How to Forecast Sales for a New Product with No History

You can forecast sales for your new product like everybody else does: by estimating and making educated guesses. It may be hard to forecast, but it's even harder to run a business without a forecast. You don't have to spend thousands on market research, although that is a nice shortcut if you have

Nobody knows the future; we are all just guessing. You don't need a degree or mathematical prowess or software to forecast your sales. I spent several years as a vice president at Creative Strategies doing new product forecasts. What I saw was that people in the industry had a very good idea of what was realistic. Yes, there are mathematical and statistical routines that can automatically project past data forward into the future, but they are all based on the assumption that the past predicts the future.

Start your forecast by breaking it down into its component parts. It is easier to estimate the pieces and add up to the whole. Break down your business into units, then project units and average prices separately. Multiply units times price to calculate sales. That also makes it easier to go back to a forecast and modify either the units or the price when you find out which seems wrong.

Most businesses divide into units even if they don't realize they do. Attorneys and accountants, classic service businesses, sell time in quarter-hour units. Consultants sell days or projects or hours. Taxis sell pickups and rides and quarter miles. Restaurants sell meals and drinks.

Breaking things down into units makes forecasting easier. For example, I have no idea how much a new restaurant could sell. But if I saw it, I could count the tables or possibly the seats. Then I'd estimate how many people can fit per table and how many tables are filled at peak hours. From there I'd calculate how many peak hours there are per day and how many days per month. I might differentiate weekend hours from weekday hours. Then I'd estimate the average consumption per person.

This simple finite math creates a forecast. Better yet, especially when I'm guessing for a new business, it creates a rationale for that forecast. I can use that rationale to explain the forecast as part of the plan, and when I implement the plan, I can use that reasoning to catch where the forecast was off.

Sometimes you can find pieces of information that can help you if you put them together like a crossword puzzle. Do a thorough Internet search for expert forecasts in your industry area, and with luck, you'll find an expert forecast in a trade magazine or newspaper. Sometimes you can find industrywide information in a trade association directory or annual report of a related company. For my restaurant example, I'd go to Business.gov to get economic census data on average sales per restaurant in my county. You can't always find a forecast with research, but you can sometimes find a reality check.

Forecasts will not be accurate. Accepting that will help you get over the block and start developing numbers. Do the best you can and then concentrate on tracking plan vs. actual. Then you can catch what is wrong with the forecast and revise it as quickly as possible. Expect to be constantly revising your forecasts -- that is called management. Your plan is just the beginning of the follow-up, tracking, and revisions.