Free tool: Decision Check
What you put in decides what you get back.
Decision Check gives you an outside read on one real issue in your business. Here is how to fill it out, and three examples from the same mid-sized LBM dealer so you can see what a good entry and a useful answer look like.
Time
Five minutes
Cost
Free, no account
Privacy
Nothing you type is saved or used to train an AI model.
On this page: How to fill it out · Three kinds of questions · Example: a situation · Example: a decision · Example: a problem · Common questions
How to fill it out
Five minutes, and most of it is context.
The form asks what you need help with, then about your business, then three questions about the issue. The same three questions apply to all three kinds.
01
Bring one real issue
Pick something you are dealing with now, not a hypothetical. One issue per run. If you have three, run it three times.
02
Choose what you need help with
Situation Clarifier, Decision Review or Problem Diagnosis. If you are not sure which fits, the section below shows the difference.
03
Tell it about your business
Industry, company size, your role and what the business focuses on. These fields are optional. Fill them in anyway. The context is what separates a useful answer from a generic one.
04
Answer the three questions with specifics
Numbers, timeframes, what you already tried and what you are worried about getting wrong. A few sentences each is enough. A vague entry gets a vague read.
05
Read it as an outside read
The answer separates what you told it from what it calculated, what it infers and what it recommends. Check the facts first. The decision is still yours.
What to put in each question
| Question | What to include |
|---|---|
| What situation are you currently dealing with? | The issue in plain terms, and the decision in front of you if there is one. |
| What is not working or causing concern right now? | What you are seeing, with numbers where you have them. What you already tried and how it went. |
| What do you believe is at risk if this is not resolved? | Cost, customers, people and timing. Any goal or deadline you are working against. |
Leave out names. Customer and employee names add nothing to the answer. Describe the role or the account type instead.
Three kinds of questions
An unclear situation, a choice, or a problem that keeps coming back.
Each one needs a different answer, so Decision Check handles each one differently.
| Kind of question | Primary job | What you get |
|---|---|---|
| Situation Clarifier | Frame an ambiguous situation | A working interpretation, the critical unknowns and the next step needed to gain clarity |
| Decision Review | Choose among known alternatives | A recommendation, confidence level, assumptions and switch conditions |
| Problem Diagnosis | Find the cause of a recurring problem | Ranked causes, the first controlled test and a clear measurement standard |
The business behind all three examples
Each example uses the same company, so you can see how the same context produces three different kinds of answers.
| Field | Entry |
|---|---|
| Industry | Lumber and building materials (LBM) |
| Company Size | Mid-sized, multiple branches |
| Role | President |
| Business Type or Focus | Pro dealer with delivery and installed sales |
Example 1 · Situation Clarifier
One Branch Does Not Feel Right
Something is off and you cannot name it yet. Start here to get underneath the symptoms.
The entry
- What situation are you currently dealing with?
One of our branches does not feel right. Customer complaints are going up and issues take longer to resolve. More decisions are coming up to me and the regional manager that the branch used to handle on its own. - What is not working or causing concern right now?
The branch manager and the team say they are overloaded and keep asking for more people. I am not sure that is the real problem. Revenue and branch profit have not dropped enough to explain what we are seeing. - What do you believe is at risk if this is not resolved?
Customers we have had for years, and the branch’s reputation. I do not want to add headcount or replace the branch manager based on a feeling and find out later it was the wrong call.
The answer
Scope of this answer This is an initial outside read based only on the information provided. It identifies the most likely interpretation and the next questions or tests; it is not a full diagnosis.
Working interpretation
The branch appears to be losing operating control before showing a clear financial failure. Customer complaints, slower decisions, more escalation and repeated requests for additional people suggest that work is no longer moving cleanly through the branch. Those symptoms could come from excess workload, but they could also come from unclear authority, weak handoffs, rework or a branch manager who is no longer resolving issues locally.
There is not enough evidence yet to approve more headcount or conclude that the branch manager is the problem. The immediate task is to determine whether the branch has a true capacity shortage or is consuming capacity by correcting mistakes and escalating decisions.
What we know
- Customer complaints are increasing and issues are taking longer to resolve.
- More decisions are being pushed upward instead of handled within the branch.
- Managers and employees describe the branch as overloaded and are asking for more people.
- Revenue and branch profit have not yet declined enough to explain what is happening.
