This page shows how the Sapien Amplified process unfolds from the first question to the finished result. Each example follows one decision, deal, or product from the opening request through the questions, calculations, reviews, and materials the system produces along the way.
See what AI finds, why it matters, and what you can do with the results.
Follow eight examples from the first request through the questions, calculations, reviews and practical materials. Each example shows how Sapien Amplified guides AI through the work, and how the findings affect the next step.
About these examples: the situations, figures, exchanges and output excerpts are fictional illustrations of the product instructions. They are not customer results or transcripts of completed AI runs. Separate AI reviews shown here are illustrative; an actual independent review requires a separate reviewing conversation. Excerpts show selected parts of longer outputs.
See the findings, the recommendation and the information that would change the next step. Includes a sample brief and full dossier for the acquisition example.
Should You Buy Another Business or Expand Your Existing One?
An extra-sales estimate enters the forecast without evidence. The original payment terms fail your cash requirement. Follow how AI checks the numbers, examines the alternatives and turns the findings into a limited review plan.
01Test the purchase against profit, cash and your time.
You own a service business and have $900,000 available. A smaller competitor is for sale. You are considering buying it or expanding your existing business. The immediate choice is which route deserves a limited investigation.
YOUR REQUIREMENTS
“Add $120,000 in annual operating profit by the end of year two, before financing and tax. Keep $300,000 available throughout the takeover. Use no more than eight additional hours of my time each week, at every stage.”
AI examines the financial summary, staffing estimates, proposed terms, cash assumptions and earlier discussion against all three requirements. Passing the profit test alone would not establish that the purchase works.
02Remove unsupported sales and recalculate the purchase.
An earlier AI suggestion proposed adding 20% of the competitor’s $2 million annual revenue, and said in the same sentence that it had no customer test to support that estimate. The buyer’s later forecast note then recorded $400,000 in additional annual revenue, with no such qualification attached. The estimate was labeled unsupported when it was made, and reached the forecast as a number. AI traces it back to that origin and excludes it from the operating-profit calculation.
Illustrative output excerpt
What the figures support, and what they do not
Supplied information
Examination finding
$320,000 annual earnings; the seller’s management work is unpaid.
$320,000 minus $110,000 replacement management, $24,000 staff raises and $12,000 systems leaves $174,000 annual operating profit, before financing and tax. The year-two timing and the net improvement across both businesses remain unestablished.
$400,000 extra sales are included in the buyer’s forecast.
There is no documented buyer test. Of 25 candidate organizations, nine already buy the proposed service. Eight of the remaining 16 would need to buy at $50,000 a year. That is a required result, not demonstrated demand.
$650,000 is due at closing, plus $70,000 in fees and setup.
$900,000 − $650,000 − $70,000 leaves $180,000, which is $120,000 below the cash minimum before operations begin.
The buyer proposes $450,000 upfront with $200,000 deferred.
The lowest calculated month-end balance over four months is $320,000. One $40,000 receipt delayed from month 1 to month 3 lowers it to $280,000. The seller has not agreed; within-month lows and later cash needs still require checking.
WHAT THIS MEANS FOR YOU
Payment timing can improve early cash without improving operating profit. Staffing affects profit and your workload. Losing the largest customer would reduce estimated operating profit from $174,000 to $30,000 with other inputs unchanged; the customer has not said it will leave. Those requirements must work together before a purchase can meet your goal.
03Challenge the next step before committing more time or money.
The Solution Spectrum compares the immediate choices: a two-week acquisition review capped at $2,000 and six hours weekly, or a six-week expansion test capped at $8,000 and six hours weekly. Running both original tests together would require 12 hours a week, above your eight-hour limit.
With Cross-Model Decision Triangulation, a separate AI model first examines the facts and requirements without the first model’s conclusion. The examinations are then compared. The dossier’s appendix describes corrections from the reported review exchange; it is not a complete raw transcript.
Illustrative output excerpt
A challenge that changes the action plan
Earlier interpretation
Correction and practical effect
Rejecting the proposed deferred payment ends the acquisition route.
One rejected proposal does not eliminate every possible price or payment arrangement. Ask which terms the seller would consider, and calculate each proposed structure separately. Do not assume alternatives are available or affordable.
Seller questions happen before the review allowance starts.
Seller contact and your own records work consume the allowance. Day 1 starts with that work; there is no unbudgeted preliminary week.
Agreement between AI models would not establish customer retention, seller acceptance or staffing coverage. Those points still need evidence from the business.
04Start with a limited review and record when to stop.
