Skip to content

How Do We Improve Forecast Accuracy?

by David Brock on May 8th, 2019

Yesterday, I wrote, Forecast Accuracy, Again. It’s an important foundation to this article. It focuses on why we need forecast accuracy, which may be different from what most sales leaders think.

I’ve actually become very accurate forecasting when and how this topic comes up. There are two peak periods.

The first is about now, midway through the second quarter. This happens because people may have struggled to make their numbers in the first quarter, they are midway through the second quarter, looking forward, seeing both the quarter and the year at risk. Jumping on the topic in the middle of the second quarter, hopefully, helps them get the right things going for the quarter and the year.

The second time this topic is hot is in the last 30 days of the third quarter. These queries come from people and organizations in deep trouble, trying to figure out some sort of recovery strategy–most often, too little too late.

It almost never comes up in the first quarter, people are focused on kick-offs, optimistic about a fresh start. Also, it never comes in the fourth quarter, every one is powering through doing whatever they can to make the numbers.

Often, the query, “How do we improve forecast accuracy,” is really masking the issue, “We’re struggling to make our numbers, how do we fix it?” I’ll focus on improving forecast accuracy, which will also address, how we make our numbers.

When I talk to people about the issue, they inevitably start the discussion in the wrong place–the forecasting process. Somehow they think that’s the problem.

They go into great detail describing the forecasting process and the various levels of “commitment to the forecast.” They use different language, “Firm commitment, commit, possible…..”

One client, the highest level commitment is called the “blood commit.” I, foolishly asked the question, “What’s a pinky finger commit?” You probably can guess their patience with my observation.

This whole concept is intriguing, in itself. We want to have very accurate forecasts, yet we set up a forecasting process that has various levels of certainty about the forecast. Implicitly, this process sets up uncertainty and inaccuracy in the forecasting process.

It seems to me, there is a commitment or not. There can’t be degrees of commitment because those degrees of commitment usually have nothing to do with the customer’s readiness/urgency to buy, but with the sales person’s opinion. In this case, it seems we would be more accurate by running trend analysis and analytics because this process has more to do with the number and not with the deal (or collection of deals).

Lesson 1: We don’t fix the forecasting process by fixing the forecast meeting/review process.

If we don’t improve forecast accuracy in the forecasting process, how do we improve forecast accuracy?

The next place to look is the pipeline. If our pipeline is filled with crap, if our pipeline is inaccurate, we can never have accurate forecasts! There’s the old saying, “Garbage in, garbage out.”

We have to have high quality and high integrity pipelines to begin to have higher degrees of accuracy in the forecast.

We have to inspect our pipelines, are the deals real or wishful thinking? Are they high quality? Are they accurately positioned in the pipeline? Do we have an accurate view of the target close date, the value of the deal, what the customer will buy?

Quickly, we realize these questions are less about the pipeline, but more about the deals themselves. The pipeline is just a representation of all the deals we are working on and where they are in the buying process.

Lesson 2: So isn’t forecast accuracy is really more about a deal and the quality of the deal/customer engagement, than the pipeline and forecasting process?

But then we have to start asking some questions about the deal? Too often, we ask the wrong questions, focusing n what we are doing and what the competition is doing, and how we are positioning ourselves to beat the composition. The customer ends up becoming the poor victim upon who we inflict these activities.

But the most important questions about the deal end up being about the customer. What problem are they trying to solve? What is the impact of this problem? What happens if they don’t solve the problem? What is the urgency the customer has about solving the problem? When do they need to have a solution in place? What happens if they miss that target date? Do they agree with this and are they committed to achieve the date?

Unfortunately, too often we don’t ask ourselves these questions, we don’t ask the customer these questions, the customer may not even ask themselves these questions. But until the customer can answer these questions, in a quality manner, we have no idea about “when” a certain deal will happen, or even the likelihood that anything will happen.

Once we and the customer have answers to these questions. We have a start to forecast accuracy.

