Scaling Ramp to the Fastest-Growing
SaaS in History

Sri Batchu

Head of Growth, Ramp
(ex-Instacart, Opendoor)

JUN 25 2023

Core Concept

The B2B Growth Ladder

"Channels get more expensive as you go further along — but also more effective as you understand your customers."

Start with founder-led sales: learn to sell yourself first
Add salespeople, then low-cost content & community
PR before paid — credibility before spend
SEO & paid together, only once you know the customer

Framework

The Ramp Growth System: Tech × Data × Every Channel

4×

revenue growth in 1 year

2yr

to $100M ARR

<500

employees at that scale

Growth eng co-owns quota: engineers accountable for pipeline driven, not just product metrics
Technologize every channel: AI for lead scoring, outreach drafting, prospect prioritization — 2 years before it was trendy
Cap table as growth strategy: co-founders put early-stage founders on the cap table → first customers who loved and referred the product
Skunk works team: small cross-channel squad tests TikTok, Reddit, referral UX, and first-party events
PR as market moment: fundraise announcements packaged with unique business data makes each round a top-of-funnel event

Sri's secret sauce The distribution mix is not unusual — what's different is every channel is driven by technology and data. Ramp builds a competitive moat inside each channel, not just across them.

Growth System Deep Dive

North Stars, Translation Layers & Payback Period

The Two-Metric Rule

Every growth team needs two north stars: one for volume (motivating, intuitive, directly impactable) and one for efficiency (payback period). Revenue is too lagged; vanity metrics are too loose.

Instacart north star: Monthly Active Orderers. Ramp: Dollars of SQL Pipeline.

The Translation Layer

Sub-teams own local metrics (app load time, checkout conversion rate). A finance+data model converts each local metric into MAO or SQL pipeline impact — enabling cross-team resource allocation on a single currency. Updated every 6 months.

Why Payback Period beats CAC

CAC focuses on cost, ignores value — you can "win" CAC by acquiring cheaper, lower-quality customers
LTV/CAC is too assumption-laden for companies under 5 years old
Payback period uses contribution margin (revenue minus all variable costs including support) — grounded in recent reality
Example: $5K acquisition ÷ $500/mo contribution margin = 10-month payback

Velocity culture: days.ramp.com

Every board deck and all-hands shows the live day count since founding (Day 1,529.43…). The decimal is always ticking. Cultural message: work in days, not quarters. Never put off tomorrow what can ship today.

Tactics

Fail Conclusively & Hire on Slope

Fail conclusively, not just fast: in B2B you can't get large N, so maximize treatment effect — throw every tactic at the hypothesis at once, then cost-rationalize if it works
30% experiment success rate is normal — build culture where learning from failure counts as winning
MECE your problem space: before brainstorming, collectively exhaust all possible root causes (revenue slow? new acq? activation? retention? product line?)
Hire on slope, not intercept: invest in people growing fast; only hire when teams are genuinely stretched
Data-driven sourcing: use SimilarWeb traffic data to find which companies are actually best at the function you're hiring for

Calendar as strategy "You ship your calendar." Weekly calendar audits — personal and team — ensure time allocation reflects actual priorities. Ramp's People team has a template for it.

Contrarian

Growth Myths Sri Batchu Rejects

✗ Reduce CAC to improve growth efficiency INSTEAD → ✓ Minimize payback period on contribution margin. Obsessing over CAC brings in cheaper, lower-value customers — you win the metric and lose the business.

✗ Team structure is what makes a great growth engine INSTEAD → ✓ Culture, rituals, and cadences matter far more than org design. A team that hypothesizes, evaluates quickly, and ships MVPs beats any org chart.

✗ Fail fast — just run small quick experiments INSTEAD → ✓ Fail conclusively. In B2B you rarely get statistical power from small N. Maximize treatment effect so a negative result closes the door permanently — not just delays the same debate.

✗ Compress comp bands for fairness INSTEAD → ✓ Design for 10× comp for 10× operators. Small elite teams outperform large average ones. Equal bands are a talent retention trap that slowly degrades your talent density.