Recommendation: investigate the acquisition first for up to two weeks, $2,000 and six hours weekly. It is cheaper and faster than the proposed expansion test, and workable terms and usable records may be obtainable within that allowance.
That recommendation weakens if useful records or workable terms cannot be obtained. A service offer ready to sell and suitable prospects could favor testing expansion first; their readiness is unconfirmed.
Illustrative output excerpt
Your review plan and next decision
What to resolve
What to do with the result
Can the terms preserve $300,000 throughout takeover?
Request actual terms and dated receipts and payments. The original terms fail. Any alternative needs its own cash calculation.
Can the purchase meet the profit and time requirements?
Reconcile earnings and costs; establish which seller duties the manager covers and what remains yours. Recalculate profit if staffing costs change.
What happens at Day 14?
Record spending, hours, findings and gaps. Pause paid acquisition work if essential terms or records remain missing. Any extension, expansion test or purchase requires a separate decision.
This is a proposed next step, not a recorded user authorization or approval to buy. The record keeps the recommendation, reasons, uncertainties and review point together.
See how Decision Alpha shaped the analysis above, and how the results compare with an ordinary AI answer.
↓
WHAT THE METHOD CHANGES
How Decision Alpha Forced AI to Think for this Specific Business Acquisition Example
Should you buy another business or expand your existing one? You want $120,000 more annual operating profit by the end of year two, at least $300,000 available throughout the takeover, and no more than eight additional working hours each week. In this fictional example, the following themes from Decision Alpha’s 27 dimensions change what AI examines, what it accepts and what it recommends.
Framing & Anchoring
A profitable business can still fail your requirements.
AI tests the purchase against your profit, cash and time limits together. The original payment terms fail your cash requirement, while your workload remains unconfirmed. The immediate question becomes which route deserves a limited investigation.
Provenance & Quarantine
An AI suggestion became a $400,000 sales forecast.
AI traces the $400,000 sales forecast to an earlier suggestion that explicitly lacked customer testing. The figure is excluded from operating profit. Repeating the estimate in a forecast does not supply the missing evidence.
Counterargument & Dissent
Buying the business must compete with expanding your own.
AI compares a limited acquisition review with an expansion test. It recommends investigating the purchase first, but identifies what could change that order: unavailable records, unworkable terms, or an expansion offer already ready to test.
Triangulation & Convergence
A separate examination leads to challenges and corrections.
The documented review corrects two interpretations: rejecting one payment proposal does not eliminate every purchase structure, and preliminary seller questions consume your review allowance. Agreement between models still cannot establish whether customers will stay or the seller will accept your terms.
Proportionality & Reversibility
The next commitment is limited before the larger one is considered.
The proposed acquisition review lasts two weeks, costs no more than $2,000 and uses six hours weekly. If essential terms or records remain missing, paid work pauses. Further investigation and any purchase require separate decisions.
Repair & Traceability
Corrections change the plan, and you can see why.
The record preserves the earlier interpretation, the objection, the correction and its practical effect. Seller contact counts from Day 1. Different payment structures require fresh cash calculations. The revised plan retains the reasons behind those changes.
Why are Decision Alpha's Results Different
Ordinary AI can perform these checks when asked. Sapien Amplified specifies them in a structured process, reducing the work you must do to request, challenge, organize and preserve the analysis. The comparison below describes the intended differences, not limits on what every ordinary AI conversation can do.
Ordinary AI Conversation
Sapien Amplified
Answers the question you asked.
Improves the question before answering it.
Your requirements are mentioned once.
Your requirements become the tests every result is measured against.
Numbers are presented.
Numbers are traced to the inputs they came from.
Estimates blend into forecasts.
Unsupported estimates are kept out of the result.
Findings are stated.
Each finding appears beside what could weaken it.
The preferred option wins.
The preferred option is compared against a priced alternative.
One answer.
Investigating is kept separate from committing.
A recommendation.
A next step with a budget, an hours cap and a stop date.
Confident prose.
Uncertainty stays visible after the report is finished.
Issues are handled separately.
Related findings are examined together.
The arithmetic is checked.
The origin of each forecast is examined, not just the arithmetic.
One model's view.
A separate model's examination, with disagreements kept.
Revisions replace the earlier text.
Corrections show what changed and why.
Gaps are smoothed over.
Unresolved questions are left unresolved.
You receive a conclusion.
You get the reasoning behind every finding.
Advice.
What to ask for, when to continue, when to stop, and when to review.
The reasoning remains spread across the conversation.
The original reasoning is kept for later comparison.
One document.
Organized so you can act on it and audit it.
