Small-business outsourcing research
What share of supplier records contain a usable contact and payment term?
A protocol for measuring supplier master data quality, with clear tests for a usable contact, a current term, and an approved bank record.

Research question and scope
This protocol asks what share of supplier records contain a usable contact and a current payment term. The unit is one supplier record in the business's vendor master, identified by the legal name and the tax identifier. The study measures whether the record names a reachable contact, a payment term, a remit-to address, and an approval date. It does not evaluate whether the supplier is a good commercial choice.
Set the scope to one small business, one vendor master, and one accounts-payable process. Record the source systems, the duplicate rules, and the approval roles at the start. If the vendor master changes during the window, keep the earlier definition for the full sample or analyze the change as a separate period. A new required field can look like better data when it is only a new form.
Population and unit of analysis
Define the population as every supplier record that was active during the window, including records with no invoice in the period. The unit is the supplier record, not the invoice, because a single supplier can send many invoices. A duplicate record for the same legal entity is a separate unit with a link to the duplicate. A record that is merged during the window keeps its original identity and the merge date.
Record the supplier category, the spend band, and whether the supplier is used for goods or services. These strata matter because a one-time supplier and a recurring critical supplier carry different data needs. Include records that were created by the owner outside the normal process, because the base rate is part of the finding. Test the unit definition on two ordinary suppliers before collection begins.
A supplier can appear under a trading name while the contract uses a different legal entity. Record both names and the relationship so a payment is not sent to the wrong account. A record that cannot be matched to the contract is not usable even when every field is filled. Test the match on one invoice before the sample begins.
Definitions that make the measure reproducible
Define a usable contact as a named person or role with a working email address and a phone number that reaches the supplier. Define a current payment term as the term in the signed contract or the approved supplier form, not a term copied from an old invoice. Define an approved bank record as one verified through a callback to a known number and stored with the approval date. A record should fail the test when any one of those elements cannot be used without asking someone else.
Do not count a generic website contact form as a usable contact, even when the form sends a reply. Do not count a term that appears only in an email as current unless the email is stored with the record and names the approver. Do not count a bank record with no verification date or no callback note. Each definition should include one positive example and one boundary example.
Evidence standard and data dictionary
For each supplier record, capture the legal name, the tax identifier, the contact role, the email, the phone, the payment term, the remit-to address, and the approval date. Add the last verified date and the person who verified it. Record whether the record is a duplicate and the link to any related record. Keep the packet with the vendor master entry.
Also capture whether the record was used in a payment during the window. A record that has never been tested by an invoice may look complete while hiding a bad address. If the bank record changed during the window, record the change date and the verification. A second reviewer should be able to reproduce the completeness decision.
A payment term can be stated in the contract, the purchase order, and the invoice, and the three may disagree at the same time. Record which document governs when they conflict and who confirmed the answer. A term copied from the invoice is the supplier's claim, not the business's approval. That distinction belongs in the evidence standard and in the reviewer's training.
Sampling and collection
Draw the sample from the vendor master rather than from a list of recent invoices. Record the export date, the filters, the row count, and the selection rule. Include suppliers from every category and spend band so the result is not dominated by the largest vendors. If the population is ordered by supplier name, a fixed interval spreads the sample across the list.
Collect the record as it existed at the export date, not as it looks after a later cleanup. If a field was added after the export, record the addition date and analyze the original gap separately. Do not infer a missing term from the payment behavior. Note the collection date separately from the record creation date.
Analysis and reporting
Report the number of supplier records, the share with a usable contact, the share with a current payment term, and the share with an approved bank record. Show the counts by supplier category, spend band, and record age. Present the share of records with no invoice in the window beside those rates. Use the same denominator for every rate.
Inspect the incomplete records and describe the pattern without naming the supplier. A cluster in one category may point to a missing onboarding step, while a cluster among old records may point to a cleanup problem. Compare the result with the spend band, because a high-spend supplier may receive more attention. Report the duplicates separately.
A useful additional view is the age of the last verification for the contact and the bank record. A record verified two years ago may be complete on paper but stale in practice. Show the age distribution and describe the consequence for the next payment. That view helps the owner decide whether the onboarding form or the periodic review needs the change.
A useful cut is the share of records that were created in the last year. New records may look complete because the onboarding form is current, while old records carry the gaps. Show the age band beside the completeness rate and describe the pattern. That view helps the owner decide whether to fix the form or run a cleanup.
Limitations and sources of error
The record cannot show whether the supplier will perform well or whether the price is fair. It can show only whether the master data is usable for a payment decision. A complete record is not proof of a good supplier, and an incomplete record is not proof of a bad one. The protocol also cannot capture a contact who answers only through a personal messaging app.
Vendor masters can contain duplicates, inactive records, and entries created for a single purchase. A record that looks incomplete may be intentionally inactive. Record those cases and their treatment instead of dropping them silently. If the person who approves the supplier also verifies the bank details, note the independence limitation and analyze that subgroup separately.
Review cadence and next action
Give the result to the owner and the bookkeeper and ask them to choose one observable change. The change might be a required onboarding field, a callback rule for bank changes, or a quarterly review of high-spend records. Record the owner, the effective date, and the measure that will show whether the change worked. Avoid changing the form and the verification rule at the same time.
Repeat the protocol with the same definitions after the change and compare the same fields. If completeness improved but onboarding time rose, the new field may be adding friction without value. If bank verification improved but contact quality did not, the form may be collecting the wrong contact. Close the review when the owner accepts the finding and the next sample is scheduled.
Give the bookkeeper the same protocol and a short list of the fields that block payment. A record that is incomplete but never blocks a payment will not be fixed by a report alone. Record the owner, the effective date, and the first invoice that tested the change. The next sample should use the same definitions.
Sources
- IRS, recordkeeping for small businesses, IRS; checked 2026-11-06
- U.S. Small Business Administration, manage your business, U.S. Small Business Administration; checked 2026-11-06
- U.S. Government Accountability Office, internal control standards, U.S. Government Accountability Office; checked 2026-11-06
- Federal Trade Commission, business guidance, Federal Trade Commission; checked 2026-11-06
Frequently asked questions
Does a favorable sample prove the routine caused the result?
No. The design is observational and supports a bounded operating decision, not a universal causal claim.
What should an owner review first?
Start with high-consequence defects, missing denominators, exceptions, and cases that required the owner to reconstruct the evidence.
Related research
How often do customer voicemails wait more than one business day for a reply?
A protocol for measuring voicemail reply delay, the handoff between phone and inbox, and the calls that never get a recorded outcome.
How often does a task handoff note name an owner and a due date?
A protocol for measuring handoff note completeness, including the next action, the named owner, the due date, and the evidence attached.
How much shared-inbox work sits unassigned after one business day?
A protocol for measuring unassigned shared-inbox work, the age of the backlog, and whether a named owner picks up each conversation.