🔷 HubSpot Sales Practical Textbook — 2026 Edition
Chapter 8

forecast
Improve the accuracy of sales forecasts,Accelerate decision making

“How many orders do you think you can receive this month?”——There are still very few sales organizations that can confidently answer this question. The optimistic bias of the person in charge, the intuition of the manager, and the "actually impossible" business negotiations that pile up at the end of the month. If forecast accuracy is low, all management decisions such as recruitment, marketing budgets, inventory, and development resources will be distorted. In this chapter,HubSpot's forecast function, AI Deal Score, forecast category, quota managementWe will explain how to combine these to build a sales forecasting system that "talks with data rather than intuition."

📖 Estimated reading time: 30 minutes
🎯 Target: Sales managers, managers, RevOps, CFO
📅 March 2026 edition

📋 Contents of this chapter

  1. 8-1Basic concept of forecasting: Why forecasts are off and 3 principles to reduce them
  2. 8-2Designing forecast categories and using the HubSpot Forecast tool
  3. 8-3Leverage AI Deal Score and Predictive Scoring
  4. 8-4How quota setting and percentage tracking works
  5. 8-5Agree on the estimate sending date, upper limit on the number of revisions, and contract review period. Create quotes with HubSpot CPQ and centralize the closing process with electronic signatures (DocuSign integration)
Section 8-1

Basic concept of forecasting: Why forecasts are off and 3 principles to reduce them

A sales forecast is a prediction of how many orders a company can receive in the current fiscal year (this month/this quarter). Highly accurate forecasts speed up management decision-making, while less accurate forecasts distort company-wide resource allocation. The first step to improvement is to understand why the forecast is off.

Three major reasons why forecasts are incorrect

causespecific symptomsHow to deal with it with HubSpot
Optimism bias of the person in charge Business deals that keep saying "I can do it this month" but can't be closed keep piling up. The person in charge is reluctant to admit the loss of an order because they want to believe that there is still a possibility. 1:1 Opened a sales email
1:1 Reply to business email Setting dates based on wishful thinking, such as ``I set it to the end of the month'' or ``Somehow it ended up being March 31st.'' Does not reflect the buyer's decision-making process Track the change history of the scheduled closing date and manage "how many times the closing date was changed" using opportunity properties. Treat deals with many changes as caution flags
Lack of pipeline quality control Opportunities for which qualifications have not been confirmed, decisions for which the decision maker has not been identified, and opportunities that have not moved for more than 6 months are mixed in the pipeline, distorting the overall numbers. Regularly cleanse your pipeline using a "pipeline quality score" that combines MEDDIC property input status, Deal Risk signal, and last activity date.

Three principles to improve forecast accuracy

Principle 01
Juxtaposing subjective and AI scores
Always display the forecast category (subjective) entered by the person in charge and the AI ​​Deal Score (objective) side by side. Negotiations where the person in charge says Commit but the AI ​​score is 32 are the ones that require the most attention. This list of deviations will be the most important agenda item in the forecast review.
Principle 02
Make sure everyone defines forecast categories
If the definition of "What is a commit state?" differs depending on the person in charge, it will be meaningless to tally it up. Document the definition such as "Commit = a state in which verbal agreement has been obtained from the other party's decision maker and you can commit to closing within this month" and create a state in which everyone can input based on the same criteria.
Principle 03
Maintain pipeline coverage
Discount management is one of the most individualized areas in many sales organizations. If it is not clearly stated who can discount the price by what percentage, prices will vary depending on the person in charge's negotiation skills, experience, and relationship with superiors.
Design a CPQ approval workflow to incorporate discount policies into your CRM and ensure everyone is playing by the same rules.
Review the difference between forecast and actual results weekly
Automatically detect intent signals
Section 8-2

Designing forecast categories and using the HubSpot Forecast tool

HubSpot has a dedicated Forecast toolis available, allowing each person in charge, manager, and VP to manage pipeline prospects at different levels of granularity. Available for Sales Hub Professional and above.

Four-level definition of forecast categories

🔵
Pipeline
〜30%
There is a possibility, but there is no certainty that it will be closed this period. Exploration/hearing stage. I'll count it, but I don't have high expectations.
🟠
Best Case
〜60%
If things go well, there is a possibility of closing this term. Detailed consideration is progressing, but blockers remain before decision-making
🟢
Commit
〜90%
Business negotiations that we can confidently promise to close within this period. Verbal agreement has been obtained from the decision maker, and all that remains is the contract and signature. Number for which the person in charge is “responsible”
🔷
Closed
100%
A firm order that has already become a Closed Won. This will be recorded as the achieved amount for the current period. Compared with Commit/Best Case as forecast “performance”

