This is exactly what Jacob Gonik set out to achieve for IBM Brazil in the 1970s.
The starting point is that if we want an effective sales plan, there needs to be some notion of expectation in the form of a goal or quota. The challenge, as most of us know, is knowing what that quota should be.
Gonik theorized that those best-equipped to set a reasonable quota, are the sales people themselves, so by allowing them to set their own quotas, IBM Brazil would end up with realistic goals.
I know what you are thinking – why on earth would any sales person set a high quota, no matter how realistic, when they could set a lower quota, smash through it quickly, earn their annual commission by March, and take the rest of the year off?
To address this, Gonik developed the OFA (Objective, Forecast, Actual) approach. So instead of the usual two variables – Objective (goal/quota) and Actual (actual sales) – we have the third variable, Forecast, which is set by the sales person.
The process goes something like this:
- The salesperson receives an Objective, ‘O’
- The salesperson reviews that Objective, and sets their own Forecast ‘F’ based on what they believe they can achieve for their territory
- F divided by O gives a ratio. For example, if the Objective is 100,000 and they believe they can achieve 150,000, the ratio (F/O) is 1.5
- At the end of the performance period we review Actual performance ‘A’ versus Objective ‘O’
- The final payout is a function of A/O and F/O as determined by a carefully calibrated matrix
I imagine you lost me at step 5 so let’s take a look a the matrix used for IBM:
So you can see above that the actual pay-out is derived both from a salesperson’s performance vs target and their own ambition with respect to their forecast. The higher their forecast relative to target, the higher the earnings potential.
I’ll go through a worked example with two scenarios:
Sally Sales-McSalesface receives an Objective of $1m. She wants to buy a Ferrari this year so reviews her pipeline, prior year sales, and opportunity and decides that she can achieve more than that – she thinks she can get to $1.5m.
So scenario 1 – let’s say that she simply returns a Forecast of $1m, the same as her Objective. Across the top of the matrix (F/O) she fits into the column labelled ‘1.0’ ($1m/$1m).
At the year end. Success! She achieves the $1.5m she thought she could. Actual results of $1.5 divided by her Objective of $1m multipled by 100 gives 150 (A/O*100).
So we start at 130 on the left, read across to 1.0 across the top, and that gives a score of 150.
Now scenario 2 – let’s say that Sally backs herself and submits a Forecast of $1.5m meaning she fits into the column across the top (F/O) labelled 1.5.
This time, when she finished the year at 150% of her Objective she gets a score of 180 – more than under scenario 1 for the same level of actual performance thus rewarding her for being both ambitious and successful.
I’m not going to pretend this isn’t more complicated than either a traditional quota-based plan or a commission scheme – it is. However with the right communication, it is clear that this achieves the stated objectives of reinforcing quota-based achievement without having to go through a detailed quota-setting process.
I know what you’re thinking – We still need to provide Objectives for this to work and of course that is true however they can be set relatively simply and consistently by sales person rather than having to allocate unevenly across territories which is often the time-consuming part of quota-setting.
I still know what you are thinking – does it actually work? Well, interestingly… I have absolutely no idea. My trawling of the internet has found no answers to this question and as far as I can tell this is the only documented use of it. As a consultant I would relish the opportunity to pilot this model and see if it works as well in practice as well as it would seem to in theory.
“Sell crazy someplace else. We’re all stocked up here” – Melvin Udall
Leave a Reply