When this question was first posed to me, my immediate response would have been “At the very beginning.” However, upon a little reflection I realized that the answer needs to be a little more nuanced and is not nearly as self-evident as I reckoned. I’d still say you should have a likely business model in mind, reasonably early in your startup’s life.
Of course, given the treacherous nature of the English language, it’s probably worth defining what a business model is.
Wikipedia defines it as
A business model describes the rationale of how an organization creates, delivers, and captures value
and in plain English, according to the fine folks at Walden University,
The term “business model” refers both to the way in which a business creates a product or service and sells it for a profit, and to any document that outlines the process. …The business model of a lemonade stand seems straightforward: make lemonade and sell it to passersby for profit.
Given all the talk of lean startups, I’ve met a number of folks who talk of figuring out their business model even as they roll out their product. Yet a number of sane heads seem to be be reminding us rightly, that a feature ain’t a product ain’t a business. To me fundamentally a business model is figuring out who is going to pay (& hopefully how much) for the service/product we offer. It could be the users, it could be their parents (in the case of children), their companies (for corporate/SMB), advertisers (for free consumer services) or the government – someone, somewhere is going to write a check or whip out a credit card or cash for your product or service.
Once you have a business model you can then figure out how much money you will actually make – your revenues. Done right, you hopefully know what is it going to cost you to deliver it and how long before you can get it to operational breakeven (making more than you spend on a quarterly/monthly/weekly basis.) Of course if the numbers don’t add up in your revenue model, you return to your business model to tweak your assumptions, targets etc., till you can iterate to something that looks like a practical (or at least realistic) business model.
“There’s not a single business model… There are really
a lot of opportunities and a lot of options and
we just have to discover all of them.” Tim O’Reilly
Back to the question of when do you need this? I’d still say start with something as simple as “We’ll provide P for Q, which {Q,R} will pay for in the form of S.
A couple of examples might help illustrate this.
We’ll
“enable consumers to discover relevant videos easily and we’ll make money through advertising” or
“answer academic questions over SMS for high-schoolers. Parents will pay for the service through subscription.”
“help solar installations achieve higher ROI. OEMs will license our technology or purchase our blackbox.“
This is more a statement of intent than a full fledged business model. Before you go spend a lot of time, figure out whether the world (or at least your target customer group) actually has the problem you reckon you’re solving for them and is willing to pay for it. Once the market need and your product/service fit to that need is established you can return to working up a real business model, that will allow you to build a real business.
Related articles
- How to Develop a Business Model (thinkup.waldenu.edu)
- The Business Model Canvas
You must be logged in to post a comment.