How to compare two custom biology quotes

Two custom biology quotes are almost never comparable as written, and the difference between them is usually not price. It is scope, unit definition, and who absorbs a failed batch.

The method is mechanical: put both quotes on one unit, add back everything one includes and the other excludes, separate non-recurring cost from per-unit cost, then read the five contract terms that decide what the number actually buys. Do that and the cheaper quote frequently changes identity. This page is that method, in enough detail that you can run it yourself and never contact us.

Normalising quotes is the service we sell, so it is fair to say plainly: this page gives away the method. What it cannot give away is the supplier list and the willingness of three vendors to quote the same document on the same week.

Where this sits against the RFQ page

What belongs in a wetware RFQ covers the outbound document — the twenty-one fields that make two suppliers quote the same thing. This page covers the inbound side: what to do with the responses when they arrive, including the common case where you did not write a good RFQ and are now holding two documents that answer different questions.

If you have not sent the RFQ yet, read that page first. It prevents most of the problems below. If the quotes are already on your desk, start here.

The nine reasons two quotes are not comparable

1. The counting convention

The most common and the most expensive. “One million cells” means total cells at freeze to one supplier and viable cells post-thaw to another. The supplier quoting the first convention looks cheaper while delivering less, and the gap is exactly the viability percentage.

A vial described as one million cells at 80 percent viability contains fewer usable cells than a vial described as one million viable cells post-thaw. If the first is priced 15 percent below the second, the first is more expensive per usable cell.

For tissue and devices the equivalent question is: is the unit an attempt or a delivered passing unit? A quote for twenty cast tissues and a quote for twenty tissues meeting a force criterion are different jobs at different prices, and both will be written as “20 units”.

2. QC scope

Certificate of analysis, STR identity, mycoplasma, sterility, endotoxin, karyotype, marker panel, functional data. Each has a real cost, some are standard, some are chargeable extras, and no two suppliers draw the line in the same place.

The failure mode is symmetric. A supplier who includes a full release panel looks expensive next to one who includes a certificate of analysis and nothing else — and if you then buy the missing tests independently, you pay retail for them and add weeks. The QC checklist is the list to normalise against; the cost of identity and mycoplasma testing is what they cost when bought separately.

3. Media, coating and shipping

Frequently required to achieve the performance on the datasheet, and priced separately from the cells. This is not hidden — one major catalogue prices “cells only” and “kit” as separate SKUs with a $100 difference on a $616 item — but it is easy to compare across.

Shipping is the same problem with more variance. Dry ice, dry shipper and ambient live have different costs, and a quote with shipping excluded is not cheaper than one with shipping included. For international lanes the difference is material.

4. NRE versus per-part

Non-recurring engineering — tooling, mould design, process development, method development — is a fixed cost that is the same whether you order two units or twenty. A quote that buries NRE in the per-unit price will look expensive at low volume and cannot be compared against one that lists it separately.

The consequence: you cannot compare a supplier who must build tooling against one who already has it unless the split is visible. This is frequently the single largest difference between two quotes for the same physical object, and it is invisible in a lump sum.

Always ask for a repeat-order per-unit price with tooling already amortised. It tells you what the part actually costs and what you are paying for the privilege of being first.

5. Whether the acceptance criterion is written down

A quote against “functional neurons” and a quote against “mean firing rate above X Hz on a named platform at day 21 in vitro, raw traces supplied per well” are not the same commercial instrument. The first is a description. The second is a specification you can reject against.

The second is more expensive, and it should be. What you are buying with the difference is the supplier’s acceptance of technical risk. A supplier who will not write down a criterion is telling you the risk stays with you, and their lower price is correct for the job they are actually quoting.

6. IP in a developed process

If the engagement involves process development — and most custom biology does — someone ends up owning a protocol that did not exist before. Three arrangements are common, and they are worth very different amounts:

• You own the developed process outright. Highest price, and you can take it elsewhere.
• The supplier owns it, you get a licence to results. Lowest price, and you are locked in for every repeat order at whatever the price becomes.
• Joint or field-limited ownership. Everything in between, and the detail matters.

