Why Aviation AI Quoting Depends on Reliable Inventory Data



Everybody Wants You to Quote Faster. Quote What?


TL;DR

Faster quoting needs reliable data. AI earns its place when the inventory, paperwork, and commitments behind the answer hold up.


Fine. Bring the RFQ.

Everybody wants to fix your quoting process.

Everybody.

Spend twenty minutes on LinkedIn and you’ll find enough companies promising more RFQs, faster answers, and agentic everything to make you wonder how aviation survived long enough for them to show up.

I’ve spent about twenty years in aviation, aerospace, and defense. I understand the appetite. Your salespeople are buried. Your buyers are chasing responses. Somebody’s looking at a hundred-line attachment, wondering when they’re supposed to find time to actually sell.

So let me put our cards on the table.

​ERP.Aero has iRFQ. It reads supported inbound RFQs and vendor responses, turns information into records, and connects the supporting attachments. Bring the spreadsheet with 256 line items. Let’s process it. Your salesperson shouldn’t have to type the customer’s request before they can start answering it.

We have ELIA, our decision engine embedded in the ERP workflow, using your company’s records and history to help your people decide what comes next.

I’ll show you both.

Then I want to keep going, because the impressive part of the demonstration shouldn’t end when an RFQ appears on a screen.

What inventory is that answer using? Which vendor response? What condition? Where’s the trace? What changed after somebody last touched the record?

I want AI taking unnecessary work off your people. Absolutely.

But I’m not going to celebrate how quickly we helped you promise a unit you can’t deliver.

That’s where this conversation needs to go.


The Intelligence Needs Something Worth Knowing

​ELIA doesn’t need to reinvent the wheel. The wheel is round for a reason. I want intelligence that understands why it needs to turn in a particular direction at that moment.

Your salesperson needs the relevant pricing history. Your buyer needs to know which supplier already responded. Your repair coordinator needs to see whether the material was ordered, received, or stopped in inspection.

Those answers should build on work your company already did.

That’s why iRFQ and ELIA matter together. Capture the information, connect it to the transaction, and make it useful for the next decision. Vendor quotations, supporting photos, certs, customer requirements, and sourcing history shouldn’t become another pile somebody has to sort through later.

Supported inputs still need review when they’re incomplete or uncertain. Attaching an 8130 doesn’t establish that it satisfies the customer’s requirement. Your people need the information and the ability to act responsibly on it.

And when those people leave, the records behind their decisions should remain. You shouldn’t have to rebuild years of sourcing knowledge because somebody accepted another job.

That’s meaningful AI.

But look underneath it.

If inventory says available, sales knows committed, and the marketplace is showing last Thursday’s spreadsheet, which answer should the intelligence trust?

You’ve given it three versions of the same business.

I don’t care how much agentic you put on top of that. Somebody still has to establish what’s true in the supply chain or the repair process. Automating the answer before resolving the disagreement just gives the mistake a head start.

Let me show you how that happens.


One Real Part. Seven Versions of Available.

Imagine a serviceable unit sitting on your shelf Monday morning.

It’s yours. Correct part number. Serial number recorded. Paperwork available. You can sell it.

Then the daily ritual starts.

Three people maintain seven spreadsheets. Two people export information and reconcile it against sales and inventory. Somebody updates the master file, combines the catalogs, adjusts the columns, and uploads the result to a marketplace.

The file passes. The listing appears.

Meanwhile, your salesperson commits that unit to another customer.

The spreadsheet still says one available.

Nobody in that chain woke up planning to mislead a buyer. They’re doing their jobs. The problem is that keeping the listing accurate requires another job after the actual job, and somebody has to remember to do it.

One aviation business owner described the result:

“We have so many errors because of all the manual back and forth bs.”

There it is. People doing work, then doing more work to tell another system about the work.

We complain about ghost inventory. Rightfully so. Listing every part you might find and every capability you might arrange just to collect RFQs should’ve stopped yesterday. I’ll lead that fight if I have to.

But honest businesses can publish unreliable listings too.

And telling them to “clean the data” doesn’t answer the operational question.

Clean it when? Before which export? How long does it stay clean after somebody sells something?

Our unit existed. The paperwork existed. The listing was accurate.

Then a perfectly legitimate sales transaction turned that listing into a bad answer.

Nothing about the upload template could prevent that.


Your Bad Answer Just Became My Commitment

Follow the unit.

A buyer sees your listing. They’re sourcing material for a repair shop. The work order is open, the bench needs the part, and the customer’s already asking when the unit will be ready.

The buyer sends an RFQ.

Your salesperson checks inventory and discovers the commitment. Maybe they reply immediately. Maybe they start hunting for another unit because they still want the order.

Legitimate brokerage has a place. Represent what you’re offering honestly and go earn the business.

But the buyer doesn’t know which listings represent stock, which represent sourcing possibilities, and which are simply out of date. What looks like a hundred supply options can turn into another round of people asking other people whether anybody actually has the part.

One buyer described their experience:

“I send 100 FQs. barely anyone replied, maybe only 20% of them.”

Those are their numbers. Picture the work behind them.

The buyer follows up. The repair coordinator calls purchasing. Purchasing says they’re waiting. The customer calls again. Somebody pencils in a completion date because the listing looked promising and the salesperson sounded confident.

Then the source falls through.

Quote what faster?

An unconfirmed lead? A unit already allocated elsewhere? A price without acceptable trace?

