The Cheapest Eye Cream Quote Is Rarely the Cheapest Eye Cream

On our last two eye cream sourcing cycles, the lowest quoted unit price lost. Both times.

I run procurement for a 240-person private-label skincare company. I've managed our eye care and treatment budget — roughly $380,000 annually — for six years, negotiated with 34 vendors, and logged every order in our cost tracking system. When we ran the 2024 cycle, the lowest bidder came in at $1.42 per unit on a 15 mL airless pump. Our incumbent was $1.79. That 21% gap didn't survive contact with the rest of the model. Once we priced in stability testing, fill variance, dead inventory, documentation, and freight, the two "cheap" quotes came out 19% and 27% higher than the incumbent on a fully loaded basis.

Cost per unit is a number suppliers choose. Fully loaded cost per saleable unit is a number you calculate. Most buyers never get past the first one, which is exactly why the low bid keeps winning deals it shouldn't.

How I actually build the number

I'm not going to pretend this is sophisticated. It's a spreadsheet with 11 rows and a column for every vendor we shortlist. But it took a failed launch to make me build it, and I've rebuilt it twice since.

In Q2 2024 we compared nine suppliers over about four months. Our procurement policy now requires a minimum of three quotes plus a documented TCO sheet before anything gets approved — I put that rule in after getting burned on a hidden fill-weight variance in 2022 that cost us roughly $1,900 in product we paid for and never sold.

The five rows that matter most are below. None of them appear on a standard quote sheet.

1. Stability and compatibility testing moves the real price

This is the row people skip, and it's the expensive one.

An eye cream for wrinkles built around peptides or encapsulated retinoids has to be tested against the elastomers and coatings inside an airless pump, not just in a jar. A brightening eye cream with a vitamin C derivative has an oxidation curve that depends heavily on package barrier properties. If the formula and the pack weren't tested together, you are the test.

Accelerated stability panels — 12 weeks at 40°C/75% RH, plus freeze-thaw — typically ran us $600 to $1,200 per formula-package combination (this was back in 2023; I'd budget higher now). Multiply that across three or four candidate formulas and you've added five figures to a program before a single unit ships.

Here's where I cost us money. I knew I should have requested the 12-week accelerated data on a brightening formula before approving the pilot run. I thought, "it's basically the same base we've used for two years, what are the odds?" The odds caught up with us at week nine, when the product had visibly discolored and failed our own visual QC — a $2,800 lot, plus a launch date we didn't hit. A lesson learned the hard way. Now the stability report is a gate item, not a nice-to-have.

2. Fill weight and packaging variance is where silent losses live

A 15 mL fill isn't 15 mL. Standard tolerance is often ±5%, and plenty of fillers run consistently to the short side of that band because nobody's checking. Run a 3% short fill across 20,000 units and you've handed over about 9 liters of product — roughly 600 tubes' worth — that you paid for and can't sell.

Packaging color is the other quiet cost. Industry convention for brand-critical colors is a Delta E under 2; a Delta E of 2–4 is noticeable to a trained eye, and above 4 is visible to most people under normal light (reference: Pantone color matching guidance). One supplier's "exact same color" caps came in at Delta E 4.1 against our incumbent. On a shelf, next to the rest of the line, it read as a different product. That's not a printing problem — that's a brand problem, and repressing the caps cost more than the unit savings on the whole order.

3. MOQ versus sell-through — the inventory you didn't budget for

I dodged a bullet on this one, and I want to be honest that it was luck as much as diligence.

A supplier quoted $1.28 per unit at 30,000 MOQ against $1.55 at 10,000. Straight unit-price math said the bigger order saved us $8,100. We were one signature away from it when I double-checked our own sell-through on a comparable eye cream for bags — a caffeine-based SKU that had moved about 4,200 units in its first year. We were about to buy roughly six years of inventory to save 27 cents a unit. Between you and me, that's not a saving, that's a warehouse rental.

The TCO line that catches it: carrying cost, write-off risk, and the cash you can't deploy elsewhere. Ours assumes 18% annual carrying cost plus a full write-off provision on anything with under a 40% chance of selling through in 24 months. Run your own numbers, but run them.

4. Documentation is a schedule cost, not a paperwork cost

An eye cream for sensitive skin lives or dies on documentation — fragrance-free substantiation, patch test data, INCI consistency across lots, and a product information file that actually exists rather than "we'll send it next week."

In the US, cosmetic facility registration and product listing requirements have been phasing in under MoCRA since it was signed in December 2022, and the compliance deadlines have moved more than once. Check current FDA guidance rather than trusting a blog post — including this one. The point isn't the specific date; the point is that a supplier who can't produce a complete PIF in ten working days will cost you a launch window, and a launch window costs more than any unit-price delta on this page.

5. Freight and lead time, including the buffer you'll forget

Sea freight on our lane ran 35–45 days door to door. Air freight on the same volume cost roughly 4x. Any supplier quoting a lead time with no buffer built in is quoting a fantasy — you need 20–30% on top of whatever they tell you, and you need to model the air-freight delta as a real line item, because at some point you will pay it.

Where an efficient process actually shows up

Here's the thing: most of the operational difference between suppliers isn't in their price, it's in their turnaround. The vendor that gave us a shared, version-controlled spec sheet and a COA portal cut our approval cycle from five days to two. The one that emailed PDFs back and forth with names like SPEC_final_v3_REALfinal.pdf took nine days on the same step, three times running.

That's not glamorous, but two days versus nine days, repeated across a development cycle, is a competitive difference. It compounds. It's also not universal — a genuinely custom formulation still benefits from slow, hands-on back-and-forth, and I'd push back on anyone who says digitizing everything automatically makes it better.

The surprise in our 2024 cycle wasn't the price spread. It was that a mid-tier supplier we'd shortlisted as a formality out-processed two premium vendors on documentation turnaround and COA completeness. Their unit price was 8% higher. They won the program anyway.

When the low quote is actually correct

I'm not going to argue that cheap is always expensive. Sometimes cheap is just cheap.

If you're sourcing a commodity hydrating eye cream with no novel actives, an established base, a supplier you've run for years, and a package you've already validated, the TCO model has very little to bite on. The variance risk is low because nothing is new. Same story for very small runs under about 2,000 units, where tooling and setup dominate the cost structure and a formal TCO comparison is more ceremony than analysis.

And one honest limitation: our numbers come from our category, our freight lanes, and our volumes. A buyer moving 200,000 units a year will see the fill-weight and MOQ lines collapse; a buyer moving 5,000 will see them dominate. The structure of the model travels. The percentages don't.

One last housekeeping note — every price in this piece reflects our quotes and test costs from 2023 through mid-2024. Verify current pricing and current regulatory deadlines before you build a budget on any of it. Rates move. The logic of the five rows doesn't.