You have run vendor relationships, negotiated contracts, managed properties and staff and advisors, and caught things in invoices that other people walked straight past. You are good at this.
Then an aircraft entered the portfolio. The vocabulary went sideways, the numbers got large, and every single person in the conversation knows more than you do about the actual object. So the quotes get approved, because the alternative is asking a question that might reveal you cannot evaluate the answer.
Here is the reframe worth starting from. Overseeing a maintenance program is not an engineering problem. It is a vendor governance problem wearing unfamiliar clothes, and vendor governance is your discipline, not theirs. You are not going to out-technical your director of maintenance and you should not try. You are going to do the thing you already do well, which is insist that decisions leave a paper trail.
The structure in most flight departments looks like this, and it is nobody's fault.
The principal trusts the flight department, usually correctly. The flight department holds essentially all of the information: what work is due, which facility should perform it, what the quote means, whether the price is fair, and whether the invoice matches what was agreed. You approve payment without the ability to evaluate what you are approving.
Nobody checks the middle. Not through negligence, but because the person positioned to check lacks the vocabulary, and the person with the vocabulary is the one being checked.
It is worth being precise about why that matters, and the general research on how organizations lose money is unusually clear on the point. The Association of Certified Fraud Examiners publishes a global study every two years. Its 2026 edition analyzed more than 2,400 real cases across 143 countries, with a median loss of $104,000 per case and an average exceeding $1.4 million. More than half of all cases involved either a lack of internal controls or an override of the controls that existed. Losses scaled sharply with authority: median losses caused by owners and executives ran more than nine times greater than those caused by rank and file employees. Earlier editions found that duration is the real multiplier, with schemes caught inside six months producing a median loss around $30,000 against roughly $250,000 for those that ran two to three years.
None of that is a statement about aviation people, and it is emphatically not a statement about your director of maintenance. It is a statement about structures. The finding that losses scale with authority and tenure is not a claim that long serving senior people are dishonest. It is the arithmetic observation that the roles with the most access are the roles where a problem, in the rare event one occurs, grows largest before anyone notices. That is an argument about control design, not about character.
And there is a second reason to build controls that has nothing whatsoever to do with suspicion.
In a system with no checking, an excellent DOM gets no credit, because there is no evidence.
The absence of controls does not protect a good employee. It denies them a record.
None of these requires technical knowledge. All of them require consistency, which is the part you are already good at.
Not for big jobs. For every job, because the exceptions are exactly where drift accumulates, and because a blanket rule spares you from having to make a judgment call you are not equipped to make.
This is not an exotic request and it is not adversarial. AOPA's guidance to maintenance providers tells shops the same thing from their side of the counter: produce an accurate written quote and secure appropriate work authorization before proceeding. You are asking a professional to do what their own trade association already advises.
"I need something in writing I can put in the file for the accountant."
Choose a number. It matters far less which number you choose than that one exists and is written down before the next event rather than during it.
For calibration, most of the economy already works this way. Universities routinely require documented competitive pricing above thresholds commonly set around $25,000. Organizations spending federal grant money work to a micro purchase threshold of $10,000, above which some demonstration of competition is expected. Those are institutions buying office furniture and laboratory equipment. Aircraft maintenance is frequently the largest discretionary spend in the entire family office and often has no threshold at all.
The threshold is what removes the personal element from the conversation. You are not questioning this quote or this person. The policy questions every quote over a number, automatically, forever.
One honest caveat, because a rigid rule applied without judgment will damage your credibility. There are legitimate reasons a second bid is not available: an aircraft on ground with no time to shop, warranty work that must go to an authorized facility, or a type with genuinely only one credible shop in range. Those are real. What you are listening for is whether the objection is specific or general. "This has to go to the OEM because it is covered under warranty" is a reason. "That is not really how this works" is not a reason, it is a deflection, and the difference is audible even to someone who knows nothing about airplanes.
"Anything over fifteen thousand needs two quotes. It is the same rule we use for everything else now."
You do not need to understand what the part does. You need the file to be complete.
Ask for three things on any significant component: the part number, the condition it was supplied in, which will be described as new, overhauled, or serviceable, and the release document, which will typically be an FAA Form 8130-3 or its European equivalent.
The reason this belongs to you rather than to the technical staff is that it is asset documentation, which is squarely your job. An incomplete parts file does not cause a problem today. It causes one during a pre-buy inspection, when a prospective buyer's technical representative pulls the records and finds a component nobody can substantiate. That comes out of the sale price or out of the closing timeline, and it does so years after everyone involved has forgotten the event.
