Startup Class Lecture 8 Notes – Doing Things That Don’t Scale

These are my notes from Lecture 8 of the Startup Class taught by Sam Altman. It involved short guest lectures by Stanley Tang on how he got started, Walker Williams on getting users and making them love your product and Justin Kan on getting press all around the central theme of doing things that don’t scale

Also, my notes from the previous lectures are here.

Stanley Tang on the origins of Doordash

  • Was interviewing a Macaroon store owner to learn about their problems and the owner showed them pages of orders she couldn’t fulfill because they were for deliveries and she didn’t have drivers
  • Interviewed 150 – 200 more store owners and learned that delivery infrastructure was a big problem.
  • Wanted to test the hypothesis, and so set up a landing page with pdf menus of restaurants in Palo Alto with a single phone number at the bottom. Wanted to see if people would call to place orders.
  • Launched this within in an hour, and within the week people started calling in and placing orders.
  • No drivers, no algorithms, no backend when they launched – just a landing page and they handled deliveries themselves early on.
  • No fancy systems early on: used Google docs to track orders, Square to charge customers, Find my friends to track where drivers were.
  • They were “doing things that didn’t scale”
  • Another benefit of this was that it allowed them to become experts in their business, from talking to customers and restaurants which allowed them to learn more about the problem to dispatching drivers manually which allowed them think through what their algorithms would look like, to doing customer support on their own thus receiving realtime feedback.
  • Three main learnings:
    • Test your hypotheses
    • Launch fast
    • Do things that don’t scale

Walker Williams of Teespring on doing things that don’t scale

Finding your first users

  • No silver bullet to user acquisition
  • First users are the hardest
  • They could come from anywhere, but a common thread is that founders spend personal time and effort
  • Don’t focus on ROI in the sense of time on those first users.
  • Those first users will take a lot of love and handholding, and that’s fine.

Turning users into champions

  • Delight the user with an experience they’ll remember
  • The best way to do this is to talk to your users
  • Nothing will give you a better sense of your product
  • Three ways to talk to your users
    • Run customer service yourself
    • Proactively reach out to current and churn customers
      • If a user leaves you want to reach out and find out why. Sometimes this brings them back but even if you can’t can avoid a similar situation being repeated for another user
    • Social media and communities
      • If users talk about bad experiences on social media, need to make it right.
  • Making it right is always important
  • One detractor can reverse the progress of 10 champions.

Finding product/market fit

  • The product you first launch with will almost certainly not be the one that takes you to scale
  • Goal is to iterate as fast as possible to reach the product that does have product/market fit
  • Optimize for speed over stability early on.
  • A good rule of thumb is to only worry about the next order of magnitude. So when you have 10 users, worry about getting to 100. When you have 100, worry about getting to 1000 and so on.

You want to do things that don’t scale as long as possible… This is one of your biggest advantages as a company, and the moment you give it up, you’re giving your competitors that are smaller and can still do these things, that advantage over you.

Justin Kan on getting press

  • Before you get press, think about who you want to reach as well as your actual goal.
  • That will determine which kind of press you will want to try to get
  • Stories that startups tend to get coverage for:
    • Product launches
    • Fundraising
    • Milestones/Metrics
    • Business Stories/Profiles
    • Stunts (when a startup does something different)
    • Hiring announcements
    • Articles you contribute on industry trends etc as a guest piece
  • Your press and news doesn’t have to be very original, but it just has to be original enough.
  • Press is good to get your first hundred/thousand users but it’s not a very scalable user acquisition strategy.

Mechanics of getting press

  • Think of it likes a sales funnel. Not everyone will convert.
  • First step is to think about a story.
  • Then, you want to get introduced to someone who might want to write about your story. Its much easier to get introduced than cold email.
    • Get entrepreneurs who were just written about introduce you to them. It’s easy for the entrepreneur, and for the reported, they are getting an introduction from someone who they thought was interesting enough to write about.
  • Contact them and give them enough time to write the story (1 week is reasonable)
  • Try to get a face to face meeting or at least a phone call. The more time they spend with you the more likely they are to write about you.
  • Then you want to pitch them. Think about what your ideal story will look like and have that in mind as you pitch them, as that way, when they take notes, it’ll be structured as you wanted.
  • Follow up a few days after with an email that thanks them for their time and includes any additional information you want such as images, videos etc
  • PR firms
    • they can help with contacts and logistics
    • they can’t help make your product interesting
    • they’re very expensive and probably not worth it early on
  • Keep in touch with people who write about you because they are more likely to do the same in the future
  • Pay it forward: help your fellow entrepreneurs get coverage because they’ll help you get coverage