You’ll be hard pressed to find a sales person who can’t quote at least a line or two from Glengarry Glen Ross. While we love the movie because it shines a spotlight on some of the more absurd aspects of sales, too many companies have taken the “Coffee’s for Closers” speech seriously. They lean on reps until the forecast goes up, not realising that such high pressure tactics don’t move the needle on what counts—closing deals. The results?
According to CSO Insights' 2013 Sales Performance Optimisation Study, the average company closes 46.5 percent of the opportunities they forecast. And according to a 20 year longitudinal study, forecast accuracy has dropped over the past two decades from 84% to 76%. Given the far-reaching impact of a broken forecast (scuttled resource plans, reduced margins and declining market cap), companies that struggle in this area need to move beyond a simplistic “no excuses” mindset and adapt a much broader view of what it takes to operationalise a bullet-proof forecasting process.
A New Approach
In the age of big data and predictive analytics, sales organisations have an unprecedented opportunity to blend new technology with the experience of seasoned professionals to realise outstanding results. Here are a few examples of how leading companies are outperforming their peers by integrating next generation forecasting tools into sales processes that emphasise both accuracy and productivity.
- Automate the process—Although sales organisations began to adopt SFA over twenty-five years ago, today over 93% are still forecasting on spreadsheets. As I meet with Fortune 500 accounts, I frequently ask how many analysts they have on staff to roll up spreadsheets – “10 or more” answers aren’t uncommon. However, as business environments get more complex with global, multiple offerings, sophisticated customers, and complex sales channels the manual processes simply don’t yield accurate results.
Fortunately, automating the forecasting process delivers a two-fold benefit: It allows companies to expend fewer resources and it drives greater accuracy. To start, companies need to define a submission cadence and then rely on software to enforce it. It’s also important to track and publicise forecast accuracy over time. These steps increase participation rates and improve accountability. This leads to companies getting a broader and deeper perspective on their pipeline with fewer resources expended.
- Top down meets bottom up—Companies that only take a top-down approach to forecasting (forecast revenue by applying historical close rates to the current pipeline) don’t account for variance in the quality of the pipeline from one quarter to the next (e.g. if this quarter’s deals are less solid than last quarter’s, I can’t apply historical close rates). On the other hand, if companies only take a bottom-up approach (build their forecast by aggregating the individual deals that are currently in pipe) they fail to account for the deals that unexpectedly slip into the forecast later in the quarter. This is a particularly relevant to businesses that have high transaction volumes and short sales cycles.
The best approach is to use data science to blend bottom-up and top-down forecasting. Bottom-up enterprises can use data science to identify the deals that will close and that should be included in the forecast. Then for top-down analysis, predictive algorithms can analyse the size and quality of the current pipeline in conjunction with external factors, such as macro-economic trends, accounting for deals that have not yet hit the pipe but will ultimately contribute to the quarterly achieve. By employing this hybrid top-down/bottom-up approach, companies are often surprised to discover that they can increase forecast accuracy by up to 80%.
- Calibrate the sales team’s roll-up—By the time the forecast number rolls up to the most senior levels of a company it includes layers-upon-layers of judgment. Using data science, companies can “calibrate” each tier of the forecast. Algorithms can assess historical forecast accuracy in conjunction with current pipeline strength and then re-factor the numbers that sales teams are submitting. Whether reps are overly optimistic or prone to sand bagging, they tend to be consistent in their biases. Data science provides the corrective lens necessary to “dial in” the real number.
Over the past 20 years, the volume of customer data that companies have collected has increased by over 1500%! Additionally, sales organisations now have access to tools that that surface insights to accelerate wins and gain a clear line of sight into what will close. There’s no longer any excuse for employing manual pipeline management and forecasting processes that divert attention away from selling and deliver inferior results. By automating critical tasks and using data science to analyse deals and formulate a reliable forecast, sales organisations can put time once lost to administrative minutia to a more productive use: winning deals.
By Justin Shriber, Vice President of Products, C9