What Those Misses Could Have Cost You
An answer can acknowledge uncertainty and still give you an incomplete basis for committing money. In the fictional business-purchase comparison, these are the specific gaps and the exposure they could leave you with.
You could commit to terms that breach your cash minimum.
The miss: The ordinary answer describes the purchase as workable on earnings without identifying the failed cash requirement.
The consequence: Original payment terms leave $180,000 against the $300,000 you require, before operations begin. You would start $120,000 below your minimum.
You could underestimate how much one customer matters.
The miss: The ordinary answer discusses customer concentration but does not calculate the profit remaining if the largest customer leaves.
The consequence: Estimated annual operating profit falls from $174,000 to $30,000, with other inputs unchanged. That is below your $120,000 goal. Customer departure is a scenario, not a prediction.
You could give an untested forecast too much weight.
The miss: The ordinary answer acknowledges that the extra sales are unproven, yet calls that upside the strongest argument for pursuing the purchase.
The consequence: $400,000 in unsupported sales could influence how attractive the opportunity looks. The examination traces that figure to an untested AI suggestion and excludes it from operating profit.
You could keep spending without a defined stopping point.
The miss: The ordinary answer recommends further investigation without specifying a spending cap, hours allowance or review deadline.
The consequence: You could spend more time and money while essential information remains missing. The defined review limits work to two weeks, $2,000 and six hours weekly, with rules for pausing paid work.
The ordinary answer recommends investigation, not an immediate purchase. These are possible consequences of acting on incomplete analysis, not observed losses or proof that the business should never be bought.
Compare the Answers and Supporting Work
Compare an illustrative out-of-the-box AI answer with the Complete Dossier and Decision Brief prepared using Decision Alpha. All three documents address the same fictional business materials.
What You Leave With
What you Receive
Without Decision Alpha
With Decision Alpha
Documents
One three-page answer
33-page dossier and three-page brief
Next steps
General direction and suggested follow-up
Defined review plan with spending, time and stopping rules
Supporting work
Selected calculations and cautions
Detailed calculations, source register and documented corrections
Materials to use
Offers to prepare further materials
Information request, evidence tracker and decision worksheet included
The out-of-the-box answer is a fictional illustration, not a transcript of a recorded AI run. The comparison shows what these sample documents contain; it does not establish how every ordinary AI conversation will perform.
Before Decision Alpha
After Decision Alpha
WITHOUT DECISION ALPHA
The Out-of-the-Box AI Answer
This is a standard AI-generated answer to our question, not a Decision Alpha result. Read it only as a baseline, then compare it with the Decision Alpha outputs on the right to see the improvements in this example.
Both outputs were produced through full triangulation, with multiple AI models independently examining, challenging and correcting the analysis.
Complete Dossier · 33 pages
Examine the calculations, sources, assumptions and documented review corrections. Includes a seller information request, an evidence tracker and a worksheet for the next decision.
At $230 instead of $200, 87 customers preserve today’s revenue. But losing low-cost customers could change the result. Follow how AI separates the arithmetic from the pricing decision.
01Keeping revenue steady does not establish a better business.
You sell a recurring service to 100 customers at $200 a month. You are considering $230 and have asked AI how many customers you can lose without reducing revenue. AI answers the arithmetic, then asks whether revenue is the right measure for your decision.
AI
“Is the goal to maintain revenue, increase contribution after delivery costs, reduce your workload, or a combination? Which customers account for the most delivery work?”
At the current price, monthly revenue is $20,000. At $230, 87 customers generate $20,010. That calculation describes revenue only. It does not establish which customers will stay, the cost of serving them, or whether the increase will be worthwhile.
02Test the new price and keep the retention question open.
Testing the new price with new prospects provides evidence about new sales without changing current customers’ renewal terms. It cannot establish how many existing customers would stay at $230. That question remains open and needs a separate decision about whether and how to test renewals.
Illustrative excerpt · Selected parts of the output
Decision Record · Limited pricing test
Chosen scope: test the $230 price with a defined group of new prospects. Existing customer renewals remain a separate decision.
Record: qualified prospects, accepted proposals, objections, delivery hours, direct costs, and contribution per new customer.
Compare: similar prospects and a comparable buying period. Explain any differences that prevent a fair comparison.
Reconsider: if higher delivery costs or a change in customer mix offsets the price increase. A small test can justify another bounded step without proving the response of all existing customers.
A higher salary does not establish that the move will work for your household. Follow how AI checks the job against childcare, your partner’s work and the time you want at home.
01Check whether the job schedule fits your family requirements.
