Financial models for startups

Glenn Kelman, by way of Guy Kawasaki’s blog, shows how he built a financial model for his startup, Redfin. This is a great example of how to create a forecast and manage the expectations of investors. At the end of the post, he shares a number of modeling tips. The one I like?

Flag your assumptions: Rather than burying your assumptions in Excel formulae, call them out in a separate tab of the workbook, so that you have a control panel for adjusting the model. This is especially important if you plan to share your model with potential investors.

The control panel gives you the ability to fly by wire – a key concept of the adaptive enterprise. So if you have the chance to look at an entrepreneur’s Excel model and see that it’s inflexible or poorly designed; this just may be a flag that her company is brittle and in danger of extinction. Of course, this concept works the other way, too. Want proof? Take a look at Redfin’s website. Someone at Redfin knows something about design, don’t they?

Excel Tip: Most people I’ve worked with don’t know about my favorite feature in Excel – Defined Names. You can define descriptive names for a range of cells or a constant. When I show this feature to finance pros who mess with Excel for hours on end, they tell me how much easier it is to create models using names instead of meaningless cell addresses.

I’ve created an Excel model that takes the rent example from Glenn’s post and uses names and a control panel. See how easy it is to read the formula on the forecast tab. And with names you’re able to refer to a cell address on a different tab, without any thought of clumsy links. All input to the model is defined on the control panel, and in seconds the model can be customized for different kinds of analysis. Take it to the limit by using scenarios: best case, worst case... Let’s pull some positive G’s by using a macro to run the best case and populate all the inputs at the speed of sound. Now you really feel like you’re flying by wire.

tags: Presentation, Product Management

Comments

[Mike]

Hi Steve,

I like your simple model, but it misses one of the key learnings I've made from doing these sorts of things: Show the assumptions and the results on a single page!

[Steve]

Hi Mike,

Your post is great. Thanks for sharing.

You're right, if you can put assumptions on same page as results, everything is in front of customer, and it's a tighter feedback loop. With large pro formas it's more difficult to put a lot of assumptions one page, that's why Guy Kawasaki's blog talks about a control panel tab.

Your post also reminds me of the concepts in the book Serious Play. And with a usability calculator, it's like you're modeling with the customer so you can create a model for the customer.

Steve