Sample Supplier Reconciliation: A Step-by-Step Process for Agreeing Completes, IDs and Invoices

SAMPLEDGE · FIELD NOTESISSUE No. 54 · 7 MIN READMARKETSResearch operations

Sample supplier reconciliation is the unglamorous last mile of every multi-source project, and it is where margin quietly leaks away. Fieldwork closes, the client has their data, and then the agency and each supplier have to agree exactly which respondents count. Done well, it takes a day. Done badly, it drags on for weeks of spreadsheets and emails, delays payment and sours supplier relationships. This post sets out a reconciliation process that works at respondent-ID level.

What reconciliation actually has to settle

At its core, market research project reconciliation answers one question per supplier: which respondent IDs are accepted as billable completes? Everything else follows from that list: the invoice amount, the incentives the supplier credits to its panellists, and the numbers you report internally on supplier performance.

To settle it, both sides need to agree on four categories for every ID the supplier sent:

  • Accepted completes. Qualified, finished the survey, passed quality checks.
  • Quality removals. Finished, but rejected for reasons such as speeding, straight-lining, inconsistent answers or duplicate identity.
  • Non-billable outcomes. Terminates, over-quotas and drop-outs, which may or may not carry a fee depending on the contract.
  • Disputed records. IDs where the two sides' records disagree and need investigation.

Why the numbers disagree in the first place

Before you can reconcile efficiently, it helps to know where differences usually come from. In our experience, most disputes trace back to a small set of causes.

CauseWhat happensHow to prevent it
Redirect errorsWrong status sent back, so a terminate is recorded as a complete on the supplier sideTest every redirect for every supplier before launch
Over-quota timingRespondent qualifies just as a cell closes and is counted differently by each sideAgree how over-quotas are recorded and paid
Quality removalsAgency removes completes after review; the supplier still counts themShare removal criteria at kick-off and reason codes at close
Duplicate IDsThe same respondent enters twice or two respondents share an IDDeduplicate on entry and flag repeats in the data
ID format changesLeading zeros dropped or prefixes added in spreadsheetsTreat IDs as text and agree the format up front
Late completesRespondents finish after the agreed field close timeSet a clear cut-off and decide how late completes are handled

Notice that almost every prevention step happens before or during fieldwork, not at the end. Reconciliation is easiest when the project was set up for it. The live fieldwork monitoring routine, with its decision log and supplier-level tracking, does much of this groundwork.

The reconciliation process, step by step

The sequence below works for most online projects with several sample suppliers. Adjust the timings to your contracts, but keep the order.

  1. Freeze the dataClose field at the agreed time and take a final export of every ID and its status by supplier.
  2. Apply quality decisionsFinish data checks and mark each removed complete with a reason code.
  3. Build the supplier fileFor each supplier, list every ID they sent with your final status and reason code.
  4. Share and request theirsSend the file and ask for the supplier's own list of IDs and statuses for the project.
  5. Match at ID levelCompare the two lists and sort every ID into agreed, removed, non-billable or disputed.
  6. Resolve disputesInvestigate the disputed IDs with evidence from the survey platform and the decision log.
  7. Sign off and invoiceAgree the final accepted list in writing, then raise or approve the invoice against it.

Building a supplier file that supports itself

The file you send to each supplier should answer questions before they are asked. At minimum, include the supplier's respondent ID, your internal ID if different, the final status, the quota cell, the timestamp of the final status and, for removals, a short reason code. Reason codes should be consistent across suppliers and projects, so a supplier sees the same label for the same issue every time.

Matching IDs without spreadsheet accidents

Most matching errors come from tooling, not judgement. Spreadsheets strip leading zeros, convert long IDs into scientific notation and silently trim spaces. Format the ID column as text before pasting, trim whitespace, standardise case, and check row counts after every step. When the match is done, the number of IDs in each category should add back up to the total the supplier sent.

Handling disputes fairly

Disputed records usually fall into a few patterns: the supplier shows a complete you show as a terminate, the supplier has an ID you never received, or a removal the supplier believes is unjustified. For each, pull the evidence: the survey platform's record for that ID, the redirect logs and the quality check that triggered the removal. Being consistent and transparent matters more than winning every record. A supplier who sees clear evidence for removals will accept them; one who sees vague reasons will challenge everything.