What we infer
The evidence points to weakening operating health, but it does not identify one cause. The leading possibilities are hidden rework, unclear decision rights, a bottleneck in one department, or leadership behavior that encourages escalation rather than ownership. A genuine staffing shortage remains possible, but it should be proven by workload and throughput data rather than by how busy the branch feels.
What remains unclear
- Whether complaints share a common failure point such as order entry, purchasing, staging, delivery or installation.
- Whether workload has grown or existing work is being repeated, delayed or passed between departments.
- Why decisions that were previously handled at the branch are now being escalated.
- Whether experienced employees are disengaging, withholding judgment or compensating for weak processes.
Three questions that matter next
- Which recent customer complaint best represents the branch’s problem, and where did the work first go wrong?
- If the branch received another employee tomorrow, what recurring work would that person perform, and what evidence shows that work cannot be removed or redesigned?
- Which decisions have moved upward during the past 60 days, and did each move because of missing authority, missing information, weak capability or fear of accountability?
Recommended next step
Run a two-week clarification review owned by the regional manager and supported by the branch manager. Establish a short baseline of complaints, corrections, overtime, unresolved issues and escalations; trace three representative customer problems from request through resolution; and hold structured conversations with the people closest to the work. At the end of two weeks, decide whether the branch needs added capacity, a process correction, clearer decision rights or a leadership response. Do not add permanent headcount or change leadership before that review is complete.
Example 2 · Decision Review
Buy a New Truck or Improve Routing First
You know the options. You have been going back and forth. Start here to get an outside read.
The entry
- What situation are you currently dealing with?
We are deciding whether to buy a new truck and hire a driver before the spring season, or put routing and scheduling software in place first and see what that frees up. - What is not working or causing concern right now?
Dispatch says we are out of trucks. Average truck utilization is about 68 percent, but some days are much tighter. About 12 percent of deliveries need a second trip because of missing material, order changes, loading errors and other causes. - What do you believe is at risk if this is not resolved?
A new truck is about $165,000 and a driver costs about $72,000 a year loaded. The software would be about $45,000 the first year, and I do not know the recurring fees yet. Guess wrong and we either carry a fixed cost we did not need or miss deliveries in the spring.
The answer
Scope of this answer This is an initial outside read based only on the information provided. It identifies the most likely interpretation and the next questions or tests; it is not a full diagnosis.
Decision
Should the company commit to a new truck and driver before the spring season, or first improve routing and scheduling while using temporary capacity for verified peaks?
What we know
- Average truck utilization is approximately 68 percent, although peak days may be materially higher.
- About 12 percent of deliveries require a second trip because of missing material, order changes, loading errors or other causes.
- A new truck costs approximately $165,000, and the driver’s loaded annual cost is approximately $72,000.
- Routing and scheduling software would cost approximately $45,000 in the first year, before any recurring fees not yet confirmed.
What we calculated
The truck and driver create a three-year commitment of approximately $381,000 before fuel, maintenance, insurance, financing and replacement risk. The 12 percent second-trip rate also means part of the apparent capacity shortage may be avoidable work rather than a shortage of trucks.
Options and tradeoffs
| Option | Advantage | Primary risk |
|---|---|---|
| Buy now | Adds durable capacity before spring and may protect service on peak days. | Creates a large fixed commitment before proving that peak capacity is the real constraint. |
| Software first | Improves visibility, routing and daily discipline with a smaller, more reversible investment. | Will not solve poor order completion, loading bottlenecks or a true peak shortage. |
| Staged approach | Tests the causes of second trips, pilots routing improvements and covers peaks with rentals or carriers. | Requires management discipline and may delay a truck purchase if the diagnostic is poorly run. |
Recommendation
Choose the staged approach. Authorize a four-to-six-week capacity review and routing pilot, but do not authorize the truck or a full software purchase until the company has validated peak demand, recurring fees, implementation timing and the causes of second trips. Use a disciplined manual routing process immediately, evaluate software in parallel and reserve rental or third-party capacity for known peak periods.
Confidence level Moderate. The recommendation is more reversible and directly addresses the avoidable work already visible in the data. Confidence is limited because average utilization can hide Monday, Friday and month-end shortages, and the current information does not show capacity by day, route or driver hour.
What would change the answer
- Peak-day analysis shows the fleet remains fully utilized after avoidable second trips are removed.
- On-time delivery or customer retention is being materially damaged by a persistent truck shortage.
- Rental or carrier capacity is unreliable or costs more than the annualized ownership cost of the truck.
- Routing software cannot be implemented or adopted before the spring season.