A quote 30 percent below another can be the cheaper one or the far more expensive one depending solely on which of these applies, and neither quote will mention it unless you asked. Related: cell line licensing for commercial use, which governs whether you can sell anything built on the material regardless of who owns the process.

7. What happens on a failed batch

The question is not whether batches fail. Some will. The question is whose money it is when they do, and there are exactly three answers: replacement, credit, or nothing.

A supplier who replaces failed units at their cost has priced that expectation into the quote. A supplier who charges per attempt has not. Both are legitimate; they are just different products. Comparing them on headline price compares a fixed-price contract against a time-and-materials contract and calls the second one cheaper.

Also ask what “failed” means. If there is no written acceptance criterion (point 5), there is no such thing as a failed batch, and the remedy clause is decorative.

8. Payment terms

Deposit percentage, milestone structure, net terms, and whether the final payment is contingent on acceptance. A quote requiring 50 percent up front and a quote at net 45 on delivery differ in working capital and in risk exposure even at identical headline prices.

The one that matters most: is any payment contingent on the deliverable passing acceptance? A supplier willing to hold 20 or 30 percent against acceptance is making a statement about their confidence that no marketing document can match.

9. Grade

An RUO quote and a GMP quote are not comparable at all, and a phase-appropriate GMP quote sits somewhere in between. Documentation burden, testing, facility cost and release process all change, and lead time moves from weeks to months. See RUO vs GMP cells.

If one supplier assumed GMP because you mentioned a clinical ambition in the introductory call, they are not expensive. They are answering a different question. Ask them to re-quote at RUO before concluding anything.

The normalisation worksheet

Run both quotes down this table. Fill every row, including the ones the quote is silent on — the blanks are the finding.

The normalised comparison number is row 5 plus row 7 plus row 8 plus the row 9 gap, divided by the row 11 pass rate, at the row 1 unit and row 2 convention — with row 4 shown separately and rows 10, 13, 14 and 16 recorded as terms, not numbers.

That sentence is the whole method.

Worked example A: two quotes that look different and are the same

You asked three suppliers to quote human iPSC-derived cortical neurons for a screening campaign. Two responded.

Supplier A looks 28 percent cheaper. Normalise.

Row 1 and 2 — put both on viable cells. Supplier A: 6 million total at freeze × 0.80 = 4.8 million viable cells. Supplier B: 3 × 2.5 million viable = 7.5 million viable cells. These are not the same order.

Row 7 and 8 — add back exclusions. Supplier A needs media and supplement (quote it at the supplier’s own catalogue price — say $760 for the campaign), coating ($120), and shipping ($340). A’s delivered cost becomes $4,200 + $760 + $120 + $340 = $5,420.

Row 5 — per viable million.

The “28 percent cheaper” quote is 45 percent more expensive per usable cell.

But that is not the finding either. Look at row 3. Supplier B is quoting a 2.5 million vial; Supplier A is quoting a 1 million vial. Volume tiering in this market is steep — a single supplier’s own price per million can fall by roughly 46 percent moving from a 1M to a 5M vial. Ask Supplier A for a 2.5M-vial price, and you will very likely find the two suppliers are within a few percent of each other on identical terms.

The real conclusion: these two suppliers are priced the same. You were comparing pack sizes and inclusion scope, not vendors. The decision should be made on the things that actually differ — viability method (trypan blue reads generously on debris-laden post-thaw samples, AO/PI does not; see post-thaw viability), lot availability, and whether either will reserve a lot.

Worked example B: two quotes that look the same and are very different

You need twenty engineered contractile tissue units. Both quotes come in at essentially the same number.

The headline gap is 5 percent. The actual gap is not measurable in percent, because these are different products.

Row 10 — is there a criterion. Supplier C is selling twenty units that meet a force threshold with per-unit raw data. Supplier D is selling twenty objects and a report. If four of D’s units are weak, you have four weak units and no basis for complaint, because “demonstrating contractile function” cannot be failed.

