Tighten valuation data and quality control

Improve the consistency of valuation data, testing, and sign-off so that assessments are easier to defend and more consistent across similar properties. Quality control is often the difference between a reform that survives and one that collapses into distrust.

• Different teams or vintages of data produce visibly uneven assessments.

• Management wants more confidence before scaling up valuation reform.

• The city needs a way to spot and correct systematic errors, not just individual complaints.

• Standard data definitions, review checks, and responsibility for correcting failures.

• At least one practical quality-control method, such as ratio checks, spot reviews, or peer review.

• Do not turn quality control into a bureaucratic layer with no practical tests or action.

• This is secondary if the city still lacks the basic data needed for any credible assessment.

Valuation reforms are usually easier to defend when they are presented as fairness and credibility measures, not simply as ways to raise more money. Sudden unexplained changes in bills tend to provoke resistance.

The full card would help the city plan data standards, spot checks, ratio tests, review responsibilities, and the feedback loops that keep valuation reform from drifting into inconsistency.

• Move to mass valuation when data and capacity allow; Put revaluations on a rolling cycle.

• Keep assessments current

• These cards are about preventing assessments from becoming stale between major revaluations.

Valuation reform survives only when the data and review process are good enough to produce assessments that are broadly consistent, explainable, and correctable. Quality control is the discipline that stops a good design from unravelling during implementation. It includes clear data definitions, spot checks, ratio or reasonableness tests, review thresholds, and feedback loops that turn identified errors into process improvement rather than isolated fixes.

• Different teams, years, or data sources are producing visibly uneven assessments.

• Management wants more confidence before scaling valuation reform more widely.

• The city needs a way to spot systematic error, not just respond to the loudest complaints.

• Start with the few quality checks that would catch the most damaging errors first, rather than building a bureaucratic review layer with little practical value.

• If the city is not yet running advanced models, simple tests such as spot checks, peer review, and range checks can still add strong discipline.

• Assign responsibility for correcting failures, not only for detecting them.

• Which tests matter most in your context: spot field verification, peer review, sales-to-assessment checks, boundary checks, or control totals by zone or property type.

• At what threshold a valuation should be escalated for review rather than flowing directly into the roll.

• How quality findings will be logged and used to improve methods, instructions, or training.

First 90 days

• List the most common data and valuation errors already visible in the system and decide which checks would have caught them earlier.

• Define standard data fields, coding rules, and review responsibilities so teams are not inventing their own practices.

• Choose a first batch of practical quality-control tests and apply them to a sample of recent or pilot valuations.

6 to 12 months

• Run the first quality cycle and document both the failures found and the time needed to correct them.

• Adjust the tests if they are too weak to catch material problems or too burdensome to apply regularly.

• Feed recurring errors back into training, forms, tables, or modelling rules rather than treating every failure as an individual issue.

12 to 24 months and beyond

• Embed quality checks into routine valuation operations and not only into special projects or donor-supported pilots.

• Periodically review whether the same failure patterns keep recurring and whether the underlying process needs redesign.

• Use quality evidence to support management decisions on when the city is ready to scale methods, revalue more zones, or move toward mass valuation.

• No major legal change is usually needed, but documentation standards and sign-off responsibilities should be formalised so reviews are defensible.

• Where external providers contribute data or modelling, quality-control obligations should be written into the arrangement clearly.

• If results from quality tests will trigger changes to bills, the city should clarify the corrective procedure and communication path.

• The city needs one point of coordination for data standards and review results, even if different teams conduct the checks.

• Field, valuation, IT, and management staff should all understand what counts as a quality failure and how it will be handled.

• Review time must be budgeted; quality control collapses when teams are expected to do it only in spare moments.

• If checks are too weak or irregular, management may assume quality exists when it does not.

• If checks are too elaborate for current capacity, staff may bypass them or treat them as paperwork.

• If failures are identified but never corrected, quality control becomes performative and can undermine rather than strengthen credibility.

• Share of reviewed records passing the main checks.

• Time taken to correct failed records or model outputs.

• Most common data or assessment errors by category.

• Whether recurrent failures decline after training or process changes.

• Bring sales, rents, and build-cost evidence into assessments.

• Move to mass valuation when data and capacity allow.

• Put revaluations on a rolling cycle.

• Which quality failures are doing the most harm today?

• What checks could catch them with the least extra burden?

• Who will own correction once a failure is found?

• How will management know whether quality is genuinely improving?

• C. Keep assessments current