Cupcakes & T-Shirts: How To Build A 9-Figure SaaS Product

When it comes to building products customers love, I’ve found there are two really important ways you need to approach product design and development.

I’ve used these two concepts to build products (mainly SaaS) that have been paid for by well over 100,000 businesses and generated hundreds of millions of dollars in revenue (including BigCommerce, which I co-founded in 2009).

I’m talking specifically from a SaaS context here, but this should also work for companies who produce physical goods (lean manufacturing, etc).

Here they are…

Build A Cupcake First

If we use the idea of cakes, most startups go out and try to build a 3-tier wedding cake as version 1.0 of their product.

It takes months, is expensive and without any customer feedback, you’ve probably gone way off course even before making a dollar.

Instead, make your version 1.0 a cupcake — and ideally a cupcake that tastes REALLY GOOD to a small/niche part of the market that’s currently eating the (tasteless) wedding cake from your competitors.

Translation — focus on the pain you find in your market from customers of the incumbents (look at,,, etc) and focus your version 1.0 (AKA cupcake) around solving that pain.

An easy way to do this? Look at what people who rate your competitors a 1/5 or 2/5 say in their reviews. Find the commonalities and if there’s a willingness to pay for a better solution to that problem, focus a big part of your cupcake around that.

Once you’ve built a cupcake that tastes good (i.e. people are paying for it), then you can iterate version 1.1 of your product to be a single-tier cake. Version 1.2, a two-tier cake, version 2.0 a fancy wedding cake, etc.

Focusing on a cupcake gives you 3 things:

  1. A faster time-to-market
  2. More customer feedback, sooner
  3. Less guess work

You can use this speed and feedback to course correct and to put you on the right path to building a wedding cake that TASTES good instead of just looking pretty.

A key takeaway here is to optimize early for customer feedback NOT revenue. The currency you want in the first few months is the feedback required to achieve product-market fit, not get to $100K MRR by the end of this quarter.

T-Shirt Sizing

When you’re deciding what to put in the next version of your product, break all potential features/bug fixes/etc down by t-shirt size, as follows:

  • XL — a few weeks of effort
  • L — a week of effort
  • M — a few days of effort
  • S — a day of effort
  • XS — a few hours of effort

Work with your developers, designers, etc to get these estimates. Don’t come up with them yourself (if you’re a non-technical CEO like me, for example).

Now here’s the thing — anything that t-shirt sizes as an XL needs to be broken down into multiple Ls. “A few weeks” of effort doesn’t mean 2 weeks. It normally means 6+ weeks.

Why? Bad scoping and too many unknowns.

Break down large units of work into smaller units of work, starting with t-shirt sizing and you’ll find less scope creep and that your dev team gets a LOT better at hitting deadlines and due dates for releases.

I’ve been using these two simple product management strategies for over 5 years and they work incredibly well. I hope they help you build better products faster and with a lot less guess work.

Join over 25,000 smart founders on my email list and get one new post sent to you each week.

About Mitchell Harper

Mitch is a 7x company founder, advisor and investor. He is best known as the co-founder of BigCommerce. His companies have generated over $200,000,000 in total revenue and he is currently building an online education company and a SaaS company.

Read his story or follow him on Facebook and Twitter.