You receive a job offer that requires relocating. The salary is higher, but the decision also affects your partner’s work, housing, childcare, and time at home. You provide the offer, current household budget, housing estimates, and the limits you want respected.
YOU
“I want the career opportunity, but we need dependable childcare and at least three evenings at home each week. I do not want the salary increase to hide what the move would cost us.”
AI separates documented facts from estimates. The salary and required office days are in the offer. Childcare availability, actual commuting time, and your partner’s future income are still unconfirmed. It examines accepting, negotiating a later move, negotiating fewer office days, and staying, where those options are available.
02Get the answers that would change the acceptance decision.
Illustrative output excerpt
Questions to resolve before accepting
Question to ask
Why the answer matters
Employer: “Which office days and working hours are required? Which changes would you confirm in writing?”
Establish the schedule that commuting and childcare must accommodate. A possible hybrid arrangement is not an agreed term.
Childcare provider: “Is a place available for our start date, what are the hours, and what happens if pickup is late?”
Check actual availability and timing against the workday and commute.
Household review: “What would remain after housing, childcare, moving costs and a period of lower partner income?”
Compare the household benefit rather than the salary increase alone. Keep unconfirmed costs and income labeled as estimates.
Illustrative excerpt · Selected parts of the output
Decision Record · Job and relocation
What matters
What you retain
Requirements
Reliable childcare and three evenings at home. These are your priorities, carried through the comparison.
Next action you choose
Request a written answer on the office schedule and confirm childcare availability before making an acceptance decision.
Uncertainty
A possible hybrid arrangement is not an agreed term. Your partner’s possible income remains an estimate.
Revisit trigger
The employer’s response, confirmed childcare, and a household budget that includes moving costs and a period of lower income.
Check whether the financial figures, forecasts and terms support the investment or fundraising description, then identify the evidence needed to resolve the important gaps.
What Should You Know Before Making This Investment?
The materials describe $2 million in recurring revenue and 18 months of cash. Follow the checks that show what those descriptions still need to establish, how the gaps affect your exposure, and which records to request.
01Make the investor’s limit part of the examination.
You are considering a minority investment in a private B2B software company. The material describes $2 million in annual recurring revenue, an $8 million enterprise valuation, and 18 months of cash runway. You provide the presentation, billing export, management cash report, and draft investment terms.
YOU
“I am considering $100,000. I will not accept a requirement to contribute more capital later. My next step is deciding what evidence to request before any commitment.”
AI carries the prohibition on required additional capital into the review of cash needs and terms. It examines recurring revenue, proposed ownership and dilution, customer dependence, and the investor’s obligations separately.
02The revenue measure and cash forecast need reconciliation.
Illustrative excerpt · Selected parts of the output
Three financial checks
Supplied information
Examination and limitation
S1 · Presentation: $2 million ARR. S2 · Annual billing export: $2 million total, including $400,000 implementation fees.
The export contains $1.6 million in subscription billings. Historical billings and current ARR are different measures. This export cannot establish the stated ARR without contract dates, recurring charges, and adjustments.
S1 · $8 million enterprise valuation.
The stated valuation divided by stated ARR is 4×. If verified ARR were $1.6 million, the multiple would instead be 5×. That second calculation is conditional, not a finding that ARR is $1.6 million.
S3 · $550,000 cash and recent net cash use of $50,000 a month. S1 · 18 months runway.
Cash divided by unchanged monthly net cash use gives 11 months. The 18-month forecast needs a bridge showing the timing and support for lower spending, greater receipts, or additional funding.
AI keeps those distinctions visible. An unexplained difference is not proof of misconduct. A forecast can have sound arithmetic and still depend on future sales that have not occurred.
03Connect cash needs to the obligation you will not accept.
The draft refers to future financing but does not clearly settle whether you could be required to provide additional capital. AI identifies the ambiguity and requests the operative provisions and related agreements. It does not infer that silence means no obligation or present a legal interpretation as a completed legal review.
04Keep each finding and unresolved question visible.
Illustrative excerpt · Selected parts of the output
Deal Sheet · Private software investment
KEY INFORMATION MISSING
PRELIMINARY DEAL SHEET. Illustrative review date: 13 September 2026. Current source set: S1 presentation, S2 billing export, S3 cash report, S4 draft terms. This excerpt illustrates an incomplete examination; intellectual-property ownership and share-transfer restrictions remain unexamined.
Material point
Current state and effect
P1 · Current recurring revenue
Needs outside proof. Obtain a contract-level recurring-revenue reconciliation. The billing export cannot settle the run rate.