Immediate actions
- Measure deliveries for the prior eight to twelve weeks by day, route, truck, driver hours, second-trip cause and promised versus actual delivery time.
- Confirm software implementation timing, integration requirements, recurring fees and dispatcher usability before committing to a vendor.
- Secure rental or carrier availability for the spring peaks, then revisit the truck decision after four to six weeks of stable data.
Example 3 · Problem Diagnosis
Recurring Picking and Delivery Errors
A recurring problem you have not been able to solve. Start here to find what is actually driving it.
The entry
- What situation are you currently dealing with?
We keep making picking and delivery errors. About 57 of our 2,100 deliveries a month need a correction. We estimate the direct cost at $42,000 a month, before the disruption to customers. - What is not working or causing concern right now?
Incorrect quantities are about 45 percent of the errors, missing material 30 percent, wrong product 15 percent, and damage or improper loading 10 percent. Errors go up on high-volume days and when an order changes after the pick ticket is printed. We have trained on it before. It helps for a while, then slides back. - What do you believe is at risk if this is not resolved?
The cost, customer confidence and the crew’s time. Our goal is under 1 percent in 90 days and under 0.5 percent in six months.
The answer
Scope of this answer This is an initial outside read based only on the information provided. It identifies the most likely interpretation and the next questions or tests; it is not a full diagnosis.
Problem definition
The company is correcting approximately 57 deliveries each month out of 2,100, creating an estimated direct cost of $42,000 per month plus customer disruption. Errors increase on high-volume days and after orders change following ticket printing. Training helps temporarily, but the improvement does not hold. The evidence points to a process-control problem that training alone will not solve.
What we know
- Approximately 2.7 percent of deliveries require a correction.
- Incorrect quantities account for about 45 percent of errors, missing material for 30 percent, wrong product for 15 percent, and damage or improper loading for 10 percent.
- Error rates rise on high-volume days and when orders change after the original pick ticket is printed.
- Previous training produced temporary improvement followed by regression.
What we calculated
Reaching the 90-day goal of less than 1 percent would reduce corrections to fewer than 21 per month. Reaching the six-month goal of less than 0.5 percent would reduce them to fewer than 11 per month. Those are meaningful operating targets, but improvement should not come at the cost of a loading bottleneck or late departures.
Ranked causes to test
| Likely cause | Why it fits | What could disprove it |
|---|---|---|
| Post-ticket changes lack one controlled version | Errors rise after ticket printing, and quantity or missing-material errors dominate. | Changed orders have the same correction rate as unchanged orders when complexity is comparable. |
| Final verification is inconsistent | The error is reaching the truck or jobsite instead of being caught at loading. | Verified high-risk loads do not perform better than similar unverified loads. |
| Peak volume exceeds process capacity | Errors rise on high-volume days and may reflect congestion, late changes or rushed loading. | Error rates remain unchanged after controlling for volume, staffing and order complexity. |
| Product data or labeling causes confusion | Wrong products and quantities may cluster around a small group of similar items or units of measure. | A SKU-level review shows no meaningful concentration by product, label or unit of measure. |
First controlled test
Run a two-week test on changed orders at one location or on one defined delivery wave. Require one clearly marked revised ticket, remove or void the old version, and name the person responsible for communicating and confirming the change. Add a brief physical verification before loading for changed orders and other high-risk loads. Keep a comparable group under the current process so the result is not confused with a light week or a different order mix.
How to judge the result
- Compare correction rates for changed and unchanged orders before and during the test.
- Track added verification time, loading delays, late departures and any workarounds.
- Confirm whether the change reduced errors at the source or merely detected them later.
- Reject the leading diagnosis if changed orders do not improve relative to the comparison group; then move the investigation toward capacity, product data or staffing patterns.
Recommended owner
The operations leader should own the test, with sales or order entry responsible for change communication, picking responsible for item and quantity accuracy, loading responsible for final condition and load sequence, and dispatch responsible for releasing only verified high-risk orders. Refresh training after the process is changed, not as a substitute for changing it.
Common questions
Before you run it.
Will it make the decision for me?
No. It gives you an outside read: what is likely going on, what it would check and what it would do next. You know things about your business it cannot. The call is yours.
What if the answer misses?
Look at the entry first. A thin entry gets a thin answer. Add the numbers, what you already tried and what you are worried about, then run it again.
Is anything I type saved?
No. Nothing you type is saved or used to train an AI model.
Bring one real decision.
Five minutes, no account. Nothing you type is saved or used to train an AI model.
