Bottom line: the delay almost never happens at the bench
The question that comes up most often is "how long will this take," but here is the real picture. From commissioning to a report in hand, the execution of the testing itself is usually the part that can be estimated and scheduled. What actually stretches the timeline is undefined scope, unacceptable sample status, plans that bounce back and forth, and documentation that never quite arrives. These four have one thing in common: they all happen before testing starts, and most of them are decided on the sponsor's side.
So if you want a shorter timeline, chasing the laboratory to "go faster" achieves little. Clean up each of the nodes below before samples ship and the effect is immediate.
Break the flow into nodes
| Node | What should happen | Typical cause of delay | Who can solve it in advance |
|---|---|---|---|
| Scope confirmation | Product category, intended use and the list of tests are defined | Only a product name is given; the laboratory guesses the test list | Sponsor |
| Quotation and contract | Tests, sample consumption, timeline and report format agreed in one pass | Tests added after the quote is agreed, forcing a requote | Sponsor |
| Sample receipt | Samples match the packing list, identification complete | Insufficient quantity, no lot number, not finished-product status | Sponsor |
| Plan and preconditioning | Plan approved in one round, preconditioning scheduled to the plan | Multiple plan rounds, preconditioning rescheduled repeatedly | Both sides |
| Test execution | Completed inside the equipment and personnel window | Sample damaged mid-way, retests needed, equipment queue | Mainly the laboratory |
| Data handling and internal review | Calculation, checking, technical review | Incomplete raw records requiring rework | Laboratory |
| Report issuance | Drafting, review, approval, release | Cover page details, company name and address corrected repeatedly | Sponsor |
| Report amendment | Correction of textual information only | Tests added or acceptance wording changed after testing ends | Sponsor |
Take this table and self-check against it. Every row where the last column says "Sponsor" is time you can save without spending money.
Scope confirmation: where most of the delay comes from
"I need some testing" and "I need testing against this intended use, this use environment and this list of tests" are two completely different commissions. The first one enters round after round of clarification. Is the device active? Does it contact the body? How is contact duration categorized? Is it supplied sterile? Does distribution need to be considered? Every round of clarification means a new quote and a new place in the schedule.
The workable approach is to answer these questions yourself before commissioning:
- what is the intended use and the use setting, and is the user a professional or a lay user;
- does the product contact the body, and how are the contact site and contact duration categorized;
- is it supplied sterile, and by which sterilization method;
- does it contain electrical parts, software, or wireless functions;
- is the report for regulatory submission, customer acceptance, or internal development verification.
The answers to those five questions largely fix the backbone of the test list. Not knowing an answer is fine, but work through it before the quote rather than after the samples have landed. The full preparation list is in sample and documentation requirements.
Samples: the ways a delivery gets rejected on arrival
Rejection at sample receipt is common and hits the timeline hard, because it means shipping both ways plus a new place in the queue. The laboratory will normally raise an issue on the spot in these cases.
Insufficient quantity. Many tests are destructive, and one sample only carries so much testing. Sponsors who prepare "just enough" find themselves topping up as soon as a test has to be repeated or a sample fails during preconditioning. A top-up costs more than time; it also brings the burden of explaining batch consistency.
Not finished-product status. Prototype parts, semi-finished parts and unsterilized units produce data that is hard to rely on even when the testing itself is done properly. The test for final status is straightforward: identical to the product that will be placed on the market in material, process, sterilization and packaging.
Missing identification. Without a model, a lot number, or manufacturing date and batch information on the sample, the laboratory cannot link the sample to its records. ISO/IEC 17025 sets explicit requirements for sample identification and condition records, so an unidentified sample is either returned or held pending written confirmation from the sponsor, and both routes cost time.
Incomplete accompanying documents. Instructions for use, structural drawings, the bill of materials and electrical schematics are inputs to the test plan. Without them the plan simply waits.
Transport damage or non-compliant packaging. Products with batteries, samples needing cold chain and samples containing liquids are the usual victims. If the shipping method is not agreed beforehand, the sample can already be unusable on arrival.
Plan rounds: the difference between one and four
Plan approval is a shared responsibility. The laboratory drafts, the sponsor approves, and one round is the ideal. Three or four is what usually happens: seeing the plan reminds the sponsor of a function that was never mentioned, a configuration that also needs coverage, or a test that is not actually needed. Every round pushes the schedule back, because equipment and personnel windows are locked in advance and an unapproved plan cannot lock a window.