P2 · Forecast runway
Needs outside proof. The supplied cash and recent net cash use imply 11 months if unchanged. The forecast bridge to 18 months is missing.
P3 · Any required future contribution
Needs outside proof. An ambiguous central term could conflict with your explicit prohibition on mandatory additional capital. No breach is established from the current draft.
P4–P5 · IP ownership and transfer restrictions
No state assigned. These two material areas remain unexamined and are carried into the next review.
What the supplied records support: the supplied export records subscription billings to identified customers. That supports a narrower description of an operating subscription business, subject to the records’ completeness and authenticity; it does not by itself establish retention or future growth.
What remains unresolved: the current documents do not settle the recurring-revenue measure, forecast cash needs or possible future obligations. Resolving those gaps could change the valuation analysis and the obligations you would accept.
Separate findings, no total score. Each material point retains its evidence, state, and consequence. Your investment decision is recorded separately.
05Prepare the requests needed to resolve the gaps.
Illustrative excerpt · Selected parts of the output
Priority evidence requests
Request
Likely holder · Why it matters
“Please reconcile current ARR to the underlying customer contracts, with recurring charges, start and end dates, one-time fees, credits, and cancellations identified.”
Finance lead or CFO · Resolves the revenue definition and the valuation calculation.
“Please provide the monthly cash forecast and a bridge from current cash and recent net cash use to the stated 18-month runway, with financing assumptions separate.”
Finance lead or CFO · Shows which parts depend on operating performance and which require new money.
“Please provide the provisions and any related agreements governing additional capital contributions, funding obligations, and the consequences of not participating.”
Company counsel or transaction counsel · Enables the appropriate qualified review of the obligation you will not accept.
These are prepared requests for you to review and send. When new records arrive, AI re-examines the affected points, preserves the earlier Deal Sheet, and dates the changes. A repaired revenue reconciliation does not automatically resolve the funding obligation. The remaining material areas still require examination.
If you later proceed, Compare With the Outcome can examine the subsequent results against this dated record. You can see what was supported, what remained uncertain, and which expectations held up. The Deal Sheet does not tell you to buy, sell, or commit capital.
Ten customers came through the founder’s relationships and two through referrals. The proposed raise funds untested outbound sales. See how AI separates evidence of current demand from evidence that the proposed growth plan will work.
01Twelve customers do not establish a repeatable outbound channel.
You are raising capital to expand a business. The presentation describes sales as repeatable, but most customers came through your own relationships. You are preparing for substantive investor diligence, where the growth assumptions and use of funds must be supportable.
AI
“Which customers came from the acquisition process you propose to fund? What records show the cost, sales cycle, conversion, and repeatability of that process?”
YOU
“We have 12 paying customers. Ten came through my relationships and two through referrals. The outbound process in our plan has not been tested.”
The paying customers support evidence of demand from those sources. They do not establish that a funded outbound campaign will produce similar results. AI examines the cash forecast, acquisition assumptions, and use of funds against that narrower evidence.
02Repair the fundraising language and the underlying plan.
Illustrative excerpt · Selected parts of the output
Fundraising preparation · Before and after
Original wording
Supportable replacement
“We have a repeatable customer-acquisition engine ready to scale.”
“We have 12 paying customers acquired through founder relationships and referrals. We have not yet established repeatable outbound acquisition.”
“The raise will scale a proven channel.”
“A defined portion of the proposed raise would fund an outbound acquisition test. Its budget, duration, success measures, and effect on runway must be documented before this funding case is ready for substantive diligence.”
Not ready for the stated discussion
The stated discussion is substantive investor diligence. The central acquisition and cash dependencies lack necessary support. This is different from saying that you cannot have an introductory conversation.
Repair the evidence: separate relationship sales from the channel you propose to fund.
Repair the economics: model the cash effect of a longer sales cycle and define a test within the available runway.
Prepare the answer: explain what the proposed test would establish, what it cannot establish, and what happens if it underperforms.
Retest: review the revised forecast and use of funds. Better language does not upgrade an untested acquisition process into proven performance.
Paid consulting shows that owners pay for pricing help. It does not establish that they will buy a workbook. Follow the comparison, product build, independent review, paid test and repair after a buyer uses it.
01Compare the workbook, course and template pack.
You advise small service businesses on pricing. You have completed paid consulting projects, saved useful worksheets, and built relationships with owners who face similar pricing questions. You want products that can sell without requiring a full consulting engagement from you each time.
YOU
“Use my pricing work to explore products I could sell. I have six hours a week and $300 available for the first test. I want buyers to get useful results from the first purchase.”
