Supplier performance reviews for peptides

Supplier quality drifts quietly. A contact changes, a purification column is replaced, a drying step is shortened, and the certificates keep arriving on time. By the time a lot fails, the relationship is usually several years old and nobody has looked at the pattern. A performance review is the routine that makes the pattern visible before it becomes a failed batch.

Why a review, rather than a feeling

Most buyers carry an impression of each supplier: “reliable”, “cheap but slow”, “good on documents”. Impressions are built from the most recent interaction and the most memorable problem, which is exactly the wrong weighting. A simple scorecard replaces the impression with a record, and it does something else as well: it tells the supplier what will be measured. Most suppliers improve on the metrics their customers actually track.

The review is not an audit. An audit examines capability; a review examines performance over time. The two are complementary, and the audit method is covered in remote supplier audit.

The metrics that matter

Metric	Definition	Why it predicts trouble
On-time, in-full (OTIF)	Order received complete, on or before the confirmed date	Chronic lateness forces expedited freight, short-dated stock and rushed release testing
Documentation completeness	COA, packing list, SDS, origin and any agreed annex present and consistent at delivery	Missing documents are the leading cause of customs delay and release delay
COA accuracy	Supplier result agrees with your receipt or third-party result within the method's uncertainty	The single most important quality signal; a certificate that does not match the material is a process problem
Deviation and OOS rate	Lots with an out-of-specification result, a deviation, or a re-test, divided by lots received	A rising rate is an early warning of process drift or capacity strain
Change notification	Advance notice of a change in site, process, method or packaging	Unannounced change is how a stable supplier becomes an unstable one
Response time	Time to answer a technical question or a complaint, in working days	Slow response during a problem is more expensive than slow response during a quote
Complaint closure	Time from complaint to a documented root cause and corrective action	Replacement without investigation means the cause is still in the process
Lead-time variance	Spread of actual lead times around the quoted lead time	Variance, not the average, is what safety stock must cover
Price stability	Change in unit price at constant specification and volume	Unexplained price moves can signal a change of source or a squeeze on quality

Nine metrics is already more than most buyers track. If you start with three, choose OTIF, COA accuracy and change notification: they cover delivery, quality and the most common way a good supplier goes bad.

Collecting the data without a system

A scorecard fails when it requires a new database. The data should fall out of the purchase record you already keep:

One row per purchase order in a spreadsheet: supplier, product, lot, quantity, promised date, received date, documents complete (yes/no), receipt result, complaint reference.
One row per complaint: date opened, description, supplier response date, root cause, corrective action, date closed.
A simple definition sheet so that “on time” and “complete” mean the same thing every time. An order received a day late because the carrier missed a flight is still late; an order received early but short is not in full.

The value comes from consistency, not sophistication. A hand-maintained sheet with twelve months of entries is more useful than a dashboard with two months of clean data.

Scoring without false precision

A 1–5 scale per metric, weighted by what matters to your work, is enough. Two cautions:

Do not average away a serious failure. A supplier with excellent OTIF and one confirmed certificate mismatch is not “80% good”; the mismatch is a stop-and-investigate event, not a data point to be smoothed.
Do not over-weight price. A scorecard that rewards the cheapest quote will select for the supplier least able to invest in documentation and testing.

A workable weighting for research material is quality and documentation at about half the total, delivery at a third, and commercial factors at the remainder. The exact numbers matter less than applying them consistently and reviewing the trend rather than a single quarter.

What the trend says

Pattern	Likely cause	Response
Stable scores, stable price	Controlled process	Maintain; periodic blind check
OTIF falling, quality stable	Capacity strain or order-book growth	Discuss capacity and lead-time commitments; increase buffer stock
COA accuracy falling	Method drift, lax release, or a changed source	Request raw data and a investigation; test the next lots independently
Change notifications increasing	Process improvement or supplier instability	Ask what changed and why; require bridging data before accepting
Complaints not closing	No root-cause capability	Escalate; consider a second source
Price falling sharply	Cheaper raw material, changed synthesis, or a different site	Ask for the basis of the change; verify with testing, not with assurance

Running the review

Quarterly, light: update the sheet, look at the trend, and raise one specific issue with the supplier. This keeps the record alive without turning every quarter into an audit.
Annually, formal: share the scorecard, review the year's deviations and complaints, confirm the site and process have not changed, and agree improvement actions with owners and dates.
After a serious failure: review immediately, not at the next cycle. The purpose is to decide whether the supplier remains qualified, and what evidence would restore that status.

The meeting should produce minutes with actions, not a rating alone. A score of 3.8 out of 5 changes nothing; “two COA mismatches in Q3, root cause due 30 days, next lot tested independently” changes something.

What to do with a poor result

Improve: a specific, time-bound corrective action with evidence — not “please improve quality”.
Contain: increase receipt testing, add a retain, and hold affected stock until the investigation closes.
Diversify: qualify a second source before the current one fails completely. The bridging work is described in qualifying a second supplier.
Disqualify: a confirmed certificate mismatch that is not investigated, a refusal to provide raw data, or a change of site without notice are reasons to stop, not to negotiate.

What not to measure

Some metrics create the behaviour they claim to detect. “Number of complaints” punishes the buyer who reports problems; “average response time” rewards quick non-answers; “price per gram” ignores peptide content and documentation. Measure the outcome you actually need — material that matches its certificate, arrives when promised, with the documents required — and keep the rest as context.

For laboratory research use only. This article describes supplier management practice; it makes no claim about the performance of any specific supplier and does not support administration to humans or animals.

Editor's take

Suppliers drift unless someone is measuring. A quarterly scorecard on lot conformance, documentation and lead time is cheaper than discovering the drift in a failed batch.

Limitations

This is commercial due diligence guidance, not legal advice. Supplier claims must be verified against batch documents and, where required, an independent audit.