The way to cut the rounds down is to bring the product-side people, R&D or process engineering, into the plan review instead of routing everything through sales or the regulatory specialist. The descriptions of sample configuration, preconditioning conditions and acceptance basis are things an engineer spots immediately and a regulatory specialist may not.
During execution: what can be sped up and what cannot
Part of the elapsed time here is hard. Tests that need preconditioning cannot start until preconditioning is finished; tests involving ageing or simulated distribution produce no data until the process has run. That time cannot be compressed, and any offer to "skip it for a rush job" deserves suspicion, because what gets skipped is the validity of the data.
What can be compressed is queue time and the degree of parallelism. Within one batch of samples, non-destructive tests go first and destructive ones later, and tests that do not conflict can run in parallel. That has to be designed into the plan; adjusting afterwards is usually too late. The overall progression of a project is set out in our service process.
There is also a class of mid-test surprises: a sample fails or is damaged during one test and there is nothing left for the tests that follow. How it is handled depends on the cause. If the product itself failed, that is a valid result and the sponsor decides whether to remediate and retest or accept the conclusion. If the cause was operational or equipment related, the laboratory handles it under its controls for the validity of results and explains what happened. When such a dispute has to be raised formally, it goes down the retest or appeal route, which is described separately in testing knowledge.
Data and reporting: the invisible but real time
Data existing is not the same as a report existing. In between sit calculation, record checking, technical review, report drafting, review and approval. The operating logic of ISO/IEC 17025 requires technical records to be complete and traceable and requires reports to be approved by authorized personnel before release. These steps are not bureaucracy; they are where the credibility of the report comes from. There is little room to compress this stretch, but you can avoid stretching it:
- give accurate cover-page information up front. Company name, address, product model, specification description: one wrong character and the report goes into a revision process;
- state in advance how many copies are needed, in which language, and whether a stamped electronic version is required;
- do not add tests during report review. Adding one means re-entering the testing flow, and every review step already completed has to be repeated.
Pre-shipment self-check for sponsors
| Check item | What "done" looks like | Consequence if not done |
|---|---|---|
| Test list | Itemized and confirmed in writing by both sides | Quotes and schedules repeatedly overturned |
| Sample status | Finished-product status, consistent with production | Representativeness challenged, retesting required |
| Sample quantity | Per the plan's sample consumption, with a margin | Mid-project top-ups plus a batch consistency justification |
| Sample identification | Model, lot number and quantity clearly legible | Held or returned at receipt |
| Accompanying documents | Instructions, drawings and bill of materials complete | The plan cannot start |
| Report details | Cover page, address, model and specification verified | Revision process after signature |
| Change control | Design changes frozen during testing | Report does not match the device; the whole report is void |
The change control row deserves a sentence of its own. When a company runs its change control process under ISO 13485, attention naturally goes to production impact, and it is easy to forget that a batch of samples is sitting in a laboratory right now. Add a fixed checkbox to the change assessment form: are any tests currently in progress affected? The action costs essentially nothing and blocks an expensive class of rework.
What a rush order can and cannot buy
A rush order can move your project forward in the queue, add personnel and equipment windows, and shorten the internal circulation of the report. It cannot shorten preconditioning or the inherent duration of a process, and it cannot skip review and approval. When a quotation promises that everything can be expedited, ask which stretch exactly is being expedited.
A rush order also assumes the earlier nodes are clean: scope settled, samples acceptable, plan approved. Talking about expediting before those three are in place only expedites anxiety. Communication and assistance during a project run through the technical support channel, and surfacing a problem early beats catching up afterwards.
On accreditation and the timeline
Some people assume that a laboratory with a broad accredited scope must be faster. There is no necessary connection between the two. To be explicit: an accreditation mark only demonstrates that the laboratory holds the corresponding technical competence within its accredited scope, and it is not a commitment as to market access outcomes. The timeline depends on scheduling, sample status and project complexity, not on the mark. The question worth asking is a different one: are all of the tests I need within your own capability, do any of them need to be subcontracted, and does subcontracting add time?
To estimate the timeline, align the list first
If you have the product documentation and a target date for the report, send the documentation over and we will work backwards from the test list to a feasible sample submission date, marking which stretches are fixed duration and which can run in parallel. Doing this at project kick-off is worth far more than chasing progress daily afterwards. SUNGO Lab (Shanghai Shage Medical Technology Co., Ltd.) is a medical device testing laboratory accredited by CNAS, CMA and IAS (USA), with laboratories in Shanghai and Hefei. Call +86 132 4819 8029 to describe your product, or request a quote and we will send the sample list and the schedule proposal together.