But there are a lot of other questions that impact the accuracy of forecasts. Unfortunately, they are about the customer again. We know customers struggle to buy. 53% of their buying journey’s end in no decision made–even though they have a need to buy.

As a result, the next series of questions that contribute to forecast accuracy has to do with the customer’s ability to navigate their buying process and make a decision–for anyone. We can make an assessment, perhaps flip a coin (at 53% No Decision Made, I’d tend to bet tails).

Do they know how to buy? Do they know how to align interests and agendas? Do they have a project plan and are managing to that? If they have a compelling urgency and a deadline they have to make something happen, we have much more certainty.

This struggle to buy has a huge impact on forecast accuracy.

Lesson 3: Forecast accuracy is more about what the customer is/isn’t doing, not what we are doing. Yet we make forecast discussions all about what we are doing and our confidence level with those activities.

But we can have a huge impact in this process–helping the customer discover how to buy. If the customer wants to buy, but fails in their buying process, they don’t buy. If we want to sell, we only do that when the customer buys. Consequently, we achieve our goals when we help the customer navigate through their buying process.

Lesson 4: A key issue in forecast accuracy is, what are we doing to help the customer buy, how are the incorporating our leadership in their buying process, is this creating differentiated value for them?

Finally we get to the issue of vendor/product selection. Yet this is where we focus our forecast discussions—“What makes you think they will choose our product over the competitors’?” The reality, is any supplier on the customer’s shortlist will solve their problem. So it probably is not the key to success in winning and in forecast accuracy.

Magically, it’s comes back to the customer–but how we are helping the customer through their buying journey.

Lesson 5: Forecast accuracy has little to do with what we sell and how we stack up against the competition. It is always all about the customer.

Bottom Line: We make forecasting and forecast accuracy about things that have very little to do with what cause a customer to buy and what causes the customer to select us. It’s no wonder that forecast accuracy is terrible. Regardless of our “blood commits,” or whatever commitment we make, we are assessing the wrong issues, so we might as well flip a coin or do some sort of trend analysis. We’d probably be more accurate and waste less of sales people time.

Second Bottom Line: We don’t need to do this stuff just to get accurate forecasts. In fact, these things are critical to building differentiated value with customers and and executing value based deal strategies.

If we are doing these things, qualifying opportunities where the customer has a compelling need to change, helping the customer understand the importance of making the change, helping the customer navigate the buying process—we are doing the most impactful things, the things most important to the customer, the things that create value for and with the customer. If we do these things we maximize our differentiation and value and maximize our ability to win deals.

So doing these things is critical to high performance selling–and, by the way, improves forecast accuracy. By doing what we should be doing in each and every deal, we will be improving our performance and our accuracy.

Final conclusion: We need to do these things to hit our numbers. We need to do these things to improve forecast accuracy.

However, it will never be perfect. Sometimes, shit happens.

Book CoverFor a free peek at Sales Manager Survival Guide, click the picture or link.  You’ll get the Table of Contents, Foreword, and 2 free Chapters.  Free Sample
Be Sociable, Share!
Please follow and like us:
  1. Brian MacIver permalink

    “I’ve actually become very accurate forecasting when and how this topic comes up. There are two peak periods.” Loved it!

  2. Joel Lyles permalink

    | But the most important questions about the deal end up being about the
    | customer. What problem are they trying to solve? What is the impact of
    | this problem? What happens if they don’t solve the problem? What is
    | the urgency the customer has about solving the problem? When do
    | they need to have a solution in place? What happens if they miss that
    | target date? Do they agree with this and are they committed to
    | achieve the date?

    Dave, that is a fascinating paragraph in your second bullet point. I’ve sat on plenty of forecasting reviews and those questions have never affected the forecast except in a vague ‘how does this make the Trial Evaluation phase more appealing’ sense.

