Sales managers perform a lot of tasks: They hire people. They onboard and train people. They segment territories and deliver quotas. They work on deals with their AEs—a task encompassing everything from strategy, demos, pricing, negotiation and more. They collaborate with marketing and customer success.

But one of the most critical things they do is forecast.

Some say forecasting is both art and science. And both methods are important. On the science side, it’s exciting to see tools like AtriumHQ, Clari, Salesloft, and Gong* help make forecasts more predictable and data driven. Some companies also employ data-science teams to analyze their data and predict outcomes.

In addition to forecast tools and data-science teams, there are two other, highly analytical ways most sales managers build their forecast: top-down conversions and bottoms-up calculation.

How to calculate tops-down and bottoms-up sales forecasts

Top-down forecasting involves applying conversion rates against your pipeline. There are a few ways to do this, but the most common one is to apply your conversion rate against either forecast stages (for example: commit, best case, pipeline) or deal stage (for example: discovery, demo, active evaluation, proposal/negotiation). Look at your past months and/or quarters to see your conversion rates against each stage and apply those figures to your current pipeline. This simple formula will give you your first forecasting data point.

Bottoms-up is where you inspect each deal individually and determine which deals are likely to close. You summarize at the end and see if that figure is close to your tops-down calculation. This is your second data point.

In my experience, the more pieces of data I can collect, the more precise I can get with forecasting. These two techniques are time-tested and easy to do.

How to crowdsource your sales forecast

For companies that may not be able to hire analysts or buy another tool, there is an additional way to get another slice of data: crowdsourcing your forecast.

This method only works if you meet the following requirements:

• You have at least five to six AEs on your team—otherwise you won’t generate enough unique data points
• You close deals monthly (this works best for companies with average deal cycles <60 days)

If you meet these criteria, I’d like to introduce a concept I call “The Bid”, which is an additional data point to guide you in forecasting.

The Bid Preparation:

• Schedule a team meeting around the 20^th of the month – this means you likely have about seven active-selling days to close the month out.
• Prep your team. Each team member needs to prepare and present deals that are in play for the month. Each deal gets a specific amount of time – say five minutes—to share and answer questions.

The Bid Presentation:

• Start the team meeting by repeating the purpose. Each AE will present ONLY their active deals for that month. Each deal gets five minutes, so be crisp.
• Share how much in bookings is closed currently and how many deals.
• At the end of the session, you’ll ask each AE to “bid” what they think the final closed bookings will be for the month as well as the number of deals.
• Announce that you as the leader will not be asking any questions. It is up to the AEs to inspect the deals.
• To encourage active participation, offer a prize. The AE closest to the final bookings number might earn a $250 gift card. If there is a tie on the bookings number, then the number of deals can serve the tie breaker. Quick note on the prize – for the first meeting, make this prize as large as you can afford to get the AEs’ attention. The first time I did it, I offered a $250 restaurant gift card and a limo for the evening.
• Sit back and watch the magic!

This session is an experiment of human behavior. I have led sessions where the AEs are ruthless! They will ask questions you forgot and will challenge and test the fences like you’ve never seen. What’s more, they will be brutally honest. They may downgrade the size of the deal, they may communicate they think the deal won’t happen this month (or at all!). But most importantly they will share ideas of what the AE running the deal can do to get the deal closed. There’s a lot of “should-ing” on people…” you should do this”, “you should do that”. And more than once, I’ve seen an AE who took a shellacking in this session immediately go out and close the deal just out of spite and competitive spirit!

A few additional benefits arise from this process:

• If you have five AEs, you get five inputs on what the group forecast will be. AEs may have happy ears, but it’s a funny thing when they’re not pitching their own deals. I would look at my high and low bids, and I would also average them. It turns out that the average of my AEs was stunningly accurate – normally within 5-10% of actual.
• You now have a third data point to use in determining your forecast.
• You will see sales leaders emerge in this session. We all try to develop our people and know some harbor management aspirations. Here’s a chance for them to “do the job before they get it”. I’ve even delegated this session to different AEs from month to month. Ask that month’s person to send out the invite, share the prep doc, run the session, and collate all the data. It’s a great way to see future leaders in action.

To summarize, forecasts are about collecting data and calculating probabilities. The more ways you can triangulate data, the better you can become at predicting the outcome.