ASSESS· Nov 16, 2026, 12:00:01 PM

Why your forecast doesn't add up, and what to fix first

Discover how to identify and fix gaps in your sales forecasting process, ensuring accurate insights and effective decision-making with Assess.

By Marie Roberts
why your forecast doesn't add up and what to fix first

When the forecast can't be trusted, the cause usually sits upstream of the report. Leads come in and get missed, quotes go out and nobody chases them, and finance works from different numbers to sales. The forecast is the last place those problems show up, which is why rebuilding the dashboard rarely fixes the number. Start by finding where the inputs break, using real records and the people doing the work.

That is what Assess is for: holding the mirror up to your revenue engine, from where leads come from to where quotes go, so you know what to fix before you spend money fixing it.

Why is my sales forecast inaccurate?

Because a forecast is the sum of hundreds of small decisions made earlier by other people. Whether a lead was followed up. Which stage a rep picked. Whether anyone moved the close date when the buyer went quiet. Whether the deal amount still matches what the client is about to sign.

Each of those decisions is minor. Together they are your number. So when the board asks how confident you are in the forecast, the honest answer is often "about as confident as the last person who updated the deal". That is an uncomfortable thing to say to a board, and most revenue leaders know it.

Where does the forecast actually break?

In our experience it tends to break in the same handful of places.

Leads come in and get missed. An inbound enquiry arrives on a Friday afternoon and routes to a rep who is on leave. Nobody sees it until the following week, by which time the buyer has booked calls with two competitors. That deal never enters the pipeline, so the forecast never knew it existed. Missing pipeline is the hardest gap to see, because there is nothing on the report to question.

Quotes go out and nobody chases them. A deal moves to quote sent and stays there. Each month the close date slides forward by a few weeks. On the forecast it looks live. In reality the buyer stopped replying some time ago, and nobody is working the deal because everyone assumes someone else is.

Sales and finance work from different numbers. Sales forecasts from the deal amount. Finance forecasts from signed order value, after discounts and phased billing. Neither is wrong, but they measure different things and nobody has agreed which one goes in front of the board. So every forecast meeting starts with a reconciliation.

Growth sits with one or two people. Two senior reps carry a large share of the number, and much of what they know about their deals lives in their heads and their inboxes. When one of them is on holiday the forecast call stalls. If one of them leaves, a good part of the pipeline goes too, and the board will ask why that risk wasn't visible sooner.

None of these is a reporting fault. Every one of them will still be there after the dashboard is rebuilt.

Why doesn't fixing the reports first work?

A new dashboard redraws the same inputs. Add weighting, add a forecasting tool, add AI on top, and you get a faster and more confident version of the same unreliable answer.

The other common response is the pre-board clean-up. Someone in operations spends a fortnight correcting close dates and stages so the pack looks right. A month later it has drifted back, because the data is a record of how work happens. Change the record without changing the work and the record goes back to describing the work.

This is the case for fixing foundations before reporting. Get the inputs right and an ordinary report becomes trustworthy. Leave them as they are and no report will be.

How do you find the real gaps?

Start with the records, then talk to the people who created them.

The records show you where to look: deals whose close dates have moved more than once, deals in quote stages with no recent activity, leads with no owner, won deals whose amount doesn't match what was invoiced. That gives you patterns, but not causes.

The causes come from the people doing the work. Ask a rep to walk you through the last deal they closed, step by step. Ask finance where their number comes from. Ask delivery when they first heard about their newest client. Wherever someone's account differs from what the record shows, you have found a gap worth understanding.

Write each finding down with the evidence behind it, what it costs you, and who it affects. Mark it as confirmed or as a hypothesis still to test. That discipline is what turns a list of complaints into something you can prioritise and defend.

Deciding what to fix first

Sort the findings into two groups.

The first is what you can fix with what you already have. Agreeing who owns quote follow-up. One definition of close date that every rep uses. A rule about which number goes to the board. These cost little and often move the forecast more than any new tool would.

The second is where further investment is justified: lead routing that handles absence, a proper connection between the CRM and the finance system, a pipeline redesigned around how your team sells. Each of these should come with the evidence that justifies it and an honest view of the benefit, quantified only where the evidence allows.

Then put both groups in order, by how much each change affects the number. What the board needs to see is a plan they can follow and a forecast you can defend, with a clear account of what is being fixed and when.

Where Assess fits

Assess is Centralise's way of doing this work in a fixed three weeks, for a flat £4,000 excluding VAT, whatever the size of your business. It produces four outputs: the Brief, which agrees the scope and the outcomes that matter; the Landscape, showing how work moves today, reviewed by the people who do it; the Report, where every finding states its evidence, consequence, affected owner and priority; and the Roadmap, which sets out what to fix first and where investment is justified.

Assess ends at the Roadmap. You keep every output, whether or not you work with us afterwards. If you're weighing up whether to assess before you build anything, [LINK: Article 1] covers that decision in more detail.

FAQs

How do I know if it's a process or a data problem?

Ask whether the data would be wrong again a month after you cleaned it. If it would, the cause is in how the work happens, and cleaning the data will only buy you a few weeks.

Do we need a new CRM?

Rarely as a first step. Moving a broken process into a new CRM moves the problem with it, at considerable cost. Decide on the system once you know where the process breaks.

How long before we see a difference?

Some changes, such as agreeing who owns quote follow-up, can be made as soon as the Roadmap is agreed. The forecast reflects them as new deals move through, so how quickly you see it depends on the length of your sales cycle.

What does the board need to see?

Where the number comes from, which inputs can be trusted today, and what is being fixed in what order. A board will accept a forecast with known limits far more readily than one that keeps surprising them.

Start with the audit

The free revenue audit asks how your revenue engine works today and shows you where it is under strain. If you go on to Assess, your answers become the starting point for the Brief.

Start the free revenue audit or find out more about Assess.