7 reasons why most B2B CRM systems get forecasting badly wrong…


Share on LinkedIn

How accurate are your sales forecasts? According to the latest research from CSO Insights, less than 50% of deals close as originally forecasted. A significant number never close at all. Think about it – the average sales forecast is no more accurate than tossing a coin. And let’s not even go near the idea of monkeys and Shakespeare.

Maybe perfect forecasting is an impossible goal – but it’s certainly possible to consistently get a great deal closer, as many best-in-class organisations have shown. Here are 7 reasons why your CRM system might not be helping.

1. Using the out-of-the box forecast percentages

Most CRM systems come with out-of-the box percentages associated with each stage in the pipeline. If our observations are anything to go by, an awful lot of CRM users run with the percentages as delivered. This is an incredibly bad idea, for reasons that ought to be obvious. But hold off changing it just yet, because…

2. Relying on sales activity, rather than buyer behaviour, to define pipeline stages

Problem number two is that most CRM implementations use sales activity (have we qualified? have we proposed?) as the basis for tracking progress through the pipeline, rather than working out where the buyer is in their decision making process. Sales activity is an incredibly inaccurate measure of true progress. So our recommendation #2 is to insist on using evidence of buyer behaviour as the basis for your stage definitions and opportunity tracking.

3. Not knowing how often deals close from the current stage

A simple analysis of the actual average close percentages from each stage in the pipeline can be tremendously illuminating. If your CRM system doesn’t provide this information, then I strongly suggest you invest in a performance management application like the one provided by Cloud 9 Analytics. You simply cannot afford not to have this information.

4. Not knowing how long it takes deals to close from the current stage

Sales pipeline velocity is a missing ingredient from many CRM reports, and a powerful antidote to the wishful thinking of so many sales people who seem to believe that they can make a miracle happen and persuade a deal to close before it’s time. There’s usually a natural cadence to winning deals. How long, on average, does it take to win a deal from each stage of your pipeline?

5. Not being aware of deal velocity

Understanding the average pipeline velocity of winning deals is one important predictor of success, complemented by an awareness of how long each particular deal has languished at the current stage in the pipeline. Sales managers know instinctively that the slower deals move, the longer they take to close – and this is now backed up by a growing body of empirical evidence. It’s hard to forecast accurately without understanding individual deal velocity.

6. Not having evidence that the prospect must do something

What are the chances of your prospect actually buying anything at the end of the day? We’ve observed an increasing proportion of deals ending, not in a competitive loss, but in in a decision to “do nothing”. How confident are you that the prospect is committed to change? And have you helped them to develop a compelling economic case for change? If not, chances are high that they will do nothing.

7. Not having evidence that you are likely to win

Even if the prospect makes a buying decision, what are the chances that you are going to win? After conducting hundreds of win-loss analyses for clients, we’re convinced that many of the factors that influence your chances of success are visible from an early stage in the process, and that comparing individual opportunities against an “ideal prospect profile” that includes demographic, environmental and behavioural factors can help identify which deals you should be qualifying out, and which you should focus your full energies towards

8. A bonus recommendation

I’d like to offer a final bonus recommendation – and that is to base your forecast not on aggregated percentages, but on a classification of each potentially closable deal into one of three categories: Commit (which should close as planned within the current forecast period at least four times out of 5), Upside (where you have at least a one in five chance of closing in the current period) and Longshot (where the sales person has a credible plan, but everything will have to fall right to achieve it).

Once you have your sales people’s assessment, hold them accountable to it, and when any deal deviates from the forecast, invest time in understanding why, and using that learning to refine your forecasting criteria.

All this seems like a lot of work…

Actually, if you get organised, it’s pretty simple to implement, and I’ll guarantee that the approach will help you deliver consistently better forecasts.

And you might want to consider the alternative – what might consistently inaccurate forecasts mean for your career, and the success of your company and the sales teams you lead?

Republished with author's permission from original post.

Bob Apollo
Bob Apollo is the CEO of UK-based Inflexion-Point Strategy Partners, the B2B sales performance improvement specialists. Following a varied corporate career, Bob now works with a rapidly expanding client base of B2B-focused growth-phase technology companies, helping them to implement systematic sales processes that drive predictable revenue growth.


  1. I don’t disagree with you, but I wonder if it’s workable? Namely, do you have a formula that you’ve created that you can use that will numerically value each of these parameters and then calculate a more accurate forecast. Items like the velocity (really cool concept, by the way) could be easily assigned a numeric and calculatable value. But what about the customer’s perception of need, or buyer behavior as opposed to sales process.

    I guess my question is: what’s the practical outcome of this line of thinking?

    Nick Carter
    “The Simple CRM Guy”

  2. Nick

    It’s best applied through a mixture of fact and formula. For recent clients we’ve added (and reported on) custom fields to their opportunity records. The velocity thing depends on time stamping stage changes. Recent changes to salesforce.com make this easier than it used to be, and we’ve had great success using Cloud9 Analytics to come up with powerful trend information.

    Bob Apollo | Inflexion-Point Strategy Partners

  3. Bob,

    So, if you’re using date-stamped status changes (which we track as well), you’re compiling a historic velocity as an average, right? What about the salesperson entering a best guess as to velocity of this sale in a similar way that typical forecasts ask for probability to close? This way, if the salesperson knows it’s going to close in 90 days, it’s not forecasted as an average 120-day cycle.

    Nick Carter
    “The Simple CRM Guy”

  4. Hi Nick

    Take a look at the approach TAS Dealmaker uses. I don’t know what the algorithms look like under the covers, but the process seems to generate extremely usable/trustable results according to users I have spoken to.

    Bob Apollo | Inflexion-Point Strategy Partners


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here