Kintsugi savings claims need cohort and baseline evidence

Reconstruct the claim population

The direct answer is to request a claim sheet for every quantified benefit. It should identify the customer cohort, industries, sizes, jurisdictions, transaction volumes, filing frequencies, prior systems, implementation stage, inclusion and exclusion criteria, observation period, attrition, missing data, sample size, statistic reported, range or dispersion, and whether the result came from measured records, customer estimates, survey responses, case studies, or modeled assumptions.

Do not use one population label for unlike metrics. The businesses counted as customers may not be the organizations measured for cost reduction, monthly time, or avoided penalties. Preserve which customers contributed to each numerator and denominator, when they were live, and whether results represent an average, median, selected example, maximum, forecast, or aggregate.

Define the baseline and counterfactual

Document the pre-implementation operating model: staff and advisor time, software and filing fees, notice response, registration work, data preparation, reconciliation, correction, penalties and interest, audit work, and management review. Normalize volumes, jurisdictions, acquisitions, seasonality, rate or rule changes, staffing, and changes in filing cadence. A lower cost after implementation may reflect a smaller scope or shifted work rather than automation alone.

Avoided penalties are counterfactual and need special care. Identify the actual exposure, authority rule, due date, transaction and jurisdiction scope, probability and basis, mitigation performed, and whether any notice or assessment existed. Do not present the maximum theoretical penalty, a corrected configuration, or a timely filing as cash actually saved without evidence connecting the intervention to an otherwise likely charge.

Measure control quality beside efficiency

Time and cost matter only with the right tax outcome. Track transaction and exemption coverage, sourcing and taxability exceptions, nexus-review accuracy, registration and account receipts, return population and reconciliation, authority acceptance, payments, amendments, notices, penalty and interest events, downtime, manual overrides, unresolved errors, and reviewer effort. Weight errors by jurisdiction, amount, customer, and reporting consequence rather than reporting one automation percentage.

Separate implementation, configured, enabled, calculated, filed, accepted, paid, reconciled, and closed states. Include internal data, finance, legal, customer, and advisor work that moves outside the software. A buyer should compare total retained operating effort and risk over a full cycle, not only the tasks the provider interface removes.

Run a buyer-specific evidence period

Before accepting an ROI case, baseline a representative period and run a controlled pilot across jurisdictions, products, exemptions, refunds, marketplace transactions, corrections, and filing cycles. Predefine the formulas, evidence sources, control thresholds, stop conditions, and owners. Reconcile every claim to invoices, time records, tax workpapers, authority receipts, payments, notices, and exceptions, then report both improvements and work shifted elsewhere.

Kintsugi's official page supports the attributed automation scope and displayed cost, time, penalty, and customer-count claims. It does not establish the study design, cohort comparability, buyer baseline, counterfactual, configuration, tax accuracy, registration, filing acceptance, payment, savings, avoided loss, compliance, or outcome for a particular business. Tax, finance, legal, data, operations, and executive owners retain those judgments.

Enterprise buyer test

Translate this change into the exact population, record type, workflow stage, decision owner, effective date, and evidence that could be affected. Ask current or prospective providers to demonstrate the named workflow with representative data and an exception—not a polished feature tour. Record what official documentation establishes, what a provider states, what the team observes, and what remains unresolved.

A defensible review also identifies the dependency outside the product. Authority interpretation, policy configuration, data quality, integrations, human judgment, approval rights, release governance, training, and retained evidence may remain customer or service responsibilities. The evaluation should preserve those boundaries instead of treating a technology claim as the complete operating model.

What we will watch next

Indirect Tax Monitor will watch the named source and affected market records for later evidence that changes status, scope, availability, implementation timing, workflow consequence, or the limits of the initial report. A later announcement does not silently overwrite this dated account; the change ledger preserves the sequence.