It is also, formally, your side of the table. The FAA has been consistent that establishing the eligibility of a part for installation rests with the owner or operator, not with whoever sold it.
"Can you send me the part number, the condition, and the tag for the file?"
Not line by line. You will drown and you will learn nothing. Category by category, looking only at what moved.
An invoice read on its own always looks like a list of reasonable things, because it is a list of reasonable things. Read against the quote, the same document becomes a variance report, and variance is where all of the information lives.
Then ask about any category that moved more than some percentage you decide in advance. Here is the part that makes this workable for a non technical reader: you do not need to evaluate the answer on its merits. You need to observe three things about it. Does an answer exist. Did it arrive promptly. Is it specific. A specific answer you do not fully understand is a good sign. A general answer you understand perfectly is not.
"Parts came in about forty percent over the estimate. Can you walk me through what drove that?"
The least glamorous habit here and the one with the longest payoff.
Track each event by category: scheduled inspection, discrepancies and findings, parts, outside services, consumables, freight, and days of downtime. A simple spreadsheet is entirely sufficient. Nothing about this needs to be sophisticated.
No single event tells you anything useful. Twelve events tell you almost everything. Markup drift, the ratio of findings to base scope at a particular shop, whether estimates trend high or low, and whether downtime is improving or degrading are all patterns, and patterns are invisible in a sample of one.
Two years of this quietly makes you the most informed person in any conversation about the aircraft, and you will have accomplished it without ever learning what a Phase 5 inspection actually involves.
The control questions change shape when a management company is handling procurement on the principal's behalf, because a second set of commercial interests has entered the transaction. That is not sinister, it is simply structure, and the questions are well established among aviation counsel.
Ask for a sample owner statement from a current client. A transparent operator will produce one without drama, and the request itself is entirely normal during diligence.
None of the signals on the right hand side proves anything by itself. Patterns across several of them are worth paying attention to.
| Healthy | Worth a closer look |
|---|---|
| Written quotes arrive before work starts, without being chased | Quotes arrive verbally, or arrive after the work is already done |
| A second bid is treated as normal procedure | A second bid is treated as a personal comment |
| Invoice variances get explained specifically and quickly | Explanations are general, slow, or faintly offended |
| Part numbers, conditions, and release tags are supplied on request | Requests for documentation produce friction or delay |
| Your DOM proposes controls before you have to ask for them | Every control in the program originated with you |
| Bad news reaches you early, while it is still small | Bad news reaches you attached to an invoice |
The last row is the single most reliable indicator in the entire business. A department that brings you problems early, while they are still small and still fixable and still slightly embarrassing, is a department operating in good faith. A department where bad news consistently arrives already attached to an invoice has told you something about how it views your role, and it is not a flattering assessment.
This is the part that actually stops people, and it is rarely written down anywhere. Your real concern is not being wrong about a number. It is becoming the person who insulted a long serving professional the principal likes and depends on.
Five things make that risk manageable.
The structure described in this article is the structure VHMX runs by default: a written scope before work starts, competitive bids as the ordinary path rather than a special event, documented awards, and a reconciliation at the end.
The reason that matters specifically to someone in your seat is the confrontation problem. Saying "we run maintenance procurement through a platform now" is not an accusation against anybody. It is a policy, and policies are not personal. Nobody has been doubted, nobody has been asked to justify themselves, and the awkward conversation you were dreading simply does not need to occur.
In our experience the flight departments doing genuinely good work are the ones most relieved by it, because it converts their judgment into a documented record that survives staff changes, ownership changes, and the eventual sale of the aircraft.
VHMX is not a repair station and does not perform maintenance. We do not overrule your director of maintenance and we do not decide what work your aircraft needs. We build the paper trail that lets a non technical person exercise real oversight of a technical program.
Sources and further reading: Association of Certified Fraud Examiners, Occupational Fraud 2026: A Report to the Nations, and prior editions. AOPA Pilot Protection Services, "Aircraft Maintenance: Tips for Aircraft Owners and Maintenance Providers." Vedder Price, "Selecting the Best Aircraft Management Company," on transparency and audit rights. Business Jet Traveler, "What You Need to Know about Aircraft Management Agreements," on third party markups and rebate pass through. FAA Advisory Circular 20-62E on parts eligibility. 2 CFR 200 micro purchase threshold.
VHMX puts a written scope, competitive bids, and a documented record behind every maintenance event, so the paper trail exists before anyone has to ask for it.