets situation? If runway is under twelve months, decision speed beats decision perfection.
Which framework for which problem
The fastest way to use this page: find your symptom, read that framework first.
| The symptom | Start here | Then |
|---|---|---|
| Prospects say "interesting" and never buy | Obviously Awesome (Dunford) | Jobs to be done (Traynor) |
| Outreach gets no replies | Best-fit segmentation (Dunford) | Cold Calling 2.0 (Ross) |
| You cannot describe who you sell to | Best-fit segmentation (Dunford) | Lean Customer Development (Alvarez) |
| Everyone says yes to the price | Monetizing Innovation (Ramanujam) | Early-stage packaging (Rachitsky) |
| Revenue flat while customers succeed | Value metrics (Campbell) | Monetizing Innovation (Ramanujam) |
| Growth stalled, tempted to add a channel | The four fits (Balfour) | Demand creation vs capture (Walker) |
| Every customer costs the same as the last | Growth loops (Balfour) | The law of shitty clickthroughs (Chen) |
| Signups but nobody activates | Bowling alley onboarding (Bush) | Activation as a repeatable event (Verna) |
| Customers leave after a few months | Retention curve diagnosis (Winters) | The Sean Ellis test (Ellis) |
| Not sure you have product-market fit | The Sean Ellis test (Ellis) | The Superhuman engine (Vohra) |
| Wondering whether to hire a salesperson | When founders can stop selling (Lemkin) | The Sales Acceleration Formula (Roberge) |
| Busy every day, nothing moves | The Mochary Method (Mochary) | OKRs for a five-person startup (Doerr) |
| Losing deals on the product itself | Lead bullets (Horowitz) | Retention curve diagnosis (Winters) |
If you would rather have these applied to your own numbers than read them, the free go-to-market tools run several of them — positioning, ICP, pricing, channel, product-market fit and sales hiring — and return a verdict without an account.
What to ignore before €1M ARR
- LTV:CAC. Meaningless before roughly $1M ARR and twelve months of cohort data, because it divides by a churn rate you cannot yet estimate. Use CAC payback instead — under twelve months is good, under six excellent — which needs no churn assumption (David Skok).
- Category creation. A later-stage play. Subdivide an existing category instead.
- Partnerships. Below about $2M ARR the answer is usually "not yet": partners amplify a motion that already works rather than creating one. The exceptions are a product that only exists inside a marketplace, and integrations built for retention rather than acquisition.
- Full enterprise OKR ceremony. One objective, three key results, graded once. That is the whole thing at this size.
- Attribution software. It over-credits the last click and cannot see the Slack group where the decision happened. Ask people instead.
- Multi-channel anything. Before 10 paying customers, most B2B companies got there through the founder’s own network and manual outreach, not a channel (Lenny Rachitsky). Three channels at founder scale means none of them gets enough volume to teach you anything.
Who wrote this
Madalena Rugeroni built Mazo. She's an ex-Googler, a startup advisor and investor, and runs a portfolio of internet companies. Before that, as Head of Growth at Amplemarket, she scaled a B2B SaaS to $10M ARR. LinkedIn
FAQ
What is the best go-to-market framework for B2B SaaS?
How many GTM frameworks does a founder actually need?
In what order should I apply go-to-market frameworks?
Which frameworks should early-stage founders ignore?
How do I know a framework worked?
Frameworks are easy to read and hard to apply
Mazo applies them to your actual numbers: a diagnosis, a 90-day plan, one move at a time with a pass line, and a check-in every week. €99 a month, 14 days free, no credit card.
Start 14-day free trial Not ready? Try the free go-to-market tools, no account needed →Sources. Frameworks are credited to their authors throughout: April Dunford (Obviously Awesome), Andy Raskin, Christopher Lochhead (Play Bigger), Des Traynor, Cindy Alvarez (Lean Customer Development), Teresa Torres (Continuous Discovery Habits), Madhavan Ramanujam (Monetizing Innovation), Patrick Campbell, Lenny Rachitsky, Brian Balfour, Chris Walker, Andrew Chen, Casey Winters, Sean Ellis, Rahul Vohra, Wes Bush (Product-Led Growth), Elena Verna, Steli Efti, Aaron Ross (Predictable Revenue), Mark Roberge (The Sales Acceleration Formula), Jason Lemkin, John Doerr (Measure What Matters), Matt Mochary and Ben Horowitz (The Hard Thing About Hard Things). Mazo is not affiliated with or endorsed by any of them. Summaries are our reading of their published work, applied to the €0–€1M ARR stage; go to the originals for the full argument.