Scaling SaaS startups: stages, metrics, and growth traps

Executive overview

Most SaaS startups stall not because they lack a good product, but because they try to scale before they've answered two foundational questions: what is it for, and who is it for. Each stage of growth demands a different leadership style, team composition, and operating model.

The baseball-diamond framework maps four stages — MVP, product-market fit, T2D3 scale, and the Rule of 40 — and shows why you cannot skip any of them.

Cheap early growth that defers hard problems (pricing, ICP discipline, service cost) accumulates into a crisis at scale.

The two questions every early-stage company must answer

• MVP stage answers "what is it for" — the pain point is real when users say "please don't stop building this"
• Product-market fit answers "who is it for" — people vote with their wallets and renew without prompting
• You cannot credibly define your ICP before reaching product-market fit; there is no proof point
• Crossing the chasm from early adopters to economically motivated buyers marks the PMF transition
• Only once both questions are answered can you "pour fuel on the fire" — and you now know which fire

The baseball diamond: four stages of SaaS growth

1. MVP — validate the problem; early adopters engage without paying
2. Product-market fit — customers pay, renew, and refer; ICP becomes definable
3. T2D3 scale — triple ARR two years in a row, then double for three more; requires parallel execution across five levers simultaneously
4. Rule of 40 — balance growth and profitability; replaces the pure-growth obsession that inflated the 2021 SaaS bubble

The five levers that must move in parallel during T2D3

• Diversify demand-generation channels beyond the single channel that got you to $5–7M ARR
• Grow revenue per customer — expand ACV, ARPU, or ASP
• Reduce churn; revenue retention becomes a primary focus
• Optimise the funnel on volume, speed, and conversion shape
• Build systems that scale without adding headcount linearly

Doing two or three of these well is enough to reach $5–7M. Doing all five simultaneously requires a fundamentally different leadership team and operating model.

TAM, SAM, and the discipline of SOM

• Most teams get TAM (total addressable market) and SAM (serviceable addressable market) right
• The harder and more important step is SOM — the obtainable portion of the serviceable market
• SOM discipline means choosing where friction is lowest and fan probability is highest
• Saying no to easy early wins is especially hard for small teams; skipping this step inflates service costs
• In SaaS, customer acquisition cost is only half the equation — cost to serve determines long-term profitability

The land-and-expand trap

• "Land and expand" made theoretical sense: sell cheap, expand later
• In practice, lowering the initial bar just moves the revenue problem downstream
• Customer success teams become de facto sales teams — expensive ones
• The economics rarely outperform a disciplined high-ACV initial sale
• This dynamic was a core driver of the 2021 SaaS bubble: growth metrics looked good while unit economics quietly worsened

Hero pricing: charging for value already delivered

• Many SaaS companies build extensive onboarding, education, and support — then give it away
• The "hero plus" plan adds a tier (typically 10–30% above the standard plan) that formalises existing extras
• New customers never knew they were getting it free; they accept the price at face value
• Existing hero customers are the easiest upsell because they already see the value
• Price increases justified by added capability are more defensible than inflation-based raises

Outbound marketing done right

• Outbound is not inherently spam — interruption happens in every human relationship
• The failure mode is irrelevance: messaging people with no signal they need your service
• The right model: identify prospects with visible pain, leave something worth keeping
• Patience matters; the largest Colungi client initially categorised outreach as spam and converted two years later
• Test subject lines, messages, timing, and channels systematically before scaling what works

ICP and content marketing

• Reaching product-market fit gives you a unique insight advantage over competitors — learn from paying customers
• Move from SAM to SOM by identifying where onboarding friction is lowest and retention is highest
• AI raises the content production bar but doesn't change the fundamental requirement: relevance to a precisely defined ICP
• If your ICP doesn't feel like you're part of their tribe, your targeting isn't tight enough

Leadership and the scaling identity problem

• The skills that get a company to $5M are not the skills that get it to $50M
• Founders who recognise they prefer starting over scaling create better outcomes by stepping aside early
• Microsoft's model: reorganise around talent, not just business need — structure follows people
• OKRs and quarterly rhythms (rocks, bowlers) create urgency without micromanagement
• The art of urgency: use numbers to enable and empower, not to crack the whip

Lessons from Akumatika (zero to $6M ARR)

• Stijn joined as CRO, not founder; built brand, team, and all commercial functions from scratch
• Hired scaling-oriented talent early — right for the next stage, harder for the startup phase
• Brought in a product marketing manager before it was "needed" — paid off at growth stage
• His successor inherited a team ready to execute T2D3; the early hiring decisions shaped what was possible later
• Key transition: from founder-led selling to repeatable, trainable sales motion

Rule of 40 and the post-bubble correction

• Pre-2021: growth at any cost; "land and expand"; defer profitability
• Post-correction: investors and public markets demanded real unit economics
• Rule of 40 (growth rate + profit margin ≥ 40%) is now the baseline expectation at scale
• Companies that pivoted fast to profitable growth are commanding significantly higher valuation multiples
• The shift from "capture land" to "earn profit" is structural, not cyclical