The repair shop now has to unwind a commitment built on information it never controlled. The technician still needs material. The coordinator has to explain the delay. Your stale listing has reached somebody else’s customer relationship.

That’s what gets me.

The person paying for bad data is often several handoffs away from the person who could’ve corrected it.

And somebody’s still selling that buyer more RFQs.


The Part Moved. Why Didn’t the Information?

Let’s go back to Monday morning. Your salesperson commits that unit.

Why does somebody else have to discover that later, change a spreadsheet, and tell the marketplace?

The business already knows. You just made the commitment inside it. But now one considers that you have competing sync schedules, and some have yet to happen.

That’s the work I want gone.

At ERP.Aero, we distinguish committed stock from available stock. Your next salesperson needs to see what they can actually offer before they promise the same unit to somebody else.

The same goes for consignment. Somebody else owns that material. You’re managing it under an agreement. Change the consignment, and your people need that change where they work. Close a supplier catalog, and those entries need to stop influencing what your team thinks it can offer.

A catalog, a price list, and a serviceable unit in your bin are three different things. Why would you dump them into one bucket and make your salesperson sort out the difference during an RFQ?

Then there’s the paperwork.

One operator described where incoming documents went:

“Right now, when they come in, we scan them, we save them all on a shared drive on everyone's computer, and they pull them from there.”

So you have the cert. Wonderful. Can the person quoting the unit find the right one? Can receiving see what was promised? Or are we calling the person who knows which folder it’s in?

We keep those documents connected to the records because the next person shouldn’t have to start another investigation.

Our supported marketplace API connections let us publish and synchronize inventory, and exchange RFQs where supported. Show me the allocation. Show me what the connection updates. If it fails, show me who knows.

That’s how you judge whether the connection does its job.

And if you’re using files, changing one consignment shouldn’t mean rebuilding every catalog and stock list around it.

Your people should be running the business. Keeping its listings accurate should follow the work they’ve already done.


Bring the Part. Bring the Paperwork. Let’s Run It.

Go back to our buyer.

They finally find another unit. It reaches the repair shop. Receiving checks the part and paperwork against the order. The job moves forward when the requirements are satisfied.

The original listing may still be online.

Tomorrow, another buyer can start the same miserable journey.

That’s why I want a harder conversation when somebody promises to consolidate your business into their interface.

You’ve consolidated my screens. Have you consolidated the truth?

Which record wins when the systems disagree? Who fixes it? Does the correction reach the next person, or have I bought another place my employees have to maintain?

Fifteen disconnected services feeding another interface can still leave fifteen disagreements underneath it. Eventually, that middleware wants to become the place you run everything from, while your people keep feeding it copies.

I want the actual operation to produce information people can rely on.

That’s why ERP.Aero connects inventory, sourcing, documents, repair activity, and the history of what happened. A good record should save work the next time somebody touches it.

And that’s how I want you to evaluate us.

Bring an RFQ with a difficult attachment. Bring the vendor response, the consignment, the allocation conflict, and the paperwork question. Follow the transaction through receiving. Change something halfway through and watch what the next person sees.

If you’re considering a six-figure investment in an agentic quoting screen, demand that same demonstration. Ask where the operational truth lives, who maintains it, and what happens when the source changes after the quote goes out.

I’ll put our workflow against anybody’s. Bring your own work and let’s run it.

Because I care whether your salesperson can stand behind the answer. Whether your buyer can trust the commitment. Whether your repair team can finish the job without chasing information your business already had.

I want you in business today, tomorrow, and next year.

Let me show you what that takes.


Thank You

Thank you for reading, and for caring about what happens after the quote goes out. I wrote this for the people who have to make that promise hold together, often while fixing something they didn’t break. If you recognized your day somewhere in here, I want to hear about it. Bring your own work, and let’s run it together.


2252 Hayes Street, Hollywood, FL 33020
​Unsubscribe · Preferences​

Ralph Merhi,

Practical insights for aviation suppliers, distributors, brokers, and manufacturers who refuse to settle for inefficiency. I believe in cutting through the noise—delivering real strategies to make things better. No fluff, no wasted time, just the knowledge, both business and personal, and tools to help you succeed. If you want my newsletter, drop your email below 👇 or feel free to look as much as you want.

Read more from Ralph Merhi,

Less id often more... Why the Best Enterprise AI Strategy Ends With Less AI The customer quotations below come from recorded aviation-industry conversations. They’ve been lightly cleaned for readability, and names and company identities are omitted. TL;DR The smartest AI strategy may leave you with less AI in production. Use agents to get a dirty process moving, observe their decisions, and turn repeated behavior into deterministic workflow. Keep AI where interpretation and judgment matter....

What We Found When We Followed One Revenue Workflow TL;DR Your next revenue opportunity may already be inside your ERP. The key is finding where disconnected habits, manual work, and underused capabilities are limiting the business. This article combines recurring patterns from real customer conversations. The quoted customer language is real, anonymized, and traceable to recorded calls.* Most companies look for additional revenue in familiar places. More leads. More salespeople. More...

How Much Did That $30 Savings Actually Cost They Chose the Cheaper ERP. Twenty-One Months Later, They Chose Again. Twenty-one months. That’s how long it took for our newest customer to come on board. I could tell that story like it was some brilliant long-term sales strategy, like we knew exactly what would happen and patiently waited for the market to prove us right. That wouldn’t be true. The truth is, when they picked somebody else, I didn’t understand it. They had told us what they...