How to view the HubSpot Forecast dashboard

📊 Sales Forecast — 2026 Q1 (January to March)
Last updated: 2026/03/08 09:00
quota
¥30,000,000
Q1 Goal
Closed Won
¥18,400,000
Achievement rate 61.3%
Commit
¥7,200,000
Total ¥25,600,000
Best Case (upside)
¥5,800,000
Assigned to
📈 Pipeline stacking by category
Closed Won
¥18,400,000
61% achieved
+ Commit
+¥7,200,000
85% expected
+ Best Case
+¥5,800,000
105% upside
Pipeline
¥14,400,000
Candidates for next term
🔥 Commit opportunities scheduled to close this month
ABC Co., Ltd.
Sales Hub Pro × 15 seats
¥1,872,000
Commit
XYZ Co., Ltd.
Sales Hub Pro × 8 seats
¥1,008,000
Commit
DEF LLC
Sales Hub Ent × 20 sheets
¥3,600,000
Best Case
⚙️ CPQ approval flow design example (automatic routing according to discount rate)
47
GHI Co., Ltd. (¥2,400,000)
Person in charge: Commit ← AI: 47 points
Decision maker not involved/Deal Risk 2 cases
31
JKL Co., Ltd. (¥1,800,000)
Person in charge: Best Case ← AI: 31 points
No response to email for 14 days/competitor comparison
88
MNO Co., Ltd. (¥3,200,000)
Person in charge: Pipeline ← AI: 88 points
AI is highly evaluated → Possibility of underestimation by the person in charge
⚡ Managers can override assignee accuracy with “Submission Forecast”

HubSpot's Forecast tool allows you to enter forecast categories thatManager overwrites “adjusted value based on own judgment”can. If the person in charge says ``Commit'' but the manager determines that ``Best Case'' is appropriate, the outlook in the ``Manager View'' can be revised without affecting the numbers on the portal. By accumulating this overwriting history, it becomes possible to quantitatively evaluate ``whose forecast is actually accurate?''

Section 8-3

Leverage AI Deal Score and Predictive Scoring

AI Deal Score is a function in which HubSpot performs machine learning based on past order/loss data and scores the probability of winning an order for each deal currently in progress from 0 to 100 (Sales Hub Professional and above). Accuracy will increase if you have data on more than 200 business negotiations over the past two years. We provide data-based business negotiation evaluation that does not rely on the subjectivity of the person in charge.

Key factors referenced by AI Deal Score

categoryreferenced elementImpact on score
Opportunity attributes Amount, industry, company size, source of negotiation, stage accuracy Opportunities with attributes close to ICP tend to have higher scores.
engagement Number of emails opened, number of replies, number of calls, number of meetings, date of last activity Opportunities with active engagement have high scores, while deals with inactivity drop sharply.
Movement of business negotiations Frequency of stage changes, number of stay days, and number of changes to scheduled closing date Business negotiations that proceed smoothly have a high score, while stagnation or postponement of the schedule results in a lower score.
conversation intelligence Competitive mentions, price concerns, positive/negative mentions in calls Competitive mentions and negative price mentions lower your score in real time.
MEDDIC input status MEDDIC custom property input rate/Economic Buyer identified flag Opportunities with all MEDDIC tend to have high scores (setting required)

Three ways to use AI Deal Score

Utilization 01
Weekly review prioritization
Filter by "Person in charge accuracy Commit and AI Score 50 or less" to prioritize and review disconnected deals. Managers should check “why AI is underestimating” and provide coaching to remove bias from those in charge.
Utilization 02
Automatic forecast correction
The “adjusted forecast” is calculated by weighting the subjective forecast of the person in charge with the AI ​​score. For example, refer to the number adjusted by "Total Commit Opportunities x Past Commit Realization Rate of Person in Charge" in the manager layer.
Utilization 03
Setting a sudden drop in score alert
Set up a workflow to detect deals whose AI Deal Score has dropped by 20 points or more from the previous week and notify managers. Establish a system that allows early detection of "business negotiations that are deteriorating without being visible" and allows intervention before it is too late.
Utilization 04
Evaluation of prediction accuracy by person in charge
A forecast accuracy ranking is created by aggregating the ``rate of actually closing deals that were said to be committed'' by person in charge. For those in charge with low accuracy, provide individual coaching on filling out MEDDIC and setting a valid closing date.
Section 8-4

How quota setting and percentage tracking works

Forecasting has meaning only when there is a "target (quota)". HubSpot's Forecast tool lets you set quotas at the rep, team, and company level and track their progress in real time. Settings are performed from "Sales → Forecasting → Set Quota".

Three options for setting quotas

Automatic approval for 10% or less, manager approval for 11-20%, and executive approval for 21% or more - these three stages are set as CPQ rules. Accumulate discount reasons and performance data through the approval flow.ContentSuitable case
Amount-based (Revenue) Set a monetary goal of "¥X orders this month" for each person/team. most common format B2B sales where the unit price of a contract varies widely - When managed using annual contract value (ACV)
Count-based (Count) Set a number goal of "close X cases this month" Product with uniform contract unit price / Low unit price and high turnover sales model for SMB
Activity-based Activity goal of "X calls/Y meetings this month." Can be set in parallel with order targets If you want to manage pre-order pipeline generation activities such as inside sales (SDR)

