TOP 10 Best FINANCIAL Practices for Startups | Chartered Financial Analyst Nathan Beckford

Build it bottoms up on a month-by-month basis. We talked about the month-by-month already. Bottoms up meaning literally at day 1, we are, like I said, hiring 1 person, that person is making 2 sales a day at $10 a sale, you know, it's kind of this bottom-up organic versus the alternative which is a top-down model which means this market for widgets is $2 billion, if we get 1%, we'll have X. That's the wrong way to do it. You really kind of want to do it from a bottoms-up basis. Identify key business drivers. I mentioned this already. Any business, even a fairly complex business, can usually be boiled down to about 2 or 3 things that really matter. A consumer internet business might be time on site, or an advertising consumer internet business might be number of users, time on site, retention, or frequency of visit, right? I mean, most businesses can be boiled down to a few different things. And so you need to spend some time really thinking about what drives your business and weave that into your model. We build the model around that. Use benchmarks and comps. You know, it's really cool. When I was building models 10 years ago, it was actually kind of hard to find anything. Nowadays with Quora and with this increasing exposure, I mean, people are blogging about their startups in a really granular way these days. You can get a lot of comp data out there. We used to have to use like 10-K filings from public companies and try and extrapolate that data to startups, which doesn't really make sense, but that was the only publicly available data out there 10 years ago. Now it's, you go on Quora and search around, you can find a lot of good stuff. But a good best practice is to use benchmarks and comps. And another part of using benchmarks and comps is, for example, if you've got a consumer internet business that relies on advertising, you can pretty easily go into Google and play around and see what average CPMs are, right? You can get this data without guessing at it these days. Focus on the first 2 years. We often build 5-year models when we're doing these financial models. But what's gonna happen in 5 years? I mean, who can really predict the future that far out? No one can, yet we still build what could be potential 5 years. But we really focus on the first 2 years, and we'll do the first 2 years on a month-by-month basis to get granular about it. Make it easy to change and edit. This is simply a function of building a dashboard in the model. We'll show you this a little bit later. But you want to be able to run scenarios in your model. So if you have a dashboard where you can easily, quickly change the numbers versus hard coding things into cells, it's what you should do. Build it to scale. This is simply a function of as your business scales, what scales along with it, right? So if you're, as a quick example, maybe your support, customer support headcount grows at a linear rate of 1 customer headcount or customer support rep for every 10,000 users, something like that. That's what I mean by building it to scale. Overscenario-ize, right? We usually build the model what we think is going to happen, our best guess, and then we take some haircuts at it, right? What if everything takes twice as long as we think it's going to to get to market? What if our— Our revenue assumptions are way less. What does that do to our burn rate? It's pretty easy. The beauty of a model is you can run all these different what-if scenarios without the pain and expense of living it in real life. Be comprehensive with headcount, especially— I don't know what types of businesses you all are running, but especially in the web world, headcount is the biggest expense for most early-stage startups. So pack more heads in there. It's good to be conservative when you're building a model. A way to do that is pack in heads that are going to be there. Multiple sanity checks. All this means is periodically step back from your model and say, does this look reasonable? Does this growth rate and ramp look like something we could do? Are there precedents in the market of companies growing as fast as what we're saying here. Same thing with margins, right? We're showing a profit margin or operating margin of X percent. Do other internet companies or do other SaaS businesses have margins like this? If not, you might want to go in and reassess a bit. And then finally, the cool thing about the model is it sort of bottom lines everything to cash flow. It sounds simple, but never run out of cache. If you have to scale back other things, you can do that in the model to make sure you don't run out of cache. So for a good—