AI
“Which pricing problems have clients actually paid you to resolve? What can you reuse, who can you reach directly, and which parts still require your personal judgment?”
AI considers a workbook, a short video course and a proposal template pack. It checks exclusions, then compares speed to test, demand evidence, buyer access, reusable assets, later purchases, maintenance, reputation risk and differentiation relevant to the buyer. The table shows the differences driving the first test; demand for each product is still unconfirmed.
Illustrative output excerpt
Why test the workbook first?
Option
Reason to choose it / main limitation
Pricing workbook
Uses the existing worksheets to let owners compare prices, customer counts and delivery costs. It can address the chosen pricing task directly, but buyers must be able to use it without a consulting engagement. That remains untested.
Short video course
Can explain the pricing method, but requires lessons to be planned and recorded. It takes more preparation than adapting the existing worksheets and still needs exercises to test different numbers.
Proposal template pack · Runner-up
Could help owners communicate a price and scope. It does not itself calculate how price and customer mix affect contribution. It becomes the stronger candidate if buyer evidence points to proposal preparation as the more pressing paid problem.
The workbook is selected for the first test because the existing calculations directly address the chosen task within the limited time. The proposal pack remains the runner-up. Neither selection establishes what buyers will purchase.
02Check the evidence and repair the first version.
Illustrative excerpt · Selected parts of the output
Product Sheet · What is supported?
Finding
Evidence label and limit
Prior consulting clients paid for pricing analysis.
DIRECT BUYER EVIDENCE for the consulting service and the problem it addressed. It is not direct purchase evidence for this workbook.
Comparable pricing tools are advertised for sale.
MARKET SIGNAL. Their existence does not establish their sales or the demand for your version.
Owners can obtain the intended result without your intervention.
UNTESTED until representative buyers use the product and their experience is examined.
Buyers will pay $149 for the workbook.
UNTESTED before the buyer test. The proposed price is an assumption to examine.
AI prepares the proposed product, comparison, sources, assumptions, and buyer-test plan for a separate reviewing conversation. The review can retain the selection, require repairs, or identify a stronger runner-up.
ILLUSTRATIVE INDEPENDENT REVIEW
“REPAIR BEFORE BUYER TEST. The spreadsheet calculates a new price but does not adequately explain which expenses belong in the delivery-cost input. A buyer could confuse revenue with contribution. Add a worked example and clarify the required inputs before testing.”
Before the test, AI separates revenue from contribution, explains delivery costs in the guide, labels excluded overhead and taxes, and adds a completed example. The illustrative review allows a buyer test after those repairs and checks; it does not establish that every buyer will understand every input. In actual use, a separate review that has not occurred remains REVIEW PENDING.
03Build the workbook, instructions and completed example.
AI creates the supported workbook files, input instructions, completed example, and a short guide. The buyer must be able to obtain the promised result from that purchase. A later consulting engagement can be optional; it cannot be necessary to make the workbook work.
Illustrative excerpt · Selected parts of the output
Inside the product · Price and customer-count comparison
The figures below are the same pricing question shown in Example 02. There, a business owner faces the decision. Here, a consultant builds the tool that owner would use.
Buyer task: enter your price, customer count, and direct delivery cost per customer. Compare the current business with a possible price and retention outcome. The changed customer count below is an assumption, not a prediction.
Monthly measure
Current
Illustrative change
Customers
100
87
Price per customer
$200
$230
Revenue
$20,000
$20,010
Direct delivery cost per customer
$120
$120
Total direct delivery costs
$12,000
$10,440
Contribution after direct delivery costs
$8,000
$9,570
Calculation: contribution equals revenue minus direct delivery costs. It does not represent profit after fixed overhead, acquisition costs, taxes, or any other excluded expense.
What the buyer must examine: is $120 still an appropriate cost when the customer mix changes? Which customers would remain? What evidence supports assuming 87 customers? The workbook keeps these questions beside the calculation.
AI checks formulas, labels, example values, and the path through the files. An independent offering review examines the actual deliverables and intended buyer result. A completed plan or an attractive cover cannot substitute for usable, checked files.
04Write the paid-test rules before approaching buyers.
Illustrative excerpt · Selected parts of the output
Buyer-test plan · First workbook sales
Test element
Written before the test
Product and price
First usable pricing workbook with worked examples, $149.
Audience and period
20 relevant service-business owners, reached directly over 14 days. The period must fit their buying cycle.
Validity checks
The right owners see and understand the product. Purchase and access work. The number reached and time available meet the prewritten plan.
3 or more net paid purchases
Begin a limited selling and delivery period within the six-hour weekly capacity. This does not prove repeatable demand.
