Who This Checklist Is For (And When I Use It)

I'm a procurement manager at a mid-sized packaging operation. I've owned our capital equipment and consumables budget — roughly $180,000 a year across filling, capping and labeling — for about six years. In that time I've negotiated with more than 40 vendors and logged every order in our cost tracking system.

This checklist is what I actually run through before I issue a purchase order for capping equipment: bottle cappers, semi-automatic units, automatic capping machines, inline capping machines, rotary capping machines. It assumes you already have a product and a cap. It does not assume you know how to write the spec — that's step one.

Nine steps total. The first six decide whether the machine fits your line. The last three decide whether the deal stays cheap after the invoice is paid.

Step 1: Send physical samples, not drawings

In my first year buying equipment, I made the classic specification error: I sent a PDF drawing of our bottle neck finish and assumed "28mm" meant the same thing to every vendor. It cost us a $600 set of change parts that fit nothing, plus three weeks of back-and-forth.

What I mean is that cap and bottle geometry — thread pitch, neck diameter tolerance, cap skirt height, torque requirements — is not a place where drawings travel well. Couriers are cheaper than change parts. Ship 20 filled bottles and 50 caps with the caps you'll actually run, and ask the vendor to run them on video.

Checkpoint: the vendor has confirmed in writing that the machine will be quoted against your physical samples, and that you'll see those samples running before shipment.

Step 2: Pick the machine class by the outcome, not the price

This is where most first-time buyers go wrong, myself included. My initial approach was to compare quotes at the same headcount and call it a day. Then I watched a semi-automatic capping machine get overwhelmed at 40 bottles per minute with two people feeding it, and understood that labor was the real variable.

Quick working guide, assuming your output target is stable:

  • Semi-automatic capping machine — low volume, frequent changeovers, one or two operators. Cheapest to buy, most expensive per bottle in labor.
  • Automatic capping machine — a defined output band, fewer changeovers, one operator watching rather than feeding. This is where most B2B buyers land.
  • Inline capping machine — the capper is a station in a continuous line, so it inherits every upstream problem. Line integration cost is the number people forget to quote.
  • Rotary capping machine — high, steady volume with consistent container size. Fast, and the least tolerant of your changeover habits.

Speed, changeover time, floor space. Rank those three with your operations lead before you ask for prices, because your ranking eliminates half the market for free.

Step 3: Get a guaranteed delivery date, in writing, with a number attached

Look, I used to treat rush fees as vendors gouging customers. Then I sat in a planning meeting where a missed first-production date meant pushing a retail launch, and the math stopped being abstract.

In March 2024 we paid a premium for a guaranteed delivery window on a rotary capping machine instead of accepting "approximately 14 weeks" at a lower price. The alternative was missing a $15,000 committed order plus the customer conversation nobody wants to have. That premium bought certainty, not speed — the machine may have arrived on the same week either way. What it removed was the coin flip.

Checkpoint: the quote states a shipment date, the definition of "on time," and the remedy if it slips. A date with no remedy attached is a hope, not a commitment.

Step 4: Build the TCO sheet before you compare prices — especially the change parts

Back in Q1 2023 I compared six vendors for a line upgrade. Vendor A quoted $18,400. Vendor B quoted $15,900. I nearly went with B until I built the total picture: B charged separately for change parts ($1,150 per format), installation ($850), operator training ($400), and a spare-parts kit that A included. Total: $18,300 for B against $18,400 for A — and B's formats were two weeks out every time we switched sizes.

So the TCO row items I now force onto one page:

  1. Base machine price, and what "base" excludes
  2. Change parts per SKU you actually run (not per SKU theoretically possible)
  3. Installation, commissioning and operator training
  4. Recommended spare parts for year one, priced individually
  5. Freight, rigging, and duties landed at your dock
  6. Estimated labor hours per shift at your target output

The lowest quoted price is rarely the lowest total cost. It's just the number that shows up first.

Step 5: Ask for video of your container, not the catalog video

Catalog footage is a highlight reel of the vendor's best-behaved bottle. It tells you nothing about your bottle — your oval flask, your small-diameter cap, your slightly out-of-tolerance neck finish. Request a run of your samples on the quoted model, filmed at your target speed, and specifically ask them to keep the camera rolling through a changeover.

Changeover footage is the most honest test I've found. It shows the tooling, the tool count, and whether the operator looks like they've done it a hundred times or are reading a manual on camera.

Step 6: Get the year-one spare parts list with prices — before signing

This is the step most buyers skip, and it's the one that bites in month nine. Ask for the recommended spare parts list, unit prices, and lead times, then check whether any of those parts are proprietary. A $40 gripper with a six-week lead time is functionally a $40 gripper plus six weeks of downtime.

Three things I want on paper: consumable wear items and replacement intervals, lead time on the three parts most likely to fail, and whether the control system or torque module can be serviced locally.

Step 7: Agree the acceptance test and put payment behind it

Define what "working" means before the machine ships, and tie money to it. I use a factory acceptance test against a written output and torque spec on your samples, then a site acceptance test at your line speed after commissioning. Payment terms: deposit, balance on factory test sign-off, final retention released after site acceptance.

If the vendor resists milestone payment on a machine of this size, that's information about how they handle problems after the invoice clears.

Step 8: Confirm electrical, safety and documentation fit

Voltage, phase, control voltage, guarding, e-stop logic, and the certification documentation your site requires — all of it should be confirmed in writing, not assumed from a spec sheet photo. Ask for the documentation package list: electrical drawings, spare parts manual, operating manual, and the compliance declarations that apply to your region.

Step 9: Write the ramp-up support into the contract

Installation day is the vendor's easy day. Week three is when output sags and nobody knows why. Get remote support terms, response time expectations, and a defined commissioning window written in. This is cheap to ask for before the order and expensive to improvise after.

Common Mistakes I've Watched People Make (Including Me)

  • Quoting one machine against another at different scopes. Half of my early "savings" were scope differences, not price differences.
  • Treating the quoted speed as your output. Quoted speed is a ceiling under ideal conditions. Ask what sustained output looks like on your container.
  • Buying the cheapest change parts. They're the parts that touch your product. The failure shows up as torque variation, not as a broken part.
  • Ignoring the changeover count. A machine that's fast per bottle but slow per format change is the wrong machine if you run 12 SKUs.
  • Accepting "approximately" on a delivery date. If the date matters to your production plan, buy the certainty and budget for it — the cost of a slipped date is almost never on the vendor's side of the ledger.

One honest caveat: the cost figures above are from my own tracking system for our operation, and your numbers will differ by volume, region and container mix. Build the sheet with your own line data before you compare anything — that's the whole point of steps four and five.