For searchers, independent sponsors, and first-time acquirers

The earnings are real.
Are the systems?

Operational and data diligence for people who will run the business, not just own it. A document review reads the data room. This reads the systems themselves — with read-only or API access to what the business actually runs on — and reports what you are inheriting, inside the same window as your Quality of Earnings.

Send me the deal →
Who it's for

You're the one who walks in on day one.

Self-funded searchers, independent sponsors, first-time acquirers, and small funds buying a business in roughly the $3M to $30M purchase-price range that they intend to operate.

The common factor is not size. It is that on day one you walk in and run it. You do not have a bench of analysts to absorb a surprise. That changes which unknowns matter.

You have signed an LOI. The data room is open. A QoE team is in it. You have somewhere between two and six weeks of exclusivity, and every day you spend deciding what to investigate is a day you are not investigating. This report is bought in that window and delivered inside it.

Nobody is lying to you. That's the problem.

Every business you look at will present a set of numbers. The QoE will confirm those numbers reconcile to the bank. Neither the seller nor the QoE will tell you that the monthly revenue figure is assembled by the owner's sister-in-law from three exports and a memorised pricing rule, that the scheduling system was abandoned in 2023 and the real schedule lives in a shared calendar, or that the largest customer relationship is a handshake with a man who is retiring at close.

None of that is fraud. None of it shows up in accounting diligence. All of it is yours on the first of the month after close.

The cost is rarely the systems themselves. It is the six months you spend rebuilding basic visibility while the business is also, simultaneously, your responsibility for the first time.

Sometimes you shouldn't bother.

Doing nothing is a defensible choice when the business genuinely runs on a mainstream, well-documented platform you have personally operated before; the seller is staying on for a real transition of twelve months or more; the purchase price leaves enough headroom that six months of operational cleanup does not threaten the debt service; or you are buying primarily for an asset — a licence, a contract, a location — rather than an operation.

Doing nothing is a bad bet when the numbers arrive as spreadsheets rather than system exports; the seller is the operator and is leaving; the business has made acquisitions of its own and never integrated them; the deal is levered tightly enough that a reporting failure in month three becomes a covenant problem in month six; or you are counting on data you have not yet seen produced.

The deliverable

The Operator’s Inheritance Report

A written report in plain English — not a deck, not a dashboard, not a verbal debrief. Typically 12 to 20 pages depending on the complexity of the business. Delivered as a PDF, with a 45-minute walkthrough call. Six sections:

01

Source-of-truth trace.

For each reported number that matters to your model: the system, export, or human that produces it, and whether it survives the sale.

02

Owner-dependency map.

The knowledge, relationships, logins, credentials, pricing judgement, and undocumented process that exist only in the seller's head, and specifically what stops working when they leave.

03

Systems and technical-debt inventory.

Every system the business actually runs on, its contract and renewal position, its integration state, and what you are inheriting versus what you will have to replace.

04

Reporting readiness.

Whether you can produce a defensible monthly close and operating report on day one, day thirty, or not until something is rebuilt.

05

Post-close work estimate, ranked and costed.

The operational and systems cleanup you are buying, in priority order, with rough effort, cost range, and sequence.

06

Deal-relevant findings.

The items that should change your price, your structure, your transition services agreement, your earnout, or your first-100-days plan, each stated as a finding, its evidence, and its implication.

Runs alongside your QoE, on the same clock

Five to ten business days from system access to delivered report. Five for a single-entity business on mainstream systems; ten for a multi-entity or previously-acquisitive business. If your window is shorter than that, say so — a scoped-down read covering sections 1, 2, and 6 only can be turned around in three business days.

The clock starts at system access, not data room access. The data room is where the documents are; the systems are where the answers are. Getting read-only or API access agreed with the seller is usually the longest lead item, so it is worth asking for it the day the LOI is signed.

Process

What the ten days look like

What I need from you

  • Data room access, or the equivalent documents
  • The CIM and your own operating model or working assumptions
  • A list of systems the business uses, however incomplete
  • Two 45-minute conversations with the people who produce the reporting
  • Your diligence deadline, stated as a date
The part that makes it different

This reads the systems, not just the data room

Anyone can read a data room. The findings that matter — that the revenue figure is assembled by hand, that two entities define a completed job differently, that the scheduling system was quietly abandoned — are only visible inside the systems themselves. So the engagement needs access to them:

  • Read-only logins to the core operating systems — the ERP, field-service or scheduling platform, CRM, billing, and payroll.
  • API keys or a read-only service account where the platform supports one, so the underlying records can be queried rather than eyeballed.
  • Where a platform allows neither, a recorded screen-share with whoever administers it, plus raw exports.