1–2 net paid purchases
Use the evidence to choose one change and prewrite another test. Do not assume the price caused the result.
0 net paid purchases
If the test was valid, use the recorded runner-up after its own independent examination. Do not treat a broken checkout as rejection of the product.
Limits
No more than $300 and 12 total hours for this initial test. Record spend and time already used.
AI also drafts the sales message, purchase-page copy, delivery instructions, and follow-up. Supported connections can be built and checked; anything awaiting an account connection is identified accurately.
Illustrative excerpt · Selected parts of the output
Purchase-page copy · A concrete buyer promise
Compare a price change before applying it to your business. Use the workbook to enter your current price, delivery costs, and customer count. Examine how different retention assumptions change revenue and contribution, with a completed service-business example and explanations beside the inputs.
The workbook does not predict which customers will stay. It makes your assumptions and their financial consequences visible.
Before broader selling, a separate buyer-journey review examines the sales promise, payment, confirmation, access, onboarding, and support. It receives the actual pages and deliverables. The record distinguishes what was verified, what was built but unverified, and what still needs an input or connection.
05Count three retained purchases after one refund.
Illustrative excerpt · Selected parts of the output
Buyer-test record · 14-day result
Observed result in this illustration
What it means
20 qualified owners reached; the test met its validity checks.
The recorded result can be assessed against the original plan.
4 purchases at $149; 1 full refund.
3 net purchases and $447 retained sales receipts before fees and other costs. The refunded purchase does not remain a successful paid result.
Several nonbuyers said the product sounded useful.
Weak interest is recorded separately. It is not counted as paid demand.
Prewritten result reached: 3 net purchases.
LAUNCH: the next limited period of selling and delivery. Preserve the original test, its costs, hours, and refund.
06Use the buyer’s error to make the instructions more precise.
In the illustrative follow-up, one buyer enters fixed monthly overhead as a per-customer delivery cost. The earlier repair added explanations to the guide, but the input label itself remained “Delivery cost.” This buyer’s error shows that the guidance beside the field still needs to be more specific. It does not establish why other buyers purchased or declined.
Illustrative output excerpt
The instruction changes where the buyer enters the number
Before the buyer test
Repair after the buyer’s error
Input label: “Delivery cost.” The separate guide explains delivery costs and excluded overhead.
Input label: “Direct cost to deliver for one customer.” Beside it: “Include costs such as per-customer service labor and materials. Do not enter total monthly rent or fixed overhead here; assess those separately when calculating overall profit.”
A completed example is available in the guide.
Put the example beside the input: “At $120 per customer and 100 customers, total direct delivery cost is $12,000.” Recheck the formula and example, then observe representative buyers using the revised field.
Illustrative excerpt · Selected parts of the output
Product repair and next-period record
Repair: place “direct cost to deliver for one customer” beside the input, with an included-expense example and a separate note on fixed overhead. Recheck affected calculations and examples.
Fulfillment: provide the corrected file and explanation to affected buyers through the authorized delivery process. Keep support obligations visible.
Next period: continue with the same initial audience source within the six-hour weekly limit. Record acquisition cost, delivery and support hours, refunds, and buyer completion.
Expansion: defer another channel until the next review can support it with acquisition and delivery evidence. A later product needs its own paid test.
Resume: return with the operating record and actual results on the chosen review date. Background monitoring is not assumed.
Test a service without leaving delivery undefined.
The first paid test produces one booking. See the delivery materials AI prepares, and what one booking does and does not establish about the price, the demand or the alternative.
01Define what the restaurant owner will receive.
You have nine years of restaurant-operations experience and have worked with 12 restaurants. You are considering a fixed-scope operational review for independent owners. Your newsletter has 600 subscribers but has not been used in two years; 900 LinkedIn connections do not establish current buyer interest. A distributor can introduce you to relevant owners.
AI examines access to those owners, the problem they would pay to resolve, your delivery capacity, and a service that can stand on its own. A spreadsheet product is retained as an alternative; it is not assumed to have demand either.
Illustrative excerpt · Selected parts of the output
Service delivery materials · The first operational review
Material AI prepares
What it makes concrete
Scope and buyer intake
Which restaurant and operating period are covered, what records the owner supplies, and which questions the review will answer.
Delivery procedure
Review the supplied purchasing, waste, scheduling, and sales records; identify inconsistencies; separate supported findings from items needing observation.
Report and acceptance checklist
List each finding, its source, its practical effect, and the next check or change. Define what the completed review includes before payment.
