A local business website is working when the right people can find accurate information, complete a useful task, and receive a dependable result. Page views alone cannot establish that. Begin with the jobs the site must perform—such as calling, requesting an appointment, checking a service area, obtaining directions, or buying—and gather evidence from the customer's path through the business outcome.
Define success as completed customer work
List the website's priority tasks and give each one an observable finish. A call button should open the correct number on a representative phone. A booking should appear in the actual calendar. A form should create a readable record, send the promised confirmation, and reach an owner who can respond. Directions, hours, prices, availability, policies, and service-area statements should match the business's current operating record.
Use invented contact details for controlled tests unless a real transaction is necessary and authorized. Record the device, browser, time, page, action, expected result, observed result, and cleanup. Test from outside the administrator account. A green form banner is not enough if the request disappears after submission.
Separate search discovery from on-site behavior
Google's guide to Search Console and Google Analytics says Search Console is the source for Google Search performance such as impressions, clicks, and queries, while Analytics describes activity on the site. The products measure different things, so clicks and sessions need not match exactly.
Use that distinction in the scorecard. Search Console can show whether relevant queries and pages earn visibility and