Stage-Based Sales Forecasting – It Doesn’t Work

Send to Kindle
Written by on . Posted in Forecasting, Sales Methodology 2 Comments

Stage Based Sales Forecasting

Forecasting by sales cycle stage has become an epidemic of sorts – almost every CRM software company out there uses this method. So what is it, and why is it so effective that they use it?

Simply put, this type of sales forecasting creates a series of linear stages for the sales cycle and assigns a probability to each stage, increasing in value as the sales person moves through them. I’ve seen this referred to as ‘sales process’ in some cases.

Lets take an example of a series of sales stages. This is from a real process used by a real company to forecast their sales.

Stage 1 – Initial Contact. 10% probability.
Stage 2 – Initial Meeting. 20% probability.
Stage 3 – Needs Assessment. 25% probability.
Stage 4 – Budgetary Proposal. 35% probability.
Stage 5 – Presentation. 50% probability.
Stage 6 – Final Proposal. 75% probability.
Stage 7 – Gaining Commitment. 90% probability.
Stage 8 – Purchase Order. 100% probability.

At first glance, this appears correct. Sales people can and should follow a process as they move through the sales cycle. However, the idea of assigning a probability to a sales stage is totally wrong. Let’s review two example opportunities:

Opportunity 1:

You have been working with a potential customer for a few months – they definitely have the budget, but have been flipping back and forth between you and your top competitor. You now arrive at Stage 5 – “Presentation.” During the presentation you demonstrate your product, but to your dismay there are a number of technical challenges and the demonstration doesn’t get off the ground. The customer tells you that you have confirmed his concerns with your reliability, and he is now strongly leaning towards your competitor.

Your process says that at this point you have a 50% probability of winning the sale.

Opportunity 2:

An existing customer calls you up to say that he is really happy with your products that he had previously purchased. In this budget season, he will likely require 3 more, but he isn’t yet sure of his exact requirements or schedule. He specifically says that he will likely not entertain any competitors as they are very happy with your company, and the difficulty in maintaining separate products on-site is too much hassle. You agree to set a meeting to review detailed requirements in a month.

Your process says that at this point you have a 10% probability of winning the sale.

Clearly in the two opportunities above, the probabilities should be reversed! This is the key point: Where you are in the sales cycle is not the sole criteria for assigning probability to a sales opportunity. Many other factors need to be considered – need, competition, relationships, value, and more.

The unfortunate impact of stage-based probability assignment is that sales people manipulate the data. They pick the stage to get the probability they want. The value of one of the two linked fields can’t be trusted – is it “Stage 1 – prospecting” or is it “10%” probability?

Time is still critically important – the more time you have left, the better the chance you have to affect probability in your favour. Further, as you move through the sales cycle, probability should become more accurate. All things being equal – you just know more.

Join in on the discussion on our SalesWays Professional Network.
  • Faisal Haneef

    Chris – I agree with your point here, the linear “Sales Process” doesn’t account for many variables. Thanks for making it so simple to understand.

    On a further note, any alternative process that is more mature?

    • Chris Hamoen

      Hi Faisal – thanks for the comment. We argue that the process should be layered over a model such as ours – the Probe/Prove/Close universal sales model, and probability should be assessed independently as the opportunity moves along. In our case, we ask two questions – Will It Happen, and, if so, Will We Win?

      FYI I posted a follow-up – I like the new ‘high velocity’ selling concepts, but they revert to classic thinking on probability again:

Social Network