    A lot of forecast accuracy only focuses on where we are in the process, as you address in Lesson 3. I think Lesson 2 and Lesson 3 contribute 90% of the problem of bad forecasts in otherwise clean pipes. I remember being absolutely stunned by deals not closing or getting pushed off because even though we did everything ‘right’ on our end — you know, getting the technical sign off, a successful demo that wowed the engineers, competitive quotes where we beat the pants off of the alternatives — simply because my team didn’t even think about basic questions to the customer.

    In one case, we would’ve realized that even if this deal was going to close, there just wasn’t any urgency because they couldn’t even install our new machines in their shop until they used the previous class of machines to finish a make-or-break contract. It ended up sitting in my pipe for six months while everyone ‘we got everything we wanted, including the CMO straight-up saying he wants our machines to replace the old ones, where’s the PO’?

    • Joel, the problem with most forecasting is it focuses on us, where we are in the process and what we are/are not doing. None of this has anything to do with the customer propensity to buy. And that’s the only thing that is important.

      • Joel Lyles permalink

        | None of this has anything to do with the customer propensity to buy.
        | And that’s the only thing that is important.

        It seems so obvious when you spell it out that way, but it occurred to me that no sales organization I’ve worked and/or interviewed with manages deals like that. For example, it makes sense for a deal to be moved ‘upwards’ in the pipe/funnel if something major changes on the customer’s end that doesn’t kill the deal (for example, if the company replaces a key stakeholder who needs to be onboarded with the deal) but every organization I’m with would consider it outrageous if I moved from, say, stage 5 to stage 3 and thus decreased the probability of the deal closing.

  3. Jose Angel permalink

    Hi David, happy new year.

    I have been reading your articles since a while and all I have read have been very useful and productive for me (even your Guide). Let’s say my experience is not that big as many of the folks around here, but, I have a question. in my case I work with representatives, where many if not all are not Too professional within their sales process, it has been very difficult battle to make them change their mindset about it, why? Here some feedbacks

    1. they have been doing it like that for years and it has worked,
    2. why do we have to change?
    3. Following your process takes time out of my working process!
    4. This opportunity will get close next month, my friend told me it will happen;
    5. ooohh I have a surprise for you, i will get a new order next week! Which one? You don’t know it I started talking last week now it was approved and signed ! (very common and biggest headache for me)
    6. Hey!!! I got an order!!! Really, that’s good, is the opp 3? Oh no, it is the opp 10 that we were working last year. Oh great, but how it happen that yesterday and in our last 3 phone calls you did not comment it? Mmm well, my sales guy did not tell me/it came at last minute! (second headache)
    7. Hey tender closed and we won! The order will come in 3 weeks, 5 weeks later, no order due to missing signature.

    And many other statements that I deal on a daily base.

    What I do normally do is, I try using the statistics to help me understand and develop an accurate FC, trying to see
    1. the conversation ratio of the rep,
    2. the avg. deal size,
    3. avg opp age,
    4. using my acknowledge about the rep , market and country.

    i do normally use several if not all those questions From lesson 2 for the opp follow up process, anyhow it is still not accurate as I would like it to be. So at the end my FC tends to be more conservative and in many many cases Number related. my FC accuracy has been improved from less than 50% up to 75/85% still a lot to do!

    Any feedback that could be helpful?

    • Jose: Thanks for the outstanding observations and feedback. You’ve made great progress in improving your own forecast accuracy, and you should always seek to be as accurate as possible. It will never be perfect, because there are so many things that impact it. I’d also remember forecast accuracy is just one indicator of performance. As you outline in your comments, there are a lot of other indicators that are meaningful in assessing your performance and improving it. The trick is to identify and balance the few most impactful and to continue to look at improving performance. Looks like you are making a great start. Regards, Dave

Trackbacks & Pingbacks

  1. "Customer Propensity To Buy" | Partners in EXCELLENCE Blog -- Making A Difference
  2. “Customer Propensity To Buy” | CustomerThink

Leave a Reply

Note: XHTML is allowed. Your email address will never be published.

Subscribe to this comment feed via RSS