When you’re out there building a business, the last thing you want to do is waste time. Between building product, finding new customers, and the constant search for qualified new employees, the last things on most young executives’ to-do lists is updating excel models. In my startup experience, budgeting and planning was always characterized as something big companies do after they lose their edge. Not something to waste time on in the early days.
The reality isn’t so clear cut. Wasting time in spreadsheets is as bad as wasting time anywhere. But managing to the numbers and the excel wizardry that comes along with it is often what separates great startups from failed ones. The challenge for entrepreneurs is that - more often than not - people jockeying the spreadsheets aren’t doing it in a useful way. Instead of trying to expose underlying business risk, challenge assumptions around hiring plans, and managing cash, they’re trying to put together pretty charts and tables to placate their managers and investors.
When it’s done right, modeling helps you expose critical business issues and manage them. It tells you what to monitor, what to ignore, and where the threats lie. When it’s done wrong, it keeps your eyes closed to larger issues, it confuses your advisors, and it wastes your time.
Knowing how to make it count can be immensely valuable. So here are a few things that keep me focused on the useful stuff when I’m playing around in Excel, in case it’s helpful.
3 things to strive for when jockeying the numbers.
Know the critical business questions you need to answer
Modeling is useful because it sheds light on what could be. If you’re building out a budget for your next year’s operations, modeling is meant to help you manage your cash. If you’re building a sales plan, it’s meant to help you plan for hiring. To get any use out of the exercise, you need to know how the model will help you make decisions. If someone asks for a projection and you don’t understand how it could be useful, dig deeper. Push them to articulate what the key business question is and how the model will help you make a decision. If you have halfway decent investors, managers, or advisors, there will be a point to the exercise. And if there isn’t a key business issue on the other end of the question, ignore the request.
Build in the complexity you need to manage
If your models don’t reflect reality in the slightest, they’re not useful. My favorite example of this comes in the form of sales planning. Every sales manager under the sun knows that salespeople aren’t productive on day one. But for some reason, early sales managers will consistently turn in hiring plans and sales projections that don’t account for ramp time appropriately. The decision to remove the ramp might simplify the model, but it makes it entirely useless in helping you hit your numbers. Whether it’s forecasting deliveries with no buffer or looking at working capital costs, you need to build in some complexity for models to help you make actual business decisions.
Have an action plan for follow-up in mind
A model helps you build an opinion. It’s not all seeing or all knowing. Have an action plan for follow-up from the beginning. Make sure you use the numbers to hone your perspective and you take action accordingly. If you’re just putting numbers together for the sake of putting numbers together, odds are you’re wasting your time.
3 things to avoid when diving into spreadsheets.
If you have an analyst with any sort of banking or consulting background, they’re likely to get “cutsie.” They build all sorts of dynamic capabilities into their models and layer on tens or hundreds of assumptions. The challenge is that when all of these unknowns enter a model, any one of them can lead you astray. You need the complexity in the model that you know exists; things like sales ramp times. But don’t fool yourself into thinking that an assumption is the same thing as a known. It’s not. And any unknown can throw you off. Especially the ones you layer in and don’t keep in mind.Insensitivity
If false precision is rampant, it’s partner in crime is most certainly insensitivity. Even when people know they have a number of assumptions, it’s rare to see people systematically see how sensitive their business is to a given assumption. For instance, whenever my CEOs are going through a financial planning process, I force them to change their assumptions around cash collections. Whatever their policy, it’s not set in stone. And modeling out sensitivities around something so simple often shows them just how valuable it can be to their business to collect payment up front - even at the cost of a substantial discount. When you simply call out an assumption in a table, but don’t test your business’ sensitivity to whatever number you plug in the hole, you’re missing a valuable opportunity.
Finally, the killer of all models is unrealistic expectations. As they say, garbage in, garbage out. Too often do we just plug unrealistic numbers into our spreadsheets to placate stakeholders. When that happens, we cheat ourselves of whatever insight they might hold. Push yourselves to put good stuff in whatever model your building. Otherwise you’re just wasting time.