What Is an Inspectable Sales Forecast? The Definition and What It Must Show
Your CFO is building the board deck and wants two things by Friday: the deals inside the $14M you called commit, and what each of them rests on.
The request that lands the week before the board meeting
Picture the email. The CFO has your commit number and wants the deals behind it, with a line on why each one will close this quarter. Listing the deals takes ten minutes. The second half of the request takes the rest of the week.
RevOps can pull the CRM fields in an hour. The deals say Commit, the close dates sit inside the quarter, and most of the MEDDPICC boxes are filled. What the fields can't say is whether anyone on the buyer side ever confirmed what's in them.
The reasons live in places a spreadsheet can't reach. One rep heard the economic buyer approve budget on a call in August. Another deal rests on a procurement email sitting in someone's inbox. A third is in commit because the rep has a good feeling about the champion, and nobody wrote down why.
None of those reasons are wrong. They just can't be checked by anyone who wasn't there.
The definition
An inspectable sales forecast is a bottom-up forecast where every number traces back to the evidence behind it. Pick any figure, from the quarter total down to one deal's close date, and you can follow it to the buyer conversation, email or recorded decision that supports it.
It is built from deals
The total is the sum of deal-level judgments. A top-down number set from last year's growth rate can be reasonable, but there is nothing inside it to inspect.
Each input carries its source
Every qualification answer records where it came from: the buyer said it on a call, the rep logged it, or the system calculated it from other signals.
The same evidence gives the same answer
If you ask about the same deal twice with nothing new in it, you get the same score. A forecast that shifts when nothing in the pipeline shifted can't be inspected, because the movement has no cause to find.
The gaps show
A deal with no confirmed economic buyer says so plainly. Missing evidence is information, and a forecast that hides it behind a confident percentage fails the test.
📊 By 2026, 65% of B2B sales organizations will transition from intuition-based to data-driven decision making. — Gartner, Future of Sales |
What it must show for each deal in the number
Use this as the minimum. If a deal in commit can't produce something in the middle column for each row, the forecast is carrying that deal on trust.
Forecast input | Evidence you should be able to open | Warning sign |
|---|---|---|
Close date | The buyer naming the date or the event that drives it | Date set by the rep, pushed more than once |
Amount | Scope or pricing discussed with the buyer | Amount changed with no buyer activity |
Economic buyer | Budget authority confirmed in a meeting or email | Named in the CRM, never in a meeting |
Champion | Someone on the buyer side actively selling internally | Friendly contact with no access to the decision |
Decision and paper process | Steps, approvers and timeline described by the buyer | Process assumed from a similar deal |
Recent activity | A buyer conversation or reply since the last review | Nothing from the buyer in three weeks |
None of these rows is a probability. The percentage on a deal is an output of the evidence in the middle column, so a forecast that shows the percentage alone has skipped the part a reviewer needs to see.
Explainable and inspectable are two different tests
Many AI forecasting tools now write a short explanation next to each deal. That helps a reader skim, and it is a reasonable feature. The problem is that the explanation is also generated, usually from call summaries, and a reader has no way to check it short of replaying the calls.
So ask a narrower question of any explanation. Can you open the thing it points to? "Budget risk: the buyer mentioned a freeze" is an explanation. A link to the moment in the March 4 call where the CFO said it is inspection.
📊 Half of RevOps teams can't name an owner when an AI-generated number turns out to be wrong. — Vasco, 2026 State of AI in RevOps |
An owner is easy to name when every figure opens to its source. The person who reviews the forecast can click into a number, see the buyer's words behind it, and decide whether they agree.
Generated explanations also have a consistency problem. Ask the same tool about the same deal on two different days and the paragraph can read differently, sometimes naming different risks. Someone comparing this week's explanation with last week's can't tell whether the deal changed or the wording did.
Rep judgment belongs in the forecast, labeled
An inspectable forecast doesn't throw out what reps know. A rep who had dinner with the champion and heard the board approved the project has real information that no recording captured. That call deserves a place in the number.