Capacity and customer communication
Reserve six delivery hours per review as an initial estimate. Track actual hours, agree scheduling, and prepare the confirmation and follow-up messages.
AI prepares the service materials and supported analysis. You carry out the restaurant-specific service, obtain the records, and perform any observation or professional work the scope requires.
Illustrative output excerpt
A service intake and completion check
Intake: “Which restaurant and operating period should this review cover? Please provide the purchasing, waste, staff scheduling and sales records for that period, and identify any missing records.”
Completion check: “For each finding, identify the source record, the practical effect and the next check or change. Separate items supported by the records from items requiring an on-site observation. List unresolved questions and deliver the agreed report within the confirmed scope.”
02One booking requires delivery and another look at the offer.
After independent review and the necessary repairs, the illustrative plan tests a $900 service with 30 qualified owners over 30 days. The prewritten plan treats three or more net paid bookings as support for a limited next period, one or two as a reason to change one thing and test again, and zero as a reason to examine the alternative, provided the test was valid. The budget and delivery slots are recorded before invitations go out.
Illustrative excerpt · Selected parts of the output
Service test · One paid booking
One qualified owner books and pays. If the test met its validity requirements, the result falls into the prewritten one-or-two-booking path. It does not prove that the price is wrong, that restaurants do not need the work, or that the spreadsheet alternative will sell.
Next work: fulfill the booked review, record actual hours and the owner’s experience, and examine the outreach and nonbuyer responses. Select one evidence-based change for a new test within the remaining budget. Do not rewrite the first test’s rules.
Orders rise from 100 to 150, but contribution after the listed costs and owner labor allowance falls from $2,500 to $800. Follow how AI checks the economics, identifies a specific defect and prepares a measurable repair.
01More orders leave less contribution after costs and owner time.
You already sell a $100 digital product with customer support. Orders increased from 100 to 150 per comparable monthly period, but you are busier and have less contribution left. You provide order and refund records, advertising spend, direct costs, support logs, and the operating record from the previous review.
Illustrative excerpt · Selected parts of the output
Operating review · More orders, lower contribution
Comparable monthly measure
Earlier period
Later period
Gross orders
100
150
Gross sales
$10,000
$15,000
Full refunds
5 · $500
15 · $1,500
Net sales after refunds
$9,500
$13,500
Acquisition spending
$2,000
$4,000
Other direct costs
$1,000
$1,500
Delivery and support hours
100
180
Owner labor allowance at $40/hour
$4,000
$7,200
Contribution after listed costs and labor allowance
$2,500
$800
This is a management comparison using the stated $40 hourly allowance. It is not accounting profit after every expense; fixed overhead, taxes, and any omitted costs still need separate treatment.
02Repair the missing instruction and measure the effect.
The support logs in this illustration repeatedly show buyers stuck at the import step. AI checks the instructions and finds that the required file format is not named. That supports a specific repair. It does not explain the separate increase in advertising cost.
Illustrative output excerpt
Replace the missing instruction with something the buyer can follow
Illustrative added detail: for this repair, assume the verified importer accepts a CSV file with the headers shown in its supplied example.
Revised instruction: “Save your data as a CSV (.csv) file. Use the exact column headings in the included example. Confirm the reporting period, keep one order per row and check the preview against your file before completing the import.”
Verification: test the instruction with a representative file and reconcile the imported order count and sales total to the source. In the next selling period, track import-related support contacts and support time per new order; an improvement is not assumed.
Illustrative excerpt · Selected parts of the output
Business improvement · Work prepared for the next period
Product and onboarding repair: replace the ambiguous import instruction with a named file format, a completed example, and a check buyers can perform before continuing.
Customer care: prepare the corrected instructions and a message for affected buyers. Fulfill outstanding support and refund obligations.
Measurement: record support contacts about the import step and time spent per new order, while keeping refund reasons and acquisition sources separate.
Spending: hold off on increasing acquisition spend until the repaired delivery experience and channel economics have been reviewed.
Next decision: IMPROVE. At the next review, use actual costs, hours, refunds, and buyer outcomes to decide whether continued selling or expansion is justified.
The result is a changed product, customer instructions, and a measurable operating plan. AI can prepare campaign revisions as well, but it does not infer the cause of rising acquisition costs from a support defect or promise to monitor the business without a configured and tested connection.
Each module works independently inside the AI account you already use. Choose one for your immediate goal, or all three for the full Sapien Amplified suite.
AI Decision Alpha
Examine important business and personal decisions before you commit.
Sapien Amplified adds no recurring subscription. Your AI provider and any optional tools have their own costs. Read the full product descriptions and terms before purchasing.