How to Track SEO Performance After a Website Redesign

A redesign can look perfect while organic acquisition quietly breaks underneath it. The fastest way to limit the damage is to monitor the sequence that creates a lead: search engines crawl a page, index the right URL, show it in results, send a visit, and that visit converts.

Each stage produces a different signal, on a different schedule. HTTP responses and forms can be tested immediately. Search Console and ranking data take longer to reflect what Google has recrawled. Your post-launch process has to respect that timing instead of waiting for a monthly traffic report.

A redesign needs an early-warning system

Google’s guidance for a site move with URL changes says temporary ranking fluctuations can occur while Google recrawls and reindexes the site. For a medium-sized website, that transition may take a few weeks or longer. Some movement is expected. Broken redirects, accidental noindex directives, or missing analytics are active failures that deserve immediate attention.

Track four connected signal families:

• Crawl and index health: Can search engines reach the new pages, interpret their directives, and select the intended URLs?
• Search visibility: Are impressions, clicks, click-through rate, and priority rankings holding up?
• On-site behavior: Are organic visitors reaching the expected landing pages and completing important events?
• Business outcomes: Are organic visits still producing qualified leads, trials, purchases, or pipeline?

A traffic total alone tells you too little. By the time it clearly falls, an upstream problem may have been running for days.

Freeze a pre-launch baseline, even if launch already happened

A useful baseline preserves the shape of performance before the redesign. Export sitewide totals, then break them down by page type, template, and commercial importance. A product page and a support article do different jobs, so averaging them together hides the patterns you need to diagnose.

What to record

• The old-to-new URL map, including unchanged URLs
• Top organic landing pages and their sessions or users
• Search Console clicks, impressions, CTR, and average-position trends
• Rankings for priority branded and non-branded keyword groups
• Organic form submissions, demo requests, purchases, calls, or other key events
• Known seasonal patterns and previous year-over-year movement

Include the pages that create business value even when they receive modest traffic. If the redesign included large-scale removals or consolidation, use the decision criteria in our SEO content audit checklist to document what was kept, merged, redirected, or deleted.

Recovering a baseline after launch

If the new site is already live, reconstruct the baseline from historical analytics, Search Console, rank-tracking history, previous crawl exports, the CMS, and your redirect spreadsheet. Google’s site-move documentation recommends compiling important URLs from sources such as sitemaps, analytics, and server logs when preparing a migration. Those same sources help rebuild the record afterward.

Add annotations for the launch and every meaningful deployment after it. Without that timeline, a redirect fix, tracking change, content release, and ranking movement blur into one chart.

The post-redesign scorecard: four layers, one diagnosis

Crawl and index health

Start with the site’s current output. Check status codes, redirect destinations, redirect chains, broken internal links, robots.txt rules, meta robots directives, canonicals, sitemap URLs, and server errors. Test representative URLs from every important template rather than validating only the homepage.

Then compare your crawl with Google’s view. Search Console provides Page indexing, URL Inspection, Sitemaps, Crawl stats, and Core Web Vitals reports. A crawler can verify the declared canonical the site returns now, along with status codes and directives. It cannot confirm which canonical Google selected. Check the indexed URL Inspection data in Search Console for Google’s selected canonical and processed view of the page.

Search visibility

Track clicks and impressions by page group, query type, device, and country. Watch the old and new URLs during a URL migration: old URL visibility should give way to the mapped destinations over time. Keep priority keyword groups in a rank tracker so a broad template regression doesn’t disappear inside an average.

Use average position as context. Google’s guidance for debugging Search traffic drops recommends focusing first on clicks and impressions, then investigating dramatic, persistent position declines. A daily wobble in one query is weak evidence. A sustained loss across the high-intent terms attached to one redesigned template is much stronger.

On-site behavior

In analytics, isolate Organic Search and review landing pages, sessions, engagement, and key events. GA4’s Landing page report can be paired with source or medium to evaluate how visitors interact with each entry page. Its Traffic acquisition report also exposes sessions, engagement, key events, and session key-event rate by channel.

Compare analytics with Search Console before diagnosing an SEO loss. If Google still reports clicks while analytics reports almost no organic sessions, inspect the tracking implementation first.

Lead and revenue signals

Finish with the outcome the redesign was supposed to improve. Track lead volume, purchases, organic conversion rate, qualified contacts, and performance by landing page. Check the whole path manually: submit the form, book the demo, complete the checkout, and confirm that the event and contact record appear where expected.

Keep the scorecard concise for stakeholders and retain the diagnostic detail underneath. Our guide to an SEO reporting dashboard that gets read shows how to separate the executive result, performance drivers, and action backlog.

A monitoring cadence for the first 90 days

Launch day

Verify analytics and conversion tracking in production. Confirm that staging crawl restrictions and noindex tags are gone. Test priority redirects, canonical output, sitemap availability, navigation links, forms, and representative templates. Run a fresh crawl as soon as the release is stable.

