YC Startup School Notes: Setting KPIs and Goals

Table of Contents

If you are startup founder, nobody tells you how you should spend your time
It’s not worth to optimize every metrics
You want to get to product market fit ASAP, prioritization is crucial
KPI
- Key performance indicator
- Measuring what matters
Prioritization
- Infinite things we can do each day, work is never done
- How i spend the day, how team spends to day
Prioritize tasks that move you toward your KPIs Faster
Choose right KPIs and be honest with ourselves
Don’t focus on Vanity metrics
- Vanity metrics are data points that may seem impressive at first glance but don’t necessarily provide meaningful insights into the performance of a startup or its progress toward its goals. Here are a few examples:
  - Total Downloads/Installs: This metric might seem impressive, but it doesn’t necessarily reflect active users or engagement with the product.
  - Website Traffic: High traffic numbers might seem like a good sign, but if those visitors aren’t converting into customers or engaging with the product, it’s not a reliable indicator of success.
KPIs and prioritization are important since nobody tells you as a founder how to prioritize your time. It’s easy to feel productive and not move business forward
Examples of feel productive and not move business forward
- Optimizing paperwork tasks
- Unnecessary perfectionism
- Premature optimization
- picking hard problem and Not building what your users want
It’s easy to run fast in wrong direction
How to prioritize
- speed matters
- align iwth your cofounder
Step 1: Indentify top KPIs
- if you launched, it should be revenue growth
- non revenue KPI is rarely the good one
- what is the KPI goal for next week
  - E.g. get 10 customers
- Early growth compounds
Step 2: Identify biggest bottleneck
- SuprDaily company
  - Aka bolt foods
  - Top KPI
    - Growth, bottlenecked by conversion late in the funnel
  - Biggest problem
    - Why are high intent users not converting?
  - Discovered they did not have milk brand so users did not finish the order.
Simple system to move your KPI(s)
- 01 Write down ideas that may help
  - Rank & choose a few (rank few and subrank according to how long it will take)
- 02 If your KPI doesn’t move, be honest on why
- 03 Do honest retros; learn and adjust course
  - If it wasn’t working, do something different
  - If it was working, double down
- 04 Move fast
  - Learn and do something differently if the stuff so far did not work. The definition of insanity is doing over and over again the same stuff and expect different results.
  - don’t let indicision to block you, just choose one path and go with it.
Things that should be on your list
- Talking to users / responding to customer support messages
  - the only way you know what customers want is by talking with them all the time
- Building features you know your customers will pay for
- Getting users to pay for what you’ve built
Things that should probably not be on your list AKA fake progress
- Investor “coffees”
- Conferences
- Arbitrary technical milestones
  - AKA launching Android app unless its smth users explicitly asked for
Common mental traps (things that makes are feel good but does not help with Product market fit)
- Checking things off a list feels good (e.g. optimizing paperwork about processes)
- It’s easy to convince yourself something is working when it’s not (easy to mistake slow growth for PMF)
- Perfectionism and indecision are your enemy: fast decisions are good decisions
  - Most decisions does not matter, and for those that do it’s okay to decide wrong and fix it later, just keep moving
  - Nobody make quick decisions all the time, second best option pretty good decisions quickly and if wrong, fix
  - don’t waste time on decisions that does not get you to PMF
  - if it’s a tough call, it actually probably measn you can’t go wrong
- Mitigating downside (rather than chasing upside) can feel falsely urgent
  - Downside risk protection is straightforward, and fixing small problems can be incredibly gratifying.
- Working on secondary problems instead of the existential one
Prioritization Recap
- You’ll never get to everything on your task list
- Use KPIs to prioritize your work
- Be honest with yourself; fail fast
How to choose good KPIs
- Primary Metric
  - This is the metric that unequivocally tells you if your business is working
  - Most commonlv revenue growth
- Secondary Metric(s)
  - Things that need to be tracked to make sure you’re not cheating/gaming your primary metric
  - Examples
    - Retention/churn
    - Unit economics
    - Acquisition cost and payback period
  - Vanity Metrics
    - Amount raised
    - Team size
    - Office space
    - Press hits
    - Celebrity endorsements
    - Instagram likes
Main message: Choose your primary KPIs
How do we set targets? How much growth is enough?
- 5-7% WoW (Week over Week) = good
- 10% WoW = exceptional
Early growth matters
- Growth rate compounds
Factors to consider
- Latent demand may boosts early growth
- Length of sales cycle
- Organic vs. paid
- Retention/Engagement
Customers that placed 5+ orders are muuuuuuch better metric than signups
when setting targets, two approaches
- Top down
  - Pick a milestone/date in the future that you want/need to hit
  - Work backwards from this for bi-weekly goals
- Bottoms up
  - What is realistic for you to get done next week?
  - Work up from this to determine a realistic future milestone
non-revenue kpis, might be tempting but not right
- CAC/LTV - customer acquisition cost / lifetime value
  - Focus on payback period
  - post PMF
  - now worry at the start only at payback period
- Free signups / daily active users (DAUs)
  - only exception is if your product requires a network effect
Exception to revenue KPIs
- Hard tech, rockets, biomed..