How to Measure Local SEO Performance During a Website Migration

Why this measurement problem matters

Migration changes URLs, templates, redirects, analytics tags, consent controls, and lead journeys at once. If all these move together, a chart cannot identify which change explains a result. Local businesses are exposed because a small set of location and service pages may generate much of the enquiry volume.

Define the unit before comparing numbers

Export pre-launch Search Console data for priority pages and query groups. In analytics, record landing-page traffic, channel definitions, call interactions, form submissions, booking starts, and completed outcomes. Preserve event definitions and note known collection gaps. Build a map connecting each old URL to its intended new destination, including redirect, canonical, and business-critical status.

Test the customer journey

Test representative journeys after launch: homepage, priority location pages, service pages, contact routes, booking flows, and language variants. Check redirects, mobile rendering, internal links, canonical tags, indexability, event collection, and delivery of test leads. A successful redirect does not guarantee stable rankings, while a ranking decline does not prove tracking has failed.

Interpret the data in layers

Diagnose in layers. First verify site availability and crawl signals. Then test analytics events and consent behaviour. Finally interpret clicks, sessions, and outcomes. Compare old and new URLs as a page group when the content has moved. Search Console and analytics measure different things, so investigate discrepancies rather than forcing totals to match.

Troubleshoot the most likely failure points

Document each repair with its deployment date. If clicks remain steady but analytics sessions fall, investigate tags, consent, filters, and hostname configuration. If both search clicks and sessions fall for the same pages, review technical SEO and demand. If only form completions decline, test the form and confirmation event before blaming search visibility.

Practical audit sequence

Start with the exact business question behind protecting measurement continuity during a site migration. Write down the decision that the result is supposed to support and the event that would count as evidence. If the team cannot agree on that definition, postpone percentage calculations until the metric has a stable meaning. Next, collect the smallest useful export from the source system. Preserve its date range, filters, timezone, and retrieval date. Check a sample of underlying records against the aggregate. If the source does not expose the stage the business wants to measure, record that limitation and select an honest proxy rather than inventing a completion event. Finally, assign an owner to any repair and specify how the fix will be verified. A task is not complete merely because a setting was changed. Re-run the relevant journey, inspect the resulting record, and note whether the intended value appears in the report. Keep evidence that another analyst could follow later.

Checklist before publishing a report

• Definition: state what the metric represents and what it does not represent. - Source: name the system that produced it and whether it is directly observed, imported, or manually classified. - Period: disclose date range, time zone, freshness, and any partial-day limitations. - Coverage: identify missing locations, pages, devices, event types, or records. - Quality: inspect duplicates, invalid values, and a sample of source records. - Context: annotate relevant campaigns, technical releases, closures, staffing, and availability changes. - Interpretation: separate verified defects from observed changes and possible explanations. - Action: name the next test, owner, and date for review. For protecting measurement continuity during a site migration, keep raw counts alongside rates whenever volume is modest. A percentage can look dramatic when its denominator is tiny. Avoid benchmarks unless a credible source, comparable definition, and relevant population are available. Internal history can be useful, but it should be labelled as the business's own baseline rather than an industry standard.

Final recommendation

Treat protecting measurement continuity during a site migration as a repeatable operating process, not a one-off chart review. Preserve the metric definition, test critical paths after meaningful changes, and keep a dated record of the evidence. When the data supports a conclusion, state it plainly; when it does not, explain the gap and identify the next practical check. A reliable report should be understandable by the business owner, the analyst who maintains the dashboard, and a future agency taking over the account. That requires clear definitions, documented limits, and recommendations that follow from observed evidence. It does not require pretending that every customer action can be attributed perfectly.