Days 1 through 7

Use an intensive internal review for critical errors and commercially important URLs. Inspect unexpected 404s, 5xx responses, blocked pages, canonical changes, and failed conversion paths each day. Compare Search Console clicks with analytics sessions to catch tracking gaps, while recognizing that Search Console’s view of a migration will lag behind the live site.

Weeks 2 through 4

Move broader reviews to twice weekly or weekly. Compare page groups for impressions, clicks, rankings, organic sessions, key events, and leads. Keep technical alerts under closer watch because a new deployment can reintroduce a template-level defect in minutes.

Months 2 and 3

Shift toward weekly and then biweekly trend reviews once the technical checks remain clean. Compare equivalent pre-launch periods and add year-over-year context when demand is seasonal. Google’s traffic-drop analysis guidance recommends longer Search Console ranges and comparable periods to help separate a site problem from recurring changes in demand.

This cadence is a practical operating model, not a Google requirement. Increase the frequency for a large migration, a revenue-critical site, or an unresolved incident.

How to interpret the pattern before changing the site again

Start at the earliest stage where data diverges. That narrows the investigation and keeps the team from rewriting content when the real failure is a missing tracking script or crawl block.

Observed pattern	Likely area	First check
Search Console clicks are stable; analytics sessions collapse	Tracking	Tag loading, consent behavior, channel classification, and duplicate or missing scripts
Indexed pages and impressions fall as crawl errors rise	Technical migration	Robots, noindex, responses, redirects, canonicals, and sitemap URLs
Impressions hold; clicks and CTR decline	Search appearance	Titles, descriptions, rich-result eligibility, device mix, and affected queries
Organic visits hold; leads decline	Conversion path	Forms, CTAs, event tracking, page journey, and lead routing
Rankings and traffic decline gradually on one template	Template or content	Internal links, rendered copy, headings, metadata, speed, and intent coverage

Scope matters. A sitewide decline points toward shared infrastructure or broad demand changes. A directory-level decline points toward a template, navigation, or content decision. A single-page loss calls for URL Inspection and a direct comparison with its previous version.

Canonical defects deserve special attention because a shared component can repeat one mistake across thousands of pages. Our walkthrough of a sitewide canonical tag bug explains how to compare current output with Google-selected canonicals and verify recovery.

Also check events outside the redesign. Google’s traffic-drop guidance recommends reviewing seasonality, changing search interest, ranking updates, manual actions, and security issues. Correlation with launch is a strong lead, but it isn’t proof.

Escalation rules that prevent dashboard watching

Every monitoring plan needs triggers and owners. Otherwise, the team watches a decline, debates whether it is “normal,” and loses the cleanest recovery window.

Contain immediately

Escalate broad crawl blocks, accidental noindex directives, widespread failed redirects, template-level canonical defects, sustained server errors, and broken conversion tracking on priority paths. Freeze unrelated SEO changes during a serious incident. Preserve affected examples, deployment times, and crawl evidence before rollback or repair changes the output.

Investigate against normal variance

A universal percentage threshold is unreliable. Build triggers from each page group’s history, business value, and usual volatility. A modest decline on a stable pricing-page cluster may deserve faster investigation than a larger swing in seasonal editorial traffic.

For each incident, record:

1. The affected URLs, templates, and business outcome
2. The first signal in the chain that changed
3. The likely mechanism and evidence
4. The containment action and owner
5. The production acceptance test and review date

Once emergencies are contained, compare remaining fixes by impact, reach, confidence, and effort. Our framework for prioritizing technical SEO fixes turns that evidence into a development queue without treating every crawler warning as equal.

Run the monitoring loop in ProjectHQ

ProjectHQ brings the operational layers into one workspace. Our Site Audit crawls the site, groups issues by severity, and supports manual rescans after a fix. Use it to find broken links, redirect behavior, noindex issues, duplicate metadata, orphan pages, and other crawl-level changes.

Our Rank Tracking monitors keyword positions, historical trends, devices, locations, and SERP features. Pair those trends with Website Analytics for organic sources, top pages, traffic behavior, and custom conversion events. Then use CRM and Contacts to review contact activity, source context, form submissions, and movement from lead to qualified contact.

Keep Google Search Console in the workflow. It supplies Google’s own crawl, index, and search-performance data, including page-level inspection and Google-selected canonical information. A practical loop is simple: crawl after launch, track the priority queries, monitor organic landing pages and events, review contact outcomes, fix the earliest broken stage, and rescan production.

Your next monitoring review

Choose a small cohort of commercially important URLs plus examples from every changed template. Validate responses, redirects, robots directives, canonicals, internal links, analytics, and forms. Then compare Search Console visibility, on-site visits, and lead outcomes against the baseline.

End the review with four clear statements: what changed, where the first break appears, who owns the action, and when production will be checked again. That gives the team enough structure to tolerate ordinary reindexing movement while responding quickly to evidence of a real migration failure.