Visualization of quota achievement rate by person in charge

👤 Quota achievement rate by team (as of March 8, 2026 Q1)
Taro Yamada
¥11,200,000 / ¥10,000,000
112%
Hanako Sato
¥7,040,000 / ¥8,000,000
88%
Ichiro Suzuki
¥6,080,000 / ¥8,000,000
76%
Keiko Tanaka
¥3,640,000 / ¥7,000,000
52%
Yusuke Ito
¥2,280,000 / ¥6,000,000
38%
100% or more (exceeded achievement)
Submitted demo application form
Less than 70% (intervention required)
⚠️ “Continue increasing” the quota is not the only correct answer

If you continue to set quotas at 120% of the previous period, the motivation of those in charge with low achievement rates will drop sharply, leading to them leaving the company. Best practices for setting quotas areA level that 60-70% of the team can achieve.”. Quotas that everyone can easily meet eliminate opportunities for growth; quotas that no one can meet destroys morale. Adjust to a realistic level each quarter by referring to HubSpot's achievement data.

Section 8-5

Agree on the estimate sending date, upper limit on the number of revisions, and contract review period. Create quotes with HubSpot CPQ and centralize the closing process with electronic signatures (DocuSign integration)

Forecasting tools are meaningless if you just look at them.Weekly Forecast Review MeetingIt is important to structure data and design it as a place to transform data into decisions and actions. Here is a typical agenda design for managers to complete in 30-45 minutes each week.

📅 Weekly Forecast Review — Standard Agenda (45 minutes)
0-5 minutes
① Confirmation of this week's numbers (warm-up)
Share this week's Closed Won amount, number of items, and remaining target amount for this month with everyone. Check the difference from the previous week and visualize the pace of achievement. Screen share your HubSpot Forecast dashboard
5-15 minutes
② Commit Confirm progress of business negotiations (top priority)
Check all deals that are set to "Commit" to close within this month one by one. Ask the person in charge what the blocker is, what the next action is, and whether the deadline is realistic. Particularly dig deep into business deals that have a discrepancy with the AI ​​score.
15-25 minutes
③ Review deals where there is a discrepancy between the AI ​​score and the accuracy of the person in charge
Pre-filter and prepare opportunities for "Person in charge: Commit, AI Score: 50 or less". The manager analyzes with the person in charge why the AI ​​is undervalued. This is often caused by lack of MEDDIC, neglect of Deal Risk, and decreased engagement.
25-35 minutes
④ Look for opportunities to “commit” Best Case business negotiations
Check to see if any of the deals classified as Best Case can be made a Commit within this month. Ask the person in charge what we need to close this month, and the manager will provide the necessary support (calling a superior, approving special conditions, providing implementation examples)
35-45 minutes
⑤ Confirm pipeline for next month/next quarter
In addition to checking this month's outlook, we also check whether next month's pipeline coverage is sufficient. Also check the status of prospecting activities to avoid depleting the pipeline for next month and beyond due to concentrating on closing this month.

What not to do in a forecast review

anti-patternwhat is the problemwhat to do instead
Ask “How are you doing?” Questions to which the person in charge is likely to give an optimistic answer. Unable to extract specific information Ask specific facts such as "When was your last conversation with the decision maker? What did they say?"
Review all items at the same depth Due to lack of time, discussions on important business negotiations become shallow. Focus your time on commit deals and AI divergence deals, and briefly check pipeline deals.
Just report the numbers The report ends with “¥X” and the next action cannot be decided. Each session concludes with an action of ``who, what, and by when''. Log tasks instantly to HubSpot
Publicly criticize those in charge of low achievement rates Forecasting accuracy worsens as staff report numbers ``more optimistic than reality'' Personnel with low achievement rates will be provided with individual 1on1 support. Design the general review as a place to share information rather than a place to find problems.

📌 Chapter 8 Summary

Structurally eliminate the cause of forecast deviations

Customer pain, competitive advantage, NG messages, and the tone of the person in charge

Make sure everyone defines forecast categories

Document the definitions of Commit, Best Case, and Pipeline, and create a state in which all people in charge input data using the same standards. Particularly for organizations whose definition of ``commit'' is vague, start here first.

AI Deal Score is used to “discover discrepancies”

The list of discrepancies between the person in charge accuracy and the AI ​​score is the most important agenda item for the weekly review. We have also set up a workflow to alert you to a sudden drop in score to catch ``business deals that have deteriorated without you even being able to see them'' at an early stage.

Set the quota at a level that 60-70% can achieve.

There is no meaning in a goal that everyone can achieve or a goal that no one can achieve. Adjustments are made quarterly while referring to HubSpot's achievement rate data, and individual settings are made according to differences in performance for each person in charge.

Weekly reviews proceed in the order of Commit → Deviation → Best Case

→ Used to provide individual coaching to personnel with low activity levels and to improve the skills of personnel with high activity and low order intake.

Manage prediction accuracy itself as a KPI

Aggregating the "close rate of deals called commit" by person in charge and team each quarter. The quickest way to raise the quality of forecasts across the entire organization is to horizontally spread the methods of people and teams with high forecasting accuracy to the entire organization.

Next Chapter
Chapter 9: ABM — Aim for large orders with target account strategy →