What Challenges Are You Facing in Your Forecast Accuracy?

I have to laugh out loud. We take the simple... "is the deal in or out" and make it complex. Tie forecasting accuracy to compensation, give them metrics by which to measure success as in +/- 10% and pay them for achieving the goal. Once you have cemented the behavior in your culture you can move on.

Maybe the reason only 50% of all sales reps make quota is because we don't dumb it down enough. Here is one KISS strategy that is easy to implement and will have immediate results. The downside is what??

My vote is to get sales management back to basics... you want something to happen then outline what your expectations are and reward the right behavior. Net/net.. do you want to analyze the crap out of it or do you want to fix it?

PS - Everyone has their own viewpoint and I respect all presented and hope that in turn the same is true.

Accurate forecasting is the lifeblood of any business. How do you grow your business if you don't know your revenue to expense ratios?

If you have a forecasting accuracy problem, consider addressing it through your compensation plan. Compensation drives behavior right? Take 20% of your rep's monthly compensation and tie it to a monthly bonus based on achievement of +/- 10% of forecast. Shine a light on the behavior you desire and measure it.. the results will make you happy.

This thread appears to be mashing up 2 issues.

How much effort you put into a forecasting system stems from how much business risk is there in the variance (inventories, time commitments, sales expenditures.) How much effort you put into your sales management system stems from how important is it to have productive sales activies (teachable, measureable, monitorable.)

If the risks and costs are high enough... applying a modicum of scientific management methods is called for in either case.

I am not a VP of Sales, but can offer perspective on what I see and have seen:

1. User adoption of the CRM system -- In a lot of organizations, sales management is attempting to use the CRM as the record of truth -- user adoption kills this. Many sales organizations simply can't get their team to use the CRM application. In some organizations, sales reps aren't using the CRM application at all and in others, they aren't updating it consistently.

2. Sandbagging -- If you aren't managing sales via the CRM, you won't know where they really are in many accounts. This gives sales people the opportunity to manage expectations downward.

3. Bad sales management "BS" detectors -- When I was a consultant, I learned something about sales meetings -- really good VP of Sales know how to figure out what's real or not. Bad ones don't.

I tend to agree with Christian, making forecasting a part of the compensation plan presents some challenges. It starts to dilute the focus of the compensation plan. Soon we might be putting all the expected behaviors into compensation plan, confusing the sales person on what is really expected.

Accurate forecasting should be a basic performance expectation of sales people. Managers need to coach sales people on their responsibilities for forecasting, they need to train people on how to forecast properly, and work with them in developing their skills.

If the person consistently forecasts poorly, it's a performance issue, not a compensation issue. Managers should manage it as a performance problem, not a compensation problem.

(Note, I'm setting aside all the organizational and systemic issues on forecasting--many of which I addressed in my earlier comment, and which have been addressed by others.)

I agree to a certain extent with what Trish says about KISS. Having said that, it is up to the sales management to instill some discipline into the sales force by making sure that when a sales person has qualified an opportunity and is forecasting it, that they can support their forecast with fact-based evidence. Simple things like access to power, acknowledged pain, established customer value, agreed a buying vision and having an element of control within the evaluation process.
So often, failure to manage even these basic things is the reason that you see spikes in the pipeline at 20% and 80% through the sales cycle.
Fact based evidence of moving from one stage to another, verified and real close-plans for qualified opportunities and sales managers/directors that are firm but fair when they analyse the sales people about their forecasts. It's not rocket science,but it is a skill and an art.

Trish, great suggestion. Jim Dickie made a similar suggestion in a webinar recently, and I totally agree.

Forecast accuracy is a thorny issue and there is no silver bullet to fix it. Trish's suggestion of addressing forecast accuracy issues theough the compensation plan seems simple and attractive at first sight. However I have experienced that already getting an acceptable definition of forecast accuracy can be a real challenge.

The compensation motive (carrot & stick) may temper forecasts but I believe as a long term process it is tenuous.

The compensation scheme described appears to drive realism into into the psyche of the salesperson through an extrinsic and extraneous factor - pay-per-guess. The opperative word here is "guess", which is what all forecasts are. This additional economic burden may (as most compensation schemes do) distort the focus and efforts of salespeople and may have unintended consequences i.e. are you prepared to "punish" a salesperson for an unexpected sales? (If you are interested in psychology lookup "intrinsic motivation".)

The basis of forecasting is modeling. Sales facilitated by representatives can have a wide range of characteristics, many of which can be measured. Good sales management strives to understand and measure those parameters (conditions and activities) that affect outcomes. There are a number of such schemes, for example: the funnel and the pipeline, that provide a framework for recognizing the time-phased nature of these parameters as preditors and guides to action.

These types of "sales process" models allow the valuation of the current situation and facilitate adaptive activities i.e. pushing, sweatening, withdrawing, etc., not to mention providing coaching, counselling and learning opportunities for "the team"... and better forecasting to boot!

I believe that understanding and engaging the sales process in this manner is more robust than tweaking compensation and will provide greater long term returns. Also consider: "sales" modelled and managed as a system of affectible processes can eliminate the need for field based forecasts altogether.

Any forecast requires inspection and discussion. All too often, we look at the numbers and take a rule of thumb from a sales leadership perspective on what we believe will happen based on historical performance of a person or an account. Important to be looking around at what is happening when opportunities become forecasted deals. There should be an amount of "chatter" in the company. Not just from the sales person but with finance, approval from legal on terms, discussions on customer support - all of this applies especially with larger enterprise sales.

Still, one of the most important aspects of accuracy to forecast is the discussion with the sales representative and sales manager about the forecasts. Talk about the company you are selling to and listen to the inputs the rep has to why they are buying your solution. Over the years, I've found that it's not selling to one person that will get the deal over the line, so inspect the contacts with the customer to ensure that it's being appropriately spoken about (chatter) in their company.

What I have found to be two major problems with forecasting, especially in B2B cloud/SaaS sales organizations, are two things:

1. Pressure to fill meeting time for pipeline updates.
2. Inadequate use of 'staging' within the sales process of a CRM implementation.

1. Reps feel pressured to show up to 'pipeline update meetings' with managements having enough to talk about to fill the meeting time. If there's a 30 minute meeting scheduled, they'll need 5-6 deals to talk about, regardless of whether or not these are actually winnable sales opportunities. I think Sales Managers are best served having daily informal drop-by discussions to see what's going on today rather than relying on weekly, monthly or quarterly meetings. This will give you far more realistic insight on a rep-by-rep basis and provide opportunities to help close deals on the brink when necessary.

2. As Craig and Dave noted above, actually using CRM is a critical first step. The next most important step is standardizing the stages within a sales process. I think having a 5-7 step process is about right. Something like (no interest, confirmed interest, has proposal, negotiating pricing, trial account, verbal commitment, closed). Once these are setup, enforce what they mean rigidly, then attach a close % to each. Pipelines will start to forecast themselves.