My notes on Startup Business Models and Pricing with YC Group Partner Aaron Epstein
Previous Blog Post:
[YC Startup School] Notes on “Should you start a startup?”[YC Startup School] Key takeaways from “How to get and evaluate ideas”[YC Startup School] How to talk to UsersNext Blog Post:
[YC Startup School] How to get your first customers9 business models of (nearly) every billion-dollar company
SaaS (Software as a Service)
Cloud-based subscription software
PRIMARY METRICS:
- Monthly Recurring Revenue (MRR) or Annual Recurring Revenue (ARR)
- Growth Rate: Measured weekly or monthly
- Net Revenue Retention: % of recurring revenue retained from a prior period
- CAC: Costs to acquire a new customer
TAKEAWAYS:
- All the benefits of recurring revenue
- Can have non-recurring revenue, but don’t include in ARR/MRR
- Usually sold to businesses, ideally on annual contracts
- Growth can be driven by direct sales, self-serve acquisition channels, or both
Transactional
Facilitate transactions and take a cut
PRIMARY METRICS:
- Gross Transaction Value (GTV): Total payment volume transacted
- Net Revenue: Fees charged for transactions (often a %)
- User Retention: % of month 1 customers that make a purchase in month 2, etc
- CAC
TAKEAWAYS:
- Usually fintech and payments businesses
- One-time transactions rather than recurring
- Often high volume with a low fee (1-3% is common)
- Best transactional businesses have extremely consistent revenue from high repeat usage
Marketplaces
Facilitate transactions between buyers and sellers
PRIMARY METRICS:
- Gross Merchandise Value (GMV): Total sales volume transacted
- Net Revenue: Fees charged for transactions (often a % take rate)
- Growth Rate
- User Retention: % of month 1 customers that make a purchase in month 2, etc
TAKEAWAYS:
- Hard to get off the ground, chicken & egg problem
- Need to scale supply and demand in sync
- Network effects at scale drive exponential growth
- When they work, often become dominant winner-take-all winners
Subscription
Product or service sold on a recurring basis, usually to consumers
PRIMARY METRICS:
- Monthly Recurring Revenue (MRR) or Annual Recurring Revenue (ARR)
- Growth Rate: Measured weekly or monthly
- User Retention: % of month 1 customers that make a purchase in month 2, etc
- CAC
TAKEAWAYS:
- Recurring revenue is the most valuable revenue
- Usually sold to consumers, often paying monthly
- Usually lower price points, from a higher volume of customers
- Growth driven by scalable, self-serve acquisition channels
Enterprise
Sell large fixed-term contracts to big companies (5k+ employees)
PRIMARY METRICS:
- Bookings: Total signed contract value (recurring + non-recurring)
- Revenue: Recognized when delivering on the contract
- Annual Contract Value (ACV): Total contract value / # of years
- Pipeline: Top of funnel → Demo → Close
TAKEAWAYS:
- Very few customers, much larger deals ($100k+/year)
- Growth driven by direct sales
- Often begin with paid pilots or LOIs
- Usually long sales cycles, with many gatekeepers
- The buyer is not always the end user
- Lumpy growth: measuring m/m growth rate doesn't make as much sense
Usage Based
Pay-as-you-go based on consumption in a given period
PRIMARY METRICS:
- Monthly Revenue (not recurring!)
- Growth Rate
- Revenue Retention: % of revenue from last month’s customers in this month
- Gross Margin: Revenue - Cost of Goods Sold (COGS)
TAKEAWAYS:
- Don’t confuse usage-based revenue with recurring revenue
- Charge per API request, # of records, data usage, etc
- Grow as your customers grow
- Product and pricing scale to support tiny startups to large enterprises
E-commerce
Sell products online
PRIMARY METRICS:
- Monthly Revenue: Total sales
- Growth Rate: Measured weekly or monthly
- Gross Margin/Unit Economics: Revenue - Cost of Goods Sold (COGS)
- CAC
TAKEAWAYS:
- Includes D2C brands and Shopify stores
- Not marketplaces, so keep 100% of each sale
- Higher COGS = lower margins
- Products often commoditized
- Need to be excellent at user acquisition and operations/unit economics
Advertising
Sell ads to monetize free users
PRIMARY METRICS:
- Daily Active Users (DAU): Unique users active in a 24 hour period
- Monthly Active Users (MAU): Unique users active in a 28 day period
- User Retention: % of active users on D1/7/30/etc
- CPM (Cost Per Thousand) or CPC (Cost Per Click)
TAKEAWAYS:
- Typically consumer social products with huge scale
- Customer is the advertiser, not the end user
- Users are the product being sold
- Need billions of impressions each month
- Registered Users is a vanity metric
Hardtech/Bio/Moonshots
Hard businesses with lots of technical risk and long time horizons
PRIMARY METRICS:
- Milestones: Progress towards the long-term vision
- Signed contracts
- Letters of Intent (LOIs): Non-binding contracts indicating interest to purchase
TAKEAWAYS:
- Often take years to get to a live product because of technical and/or regulatory risk
- Impressive technical milestones or experimental data can de-risk the tech
- Revenue is often years away, so signed LOIs are usually the best way to show customer interest
Business model lessons from top 100 YC companies
Interesting points
SaaS business → 31% of the top 100 YC companies (recurring revenue make for great businesses)
Transactional → 22% of the top 100 YC companies but create 29% of the overall value
Marketplaces → 14% of the top 100 YC companies but create 30% of the overall value
5 of the top 10 YC Companies are Marketplaces. But tough to get off the ground because you have to supply for both sides of the marketplace: supply and the demand
Only 3% of the top 100 YC companies are advertising because they need organic virality (Twitch, Reddit)
Don’t sell ads unless you will be a top 10 site on the internet.
What’s not in the top 100
→ Services or consulting businesses
Problems:
- Non-recurring revenue
- Scaling with people
- Low margins
→ Affiliate businesses
Problems:
- Too far away from the transaction
→ Hardware businesses
Problems:
- Requires lot of capital
- Low margins
→ Businesses built on other platforms
Problems:
- Platform risk if the business gets shut down
Recurring revenue consistently creates winners
- High predictability
- Higher customer Life Time Values (LTVs)
- Lower Customer Acquisition Costs (CACs)
Biggest winners built with moats
- Network effects
- Lock-in or high switching costs
- Technical innovations: Cruise with self-driving cars
- Higher margins or better unit economics
- Organic distribution through virality or word-of-mouth
Characteristics of the best businesses
- Generate recurring revenue
- High retention rate
- Build defensible moats
- Close to the transaction
- Scale with software, not people
- Proven and business models that are familiar to customers
Startup pricing insights
What pricing can teach you:
- Who wants your product
- How much they want it
- How much value your product provides
- Which channels you can use to acquire customers
You should charge for your product or service!
→ Can learn about whether your users are willing to pay or not
→ Find the right “order of magnitude”
→ Pricing isn’t permanent, will change over time so don’t overthink it
Price on perceived value, not costs
How to find your value?
→ Talk to your users: “What is the problem that you are hoping our product could solve?”
- make more money
- reduce costs
- move faster
- avoid risk
→ Keep raising prices until you get pushback.
Ideal price
is when customers complain but they still pay→ Don’t undercharge that your competitor. Don’t have price as the only differentiator.
Pricing implies value
When it is okay to give a lower price
→ for your first user
→ for a valuable logo (recognizable logos for social proof)
→ if you get lock-in
→ if you can renew at a higher price
Keep your pricing page simple!
Example: GitLab