Read-only throughout. Nothing is written, changed, or exported beyond what the engagement needs, and access is revoked at delivery. If the seller will only grant some of this, the scope and the fee come down accordingly and the report says plainly what could not be verified.

What this is not

This engagement does not include, and is not a substitute for:

  • A Quality of Earnings report or any accounting diligence. Run one. This sits beside it.
  • Legal diligence of any kind.
  • Tax diligence or structuring advice.
  • Cybersecurity assessment or penetration testing.
  • A valuation, a fairness opinion, or a recommendation on whether to do the deal. The report describes what is there; the decision is yours.
  • Post-close integration or implementation. That is separate work, quoted separately, and you are under no obligation to buy it.
  • Any representation about the accuracy of what the seller has provided beyond what can be observed in the systems and documents made available.

After close, if the reporting turns out to be as broken as the diligence suggested, that is the work Aptum AI does next. It is quoted separately and you are under no obligation to buy it.

What you receive

The PDF report. The 45-minute walkthrough. The ranked post-close work estimate as a separate working file you can drop into your model. A one-page summary of deal-relevant findings you can send to your lender, your investors, or your QoE team without forwarding the whole report. All of it is yours, whether or not you close, and whether or not you ever work with Aptum AI again.

Risk reversal

Two ways you don't pay.

Delivered on your date, or you don't pay. We agree the delivery date on the scoping call. Miss it and the fee is refunded in full.

A finding you act on, or you don't pay. If the report contains nothing you take to your seller, your lender, or your first-100-days plan, tell me within seven days of the walkthrough and the fee is refunded in full. You keep the report.

Whether a finding is worth acting on is your call, not mine. I am not going to argue with a buyer about whether their own diligence was useful.

Proof

Who reads your deal

Nick Graham founded Aptum AI. Twelve years in the U.S. Army, ten as a Green Beret. BA Mathematics, MS Finance, Stanford Ignite. Former VP of Operations at two venture-backed cybersecurity startups, and Senior Data Scientist on the Enterprise AI team at a publicly traded enterprise SaaS company. Selected to the inaugural cohort of Palantir's American Tech Fellowship for Veterans, 2026.

What Aptum AI builds. Data warehouses, automated data pipelines, workflow automation, and custom data-driven software. Currently in production: a warehouse, pipelines, and pricing-engine software for a PE-backed operating company in home services — Python, Google Cloud, BigQuery, dbt, n8n, Airtable. The systems this report evaluates are the systems Aptum AI builds.

Prior portfolio work. Subcontracted through a partner agency: built a portfolio-wide operational KPI capture system for a private-equity owner of defense-focused IT companies.

What the report looks like. A redacted excerpt showing the structure and the finding format is available on request. Ask on the scoping call, or request it here.

How he thinks about this. Eleven issues of Dead Reckoning, including Buy Second, Integrate First and Legibility Is the Prerequisite.

Questions

Before you send anything

  • How is this different from my QoE?

    Your QoE answers “are the earnings real.” This answers “what produces them, and what happens to it at close.” Different question, different evidence, different deliverable. The two do not overlap and they do not compete for the same budget line.

  • Will this slow my deal down?

    No. It runs off the data room your QoE team is already in, on a parallel track, on a schedule you set.

  • What does it cost?

    A fixed fee agreed on the scoping call, based on entity count and systems complexity. You will have the number before you commit to anything.

  • What if the seller won't give system access?

    It happens, and it changes the engagement. Without read-only or API access this becomes a document review, which is a materially weaker product — so we scope it down, price it down, and label plainly in the report what could not be verified. A seller's refusal to show the systems is also itself a finding, and usually an important one.

  • Can my lender or investors see it?

    Yes. The one-page summary of deal-relevant findings exists for exactly that.

  • Who actually does the work?

    The engagement is delivered by the same people who build these systems, senior end to end. No offshore analyst, no handoff to someone you have not met.

  • Do I have to buy the post-close build?

    No, and there is no discount for bundling. The report is written to be useful to a buyer who never speaks to Aptum AI again.

  • What if I walk away from the deal?

    Then the report did its job. You keep it either way.

The CIM, or a paragraph. Where you are in the process, and your deadline. Within one business day you'll get a straight answer on whether an operational read is worth doing on this deal, what it would cover, and what it would cost.

Send me the deal.

Send me the deal →