Row 11 — effective per-passing-unit price. C’s cast-to-deliver ratio is 26:20, and C is absorbing the six. So C’s $2,500 per unit is a per-passing-unit price. If D’s yield is similar and D charges per attempt, D’s effective per-passing-unit price is $71,500 ÷ 20 × (26 ÷ 26) only if all 20 delivered units are acceptable — and if four are not, your usable-unit price rises by 25 percent to $4,469 while C’s stays at $2,500.

Row 4 and 6 — the second order. C’s repeat order is $50,000 for twenty units, because the tooling exists and you own it. D’s repeat order is unquoted, and D owns the platform, so it is whatever D says it is next year. Over two orders, C is $118,000 against D’s $71,500 plus an unknown. Over four orders the gap is decisive.

Row 13 — the process. If this geometry works and becomes part of your programme, C leaves you able to take the mould and the protocol to a second supplier. D does not. That is not a line item; it is the difference between having a supply chain and having a dependency.

Row 16 — payment. C holds 30 percent against acceptance. D takes half on signature. C is staking money on hitting the criterion. D is not.

The real conclusion: Supplier C’s quote is materially better value at a 5 percent higher headline, and the gap widens on every repeat order. If your budget genuinely only supports D, the correct move is not to buy D — it is to go back to C and ask for ten units instead of twenty, keeping the criterion, the tooling ownership and the acceptance holdback intact.

The five questions that do the most work

If you only ask five things about a custom biology quote, ask these.

1. “What exactly is one unit, and is it an attempt or a delivered passing unit?” Resolves the single largest source of non-comparability in this market.

2. “Please split non-recurring cost from per-unit cost, and give me the repeat-order per-unit price with tooling already amortised.” Turns two lump sums into two comparable structures and exposes who already has the capability.

3. “What is the written acceptance criterion, and what is your remedy for a unit that fails it?” Converts a purchase into a specification, and tells you where technical risk sits.

4. “Who owns the process, the tooling and the data at the end of this?” The term with the longest tail and the one most often unstated.

5. “Give me the lead time broken down by phase.” See lead times. Reveals whether release testing and transit are inside or outside the quoted number.

Every one of these is answerable in a paragraph by a competent supplier. A supplier who bristles at all five is not necessarily bad — but they are quoting you a relationship, not a specification, and you should price that accordingly.

Reading the responses

Two things to look for once the answers come back.

What they declined to quote, and how they said so. A supplier who writes “we will not contract to a force acceptance criterion because we have not run this geometry before, but we will do a paid first article and then commit” has told you the truth and given you a path. A supplier who answers a question you did not ask is avoiding one you did.

Whether the numbers come with conditions. A specification with no method, timepoint or conditions attached is not yet a specification. This applies uniformly across viability, purity, force and firing rate, and it is the difference between a quote you can plan against and one you cannot.

Two failure modes on your own side

Comparing a first-article quote against a production quote. They look like the same document and are not. Always ask which one you are holding.

Comparing an integrator against a component supplier. If one quote covers cells, casting, maturation and force testing and the other covers casting only, the second is not cheaper — it is a fraction of the job, and you are the integrator for the rest. Cost your own time into the comparison at a real loaded rate, including the failure rate, or you will systematically under-price doing it yourself.

If you would rather not do this

We run this process as a service. Send the specification in whatever state it is in, we put the same document to named suppliers, normalise the responses onto one unit and one counting convention, and return a comparison that shows where the quotes genuinely differ — including the terms, not just the numbers.

We are a facilitator. We do not manufacture and we are not a distributor for any supplier named or implied here, which is why we can tell you when the right answer is to buy the adjacent cheaper thing, to build it yourself, or to walk away from the vendor you had already chosen.

The spread between three quotes for one written specification is market data that does not otherwise exist in this industry. That is the thing worth having, and it is the thing this page cannot give you on its own.