Forecasting sales can be wildly difficult in even the best of times. And in the midst of a wobbly economic recovery following a substantial recession, projecting revenue takes on an even more Herculean quality.

Most CEOs got a nasty reality check toward the end of 2008 when it became crystal clear that consumers were spending less and customers frequently bailed out of big contracts at the last minute. 'We were 15 minutes away from a $100,000 purchase order and we got a call saying ‘the boss said don't sign anything,'' recalls Seth Earley of Earley & Associates, an IT consulting firm in Stowe, Massachusetts.

Can you really plan for a scenario like that? It's tough. But the good news is that a number of smart CEOs we spoke to have taken what they've learned over the past three years and come up with some pretty solid ways to forecast revenue. What they all seem to have in common: a more conservative approach to forecasting; more frequent adjustments to those forecasts; and a cautiously optimistic outlook for 2011. Here are some of the forecasting strategies they're using:

Sales Forecasting: Update Constantly

At the beginning of each fiscal year, the leadership team at T3, a $35 million Austin-based advertising agency, takes a stab at forecasting sales for the entire year, but then adjusts those numbers every month based on input from eight to 10 account directors.

'We go to each of them every month and ask them to give us a 12-month projection,' says COO Lee Gaddis. 'We ask everyone to be conservative.'

At the end of every month, Charles Kiley, the firm's CFO, sits down with the account directors to compare invoices to the forecast, plots a trend line, and reports back to senior management, including chief executive Gay Gaddis. At that point, the team makes a 'buy, hold, or sell' decision on hiring and expenses. 'We think forecasting is one of the most critical things we do,' says Lee Gaddis. 'Employees are our single biggest expense, so we need to know what we have in the pipeline.'

Forecast for 2011: 'We're thinking the economy may ease a bit in 2011 but it's still going to be tough.  In times like these you have to make your own good luck.'

Sales Forecasting: Plan for the Best, Average and Worst

Orb Audio CEO Ethan Siegel makes three different spreadsheets for his high-end speaker manufacturing business. 

'We try to walk the fine line of making sure we are profitable if the worst case comes true, but also have enough product and staff to support the best case predictions,' says Siegel, whose company is based in New York City.

This year, the company grew by 30 percent, with growth fueled largely by foreign sales driven by a weak dollar. 'We never would have guessed that we'd grow that much,' says Siegel. 'But because we also planned for best case and built up inventory, we weren't caught with our pants down.'

In the short term, Siegel looks carefully at the month-to-month growth rate in the previous year to predict revenue in the coming months. 'It's been very accurate for us since speaker sales do tend to follow a seasonal pattern,' he says. 'People are more likely to spend their money on indoor entertainment in cold months.'

Forecast for 2011: 'Although we expect 2011 to have some challenging months, we also anticipate annual sales growth greater than 20 percent due to new product releases and strong international sales momentum.'

Sales Forecasting: Manage Variables Creatively

Every business deals with uncontrollable variables that can wreak havoc on sales forecasts. For Terry McBride, CEO of Burns & McBride Home Comfort, a  $25-million heating oil distributor in Wilmington, Delaware, those variables include the weather and the price of oil.

'Our forecasts were usually 7 to 8 percent off because we were overly optimistic,' says McBride. 'And we'd always blame it on the weather.'

To narrow such misses, McBride partnered with grad students at Marquette University to create proprietary software that measures the consumption of individual customers and plots it against the weather. The result: more accurate predictions on oil usage, lower delivery costs, and happier customers who are less likely to run out of oil unexpectedly. That all added up to sales forecasts that were off by as little as 1 or 2 percent.

Forecast for 2011: "We expect flat gross revenue sales and customer growth when compared to the prior year. The bright spot, however, is in HVAC retrofit sales area  (replacement of existing residential heating and A/C systems) where we project a significant lift (20 percent) from last year's levels."

Sales Forecasting: Measure Customer and Employee Satisfaction

This tip might feel a little easier said than done and possibly a headscratcher for the purposes of sales. But not for Spencer Brown, the CEO of RentAGreenBox in Costa Mesa, California. Brown, whose company makes moving boxes from recycled material and is on track to gross more than $3 million a year, doesn't believe in annual forecasting.

'Looking at past sales to determine future sales is a flawed model,' he says. 

Instead, he forecasts quarterly sales using his own formula, based on six years of experience. Ninety percent of his projection is equally based on the company's quality of customer service, which he measures through social media interaction, including Yelp ratings (the company has a five-star rating); the enthusiasm level of his employees and how efficiently they're serving customers; and the company's ability to stock enough inventory to meet every customer's needs. Lastly, he layers in a 10 percent 'unknown factor' based on his personal experience in the marketplace – conversations with shop owners, mechanics, plumbers, and farmers market vendors. If the first three factors are all on track, Brown forecasts 13 percent growth the next quarter, but he adjusts that up or down based on how those conversations go. "We're accurate within 1 percent every quarter,' he says.

Forecast for 2011: "We're going national in 2011 and our growth will be phenomenal!"

Sales Forecasting: Pick a Number, any Number

Jason Loeb, chief executive of Sudsies, a multi-million dollar dry cleaning company in Miami, is adamant about 'not letting the numbers drive my business.' Other people's numbers, that is. When the recession hit, revenues in the dry cleaning industry plummeted. But Loeb had his own ideas about where he wanted his own revenues to go.

'I made a forecast, then came up with a way to fulfill it,' he says. When the economy started tanking and others were panicking, he added new services, such as environmentally friendly and couture cleaning, and forked over the marketing dollars to promote them. 

'People who spend $7,000 on a suit need someone who knows how to take care of it,' he reasoned. The result: his sales were up 9 percent in 2009 and are on track for a 40 percent increase this year.

Forecast for 2011: 'We are planning for a 40 percent growth in 2011. We are finding new innovative ways to improve the customer experience which will continue to help attract new clients and retain our existing ones.'

Sales Forecasting: Final Thoughts

Forecasting involves a little art, a little science, and would definitely be a lot easier with a very large crystal ball. How you forecast depends a lot on what industry you're in, your company culture, and the number of variables that are just beyond your control. The good news: solid experience with forecasting sales in good times and bad should help everyone make more accurate predictions in the coming years.