When to demand named sources in a local SEO review methodology

What to take away

• Procurement panels now ask for the source behind every score, so scoring rules, source lists and review dates have to sit in the methodology document itself.
• Anything a vendor states about its own results counts as a claim, not evidence, and must be labelled as such wherever it appears in a comparison table.
• Independent material sits in three places: regulator enforcement records, accredited membership criteria and published advertising rulings.
• Before/after comparisons only hold up when both columns use the same measurement window and the same definition of a lead.
• This article explains how the methodology works. It does not report hands-on testing of any agency, tool or campaign.

What a review methodology actually assesses

A methodology sets out the questions asked and the proof demanded for each answer. For local SEO services it should cover service scope, reporting cadence, contract terms, data handling and the named owner of each deliverable.

It should also state the limits. A desk review of published material cannot verify delivery quality, staff seniority or the accuracy of a client case study. Those gaps belong in the write-up, not in a footnote.

Name the evidence the method will not accept, such as a screenshot with no date, or a case study that names neither the client nor the measurement window.

Buyers comparing suppliers for 2027 need that boundary drawn early, because a score without a stated evidence base is just an opinion with a number attached.

The wider programme behind this piece is set out in Local SEO: reviews and comparison methods for 2027, which covers how the review categories fit together.

Vendor claims, independent evidence and editorial judgement

Keep three columns in your working file. The first holds what the supplier says about itself. The second holds what an outside body has published. The third holds your own assessment against your brief.

Mixing them produces a comparison that reads well and decides badly. A claim such as "average 40 per cent uplift in local pack visibility" is a marketing statement unless a named source with a date sits beside it.

Independent evidence is narrower than buyers expect. Regulators publish enforcement outcomes, and the ICO enforcement action examples for due diligence and risk assessment show how data protection failures are recorded in practice.

Editorial judgement covers fit: budget shape, sector experience, contract length and whether the reporting matches your internal calendar.

Ask for the underlying record rather than the summary. If a supplier cannot point to the page, the date and the figure behind a claim, that claim stays in the vendor column.

Where accredited status and rulings fit

Accreditation is a filter, not a verdict. Trade body membership criteria describe how an agency was admitted, which helps procurement shortlist, and the criteria for choosing accredited agencies explain what that status does and does not cover.

Advertising claims are a separate check. The independent reviews of ASA decisions for regulatory risk assessment matter when a supplier's public marketing promises outcomes that local SEO work cannot guarantee.

Neither source tells you whether the agency will hit your targets. Both tell you whether the paperwork around it is clean.

Both checks are point-in-time records. An entry dated 2024 says nothing about how a firm handled a complaint in 2026, so re-check before you sign.

For the raw inputs behind these checks, see Local SEO agency reviews: data and sources, which lists the record types worth keeping on file.

Before and after: how a score changes with sources

The table below shows the same fictional supplier scored twice. The second column is the version that survives a procurement panel. Figures are illustrative.

Assessment point	Before: claims only	After: claims plus named sources
Reporting promise	"Monthly dashboard"	Monthly dashboard, sample supplied, measured over a 90 day window
Data handling	"Fully compliant"	Policy dated, plus ICO enforcement check recorded
Accreditation	"Award winning"	Membership category named and verified
Advertising claims	"Guaranteed rankings"	Claim wording checked against published rulings
Contract	Rolling retainer	12 month term, exit notice stated in writing
Score	62 of 100	48 of 100

Treat the gap between the columns as the finding you report. The score usually falls once claims are separated from evidence. That is the point of the method.

Common questions

Does a desk review count as testing?

No. It records what is published and what is missing. Nothing here is based on running campaigns, auditing live accounts or speaking to clients.

How often should a methodology be refreshed?

At least once a year, and again whenever a cited rule, enforcement record or membership criterion changes. Each entry should carry the date it was checked.

What if a supplier refuses to name its sources?

Record the refusal and score the claim as unverified. Do not average it into the rest of the evidence as if it were confirmed.