Valuation QA Procedures That Catch Report Errors

A valuation can be well reasoned and still contain a report-level inconsistency. The market value may be £2,450,000 on the valuation page but £2,400,000 in the executive summary. A comparable may show 1,250 sq m in one table and 1,205 sq m in the analysis. These are the cases valuation QA procedures are designed to catch before a report reaches the lender.

This is not about second-guessing the valuer's opinion. It is about reviewing the finished document as a lender, panel manager or experienced colleague might read it: as one connected piece of evidence, instruction compliance, calculations and conclusions. Time pressure and familiarity with the file make this difficult to do alone, particularly when a report has been amended several times.

What valuation QA procedures need to test

A useful review process tests more than spelling, formatting and whether every heading is present. It should test whether the report tells one consistent story from instruction to conclusion.

That means reconciling market value, market rent, yield, floor area, tenancy details, dates and comparable evidence across every relevant section. It also means testing the report against the specific instruction. A report can be internally consistent and still fall short where the lender has asked for a stated marketing period, a particular level of comparable evidence, or an EWS1 disclosure.

The right level of checking depends on the asset and instruction. A straightforward owner-occupied residential report does not carry the same review points as a multi-let industrial investment with layered lease events and a special assumption. But the underlying discipline is the same: identify the figures and statements that affect the conclusion, then check that they reconcile everywhere they appear.

Start with the instruction, not the narrative

Many avoidable kick-backs are not valuation disputes. They arise because a requirement was not expressly addressed. The instruction should therefore become a short review framework before drafting is complete.

For example, if an instruction requires evidence of a minimum marketing period, the reviewer should not simply search for the words "marketing period". They should confirm that the stated period is present, relevant to the property and consistent with the evidence relied on. If the report refers to an EWS1 form, it should be clear whether it has been seen, what it states and how that position has been reflected in the valuation, where relevant.

This distinction matters. A phrase included as a standard paragraph is not necessarily an adequate response to the instruction. The report needs to show that the valuer has considered the matter in the context of the property.

Reconcile the figures that drive the conclusion

The most valuable checks tend to be simple, but they need to run across the whole report. A manual proofread often checks a page in isolation. A proper audit compares the same fact wherever it appears.

Consider a retail investment valued at £1,875,000. The capitalisation section uses a 6.25% yield, while the summary table refers to 6.50%. Either may be justifiable, but both cannot be the final adopted yield. The difference could be a drafting remnant from an earlier sensitivity run. If it is not resolved, the reader is left uncertain about the basis of the valuation.

The same applies to rent. A report may state passing rent of £132,000 per annum in the tenancy schedule, then analyse £128,000 per annum in the investment valuation. There may be an explanation, such as a rent-free period, side letter or a tenant contribution excluded from the adopted income. If so, that explanation needs to be clear. Without it, the figures look like an error rather than a considered adjustment.

Floor area is another common pressure point. Where a comparable is quoted at £3,100 per sq m based on 840 sq m, a transposed area of 804 sq m changes the implied rate and may distort the comparison. A review should check the area in the comparable schedule, narrative, calculation and source record against the adopted measurement basis.

A good audit does not flag every difference as wrong. It asks whether the difference is explained, deliberate and reflected consistently in the conclusion.

Read comparable evidence as evidence, not decoration

Comparable tables can look complete while containing weak links. The address may be right but the transaction date may have been carried over from another entry. A sale may be described as unconditional in the narrative but marked as subject to contract in the table. A deal analysed at 5.80% may be quoted elsewhere at 6.08%.

Each point can affect the weight given to the evidence. A reviewer should test the key facts for every adopted comparable: property identity, date, price or rent, area, analysed rate, tenure, lease position and source. The review should also compare these details with the narrative explanation of relevance.

This is especially helpful where the report relies on a small number of transactions. If three sales are described as the principal evidence, the reader should be able to see why they are comparable and how their differences have been adjusted for. If one is outside the stated geographical search area or significantly older than the others, that may be entirely reasonable, but it should not pass without explanation.

Check lease terms and covenant statements in context

Investment reports often contain the same lease information in several places: property description, tenancy schedule, valuation rationale and risk commentary. A lease expiry date of 24 June 2031 is not interchangeable with an unexpired term of six years if the valuation date is 31 December 2024. The former may be correct while the latter is a rounded or outdated statement.

The review should test whether expiry dates, break dates, rent review dates, passing rents and unexpired terms reconcile. It should also identify where a lease event has an effect on value but is absent from the narrative.

Tenant covenant wording requires the same care. A statement that a tenant has a "strong covenant" may be broadly understood, but it is not a substitute for the factual basis set out elsewhere in the report. Where the report refers to group support, guarantees or financial information, those statements should be consistent and appropriately qualified. The surveyor decides the covenant assessment. The review process helps make sure the language supporting it has not drifted between draft sections.

Use calculations as a cross-check, not a black box

Valuation reports contain enough calculations to create small discrepancies that become awkward under scrutiny. A gross-to-net adjustment may use the right percentage but the wrong base figure. A years' purchase factor may be correct, but attached to an earlier rent. A stated net initial yield may not reconcile to price, rent and purchaser's costs.

The audit method should recalculate material lines independently and compare the result with the report. It should also check sensible tolerances. A rounding difference of £1 is not the same as a £50,000 mismatch between the valuation table and the certificate.

Where a calculation does not reconcile, the reviewer needs the inputs as well as the output. That makes it easier to establish whether the issue is a formula, an assumption, a transcription error or simply a figure that has been updated in one place but not another.

Build review into the drafting workflow

The best time for a detailed report review is after the valuation conclusion and narrative have settled, but before the report leaves the valuer's desk. Reviewing too early creates repeated work. Reviewing only at sign-off increases the chance that an issue will be found when there is little time to correct it.

For a small team, this can be a structured final review by another valuer where capacity allows. For higher-volume work, a consistent digital audit can provide the first pass, flagging contradictions and missing requirements for the valuer to consider. It is particularly useful for checks that are repetitive and easy to miss after several revisions.

WriteUp is built for that role. It reads a draft report as one document and flags matters such as conflicting values, rents, areas, yields, lease details, comparable figures and instruction-specific omissions. Findings remain for the surveyor to assess. The tool is a second pair of eyes, not the decision-maker.

Confidentiality should be part of the workflow decision, not an afterthought. Before using any review system, firms should establish where data is processed, whether it is encrypted, whether reports are stored, and whether client information is used to train models. WriteUp uses private encrypted processing and does not store report data or use it for model training. That allows the review to support quality without treating sensitive valuation material casually.

A review standard that remains practical

There is a trade-off between reviewing every possible detail and getting a report issued on time. The answer is not an endless checklist. It is a proportionate process focused on the information most likely to affect instruction compliance, the valuation conclusion and the reader's confidence in the report.

For a low-complexity instruction, that may be a focused check of the conclusion, areas, comparables and required disclosures. For a complex investment or development valuation, it may include more detailed testing of lease events, cash flows, assumptions and sensitivity wording. The review should follow risk, not habit.

A careful final audit will not remove the need for professional judgement. It gives that judgement a cleaner document to work from, so the report is more likely to stand up under scrutiny before it is issued.