It just needs a label. When the forecast shows which inputs the buyer confirmed and which ones the rep logged, a reviewer can weigh both. Reviews get shorter, and the rep can point to exactly what backs the deal.
Labels change how a forecast call runs, too. A review that opens with "the buyer has confirmed four of six inputs, and these two rest on the rep's read" goes straight to the two open items. Without labels, the first twenty minutes go to establishing what anyone actually knows.
How Spotlight.ai builds an inspectable forecast
Spotlight.ai is the Enterprise Revenue World-Model. It runs on Spotlight's Calibrated-Decision AI-Model (CDAM), which reads the buyer conversations and email threads on every deal and answers each qualification question inside a closed answer space defined before the model is called. Each answer is stored as a typed value with its source and a confidence score, and it only counts once it clears the bar.
The deal conclusion is then calculated in code from those stored answers. The Spotlight Score adjusts each deal's stage probability on MEDDPICC status, stakeholder engagement, slippage and time on stage, and the bottom-up forecast rolls up from there. Because the score is computed from stored values, the same evidence gives the same answer, so a change in the number points to a change in a deal.
The same evidence works across the revenue team. Sales gets every deal qualified, inspected and forecast on what the buyer said. RevOps gets complete deal data and the winning, losing and slippage patterns behind the number. Value teams get a business case on every deal built from the buyer's own words. Customer success runs renewals with the same rigor, so retention can be forecast like new business.
Gary Olson, CRO at Sysdig, described the gap it filled: "What I didn't have was an ability to integrate and access all the different data points. Spotlight brings it all together and spotlights what needs to happen next." Sysdig reported 3.8x pipeline conversion over 12 months.
Test your current forecast
Pick one deal at random from commit. Try to reach the buyer evidence for its close date in under two minutes. If you have to message the rep, the forecast isn't inspectable on that deal.
Check for source labels. Look at whether the CRM distinguishes a buyer-confirmed economic buyer from one the rep entered.
Ask the same question twice. If you use an AI forecast, re-run a deal that hasn't changed and compare the scores.
Count the visible gaps. A healthy commit list shows its missing evidence. Zero gaps across twenty deals usually means the gaps are hidden.
Read one AI explanation to its source. Follow the claim back to the call or email. Note how long it takes and whether the source says what the explanation says.
Pick five deals from this quarter's commit
Choose the five largest deals in commit and fill in the table above for each one, with a link or timestamp in every cell. Most teams find two or three cells per deal they can't fill, and those cells are the conversation for next week's forecast call.
The fastest test is one deal you already lost. Pull its last forecast entry and check whether every number on it traces to something the buyer said or did. The gaps you find are where an inspectable forecast starts.
FAQs About Inspectable Sales Forecasts
What is an inspectable sales forecast?
An inspectable sales forecast is a bottom-up forecast where every number traces back to the evidence behind it, such as a buyer conversation, an email or a recorded decision. Anyone reviewing it can open the source of a deal's close date, amount or qualification status without asking the rep.
What is the difference between an explainable and an inspectable forecast?
An explainable forecast gives a written reason for each number, which is often generated from call summaries. An inspectable forecast lets you open the underlying evidence, so you can check whether the reason holds.
What should an inspectable forecast show for each deal?
At minimum, the evidence behind the close date, amount, economic buyer, champion, decision process and most recent buyer activity. It should also show which inputs the buyer confirmed and which the rep logged.
Does an inspectable forecast replace rep judgment?
No. Rep judgment stays in the forecast with a label showing it came from the rep, so reviewers can see what the buyer confirmed and what the rep assessed.
Can an AI forecast be inspectable?
Yes, if every AI output is tied to the evidence it came from and the same evidence always produces the same answer. A forecast whose scores move when nothing in the deal changed can't be traced to a cause.
How do I test whether my forecast is inspectable?
Pick a random commit deal and try to reach the buyer evidence for its close date in under two minutes. If you need to ask the rep, the forecast isn't inspectable for that deal.
Comments