Who owns reconciliation, and by when

Reconciliation stalls most often because nobody owns it. Name one person on each project, usually the project manager or a dedicated field executive, who sends the supplier files, chases replies and signs off the final list. Agree a turnaround with suppliers at kick-off, for example that your file goes out within a set number of working days after close and that replies are expected within a similar window. Put both dates on the project plan, not in someone's head.

Remember the respondent at the end of the chain

Reconciliation is not only a commercial exercise. Many suppliers use the final accepted list to decide which panellists receive their incentive, so a slow or careless reconciliation means real people waiting for rewards they earned. That is one more reason to send removals promptly and with clear reasons.

Quality removals: agree the rules before you need them

Quality removals are the most common source of friction in invoice reconciliation, and most of that friction is avoidable. Before launch, share with each supplier:

  1. The checks you will run, such as speed, attention, open-end quality and consistency checks.
  2. How you will treat borderline cases, for example whether a single failed check is enough.
  3. The deadline by which you will send removals after field closes.
  4. Whether removed completes are replaced and, if so, how replacements are counted.

Professional codes such as the MRS Code of Conduct set standards for how research practitioners behave, and service standards such as ISO 20252 set requirements for research providers. Neither will tell you which completes to remove on a specific study, but both reward the same habit: documented, consistent criteria applied the same way to every source. Our guide to online survey data quality checks goes into the checks themselves.

From agreed IDs to a clean invoice

Once the accepted list is signed off, invoice reconciliation should be mechanical. Multiply accepted completes by the agreed rate per cell or audience, add any agreed fees for over-quotas or terminates, and compare the result with the supplier's invoice line by line. If the invoice and the agreed list differ, the conversation is now short and specific: which IDs, which rate.

  • Accepted complete list signed off by both sides
  • Rates applied per cell or audience as contracted
  • Over-quota and terminate fees applied only where agreed
  • Invoice matched to the accepted list before approval
  • Reason codes stored with the project for future reference
  • Supplier performance notes recorded for the next bid

Keep the signed-off file with the project. It is the evidence for the client invoice, the basis for supplier performance reviews and the first thing you will want if a query comes back months later. If your finance process runs separately, our post on invoicing and billing software covers how project figures can flow into billing.

AreaEnd-of-project scrambleReconciliation built into the project
Status definitionsDebated after field closesAgreed at kick-off with every supplier
Removal reasonsWritten up from memoryRecorded with each ID as decisions are made
ID matchingSpreadsheets emailed back and forthSupplier lists checked against one project record
DisputesMany, slow, frustratingFew, specific and evidence-based
Invoice approvalWaits on manual re-checksFollows directly from the agreed list

Using reconciliation data to choose better suppliers

Reconciliation produces some of the most honest supplier data you will ever have. Over a few projects, you can see which suppliers deliver clean completes for which audiences, whose redirects are reliable and who responds quickly to queries. Feed that back into supplier selection and bidding. Our article on market research supplier management covers how to structure that review.

FourStatus categories for every ID
SevenSteps from data freeze to invoice
OneSigned-off list per supplier

Where Sampledge fits

Sampledge comes from Frequent Research's own work running multi-supplier projects, where reconciliation is a weekly reality rather than an occasional chore. In SampTrek, suppliers, quotas, fieldwork outcomes and reconciliation sit on the same project record, so close-out starts from the numbers the team tracked during field instead of a fresh round of exports. Finance teams can then work from agreed figures in Talk2Tax for invoicing and payables.

If supplier reconciliation is taking longer than the fieldwork itself, book a demo and we will walk through a reconciliation on a sample project with you.

Frequently asked questions

It is the process of agreeing with each supplier, at respondent-ID level, which completes are accepted and billable once fieldwork closes, so invoices match the delivered sample.

Common reasons are quality removals, redirect errors, duplicate IDs, over-quotas recorded as completes and timing differences between the survey platform and the supplier's own system.

On respondent IDs. Totals can match by coincidence while the underlying records differ, and IDs are what both sides can actually check.

Preparation starts at kick-off, with agreed status definitions and ID formats, and the ID files should be shared soon after field closes while details are fresh.

Send each supplier the removed IDs with a reason code for each, ideally against criteria agreed before launch, so they can check and update their own records.

Often yes, because suppliers usually use the final accepted list to decide which panellists are credited, so delays and errors reach respondents too.

SampTrek keeps suppliers, fieldwork outcomes and reconciliation on the same project record, so the counts at close start from figures already tracked during field.

See it with your own projects in mind.

Book a short demo, or create a free account and look around.

WhatsApp us +91 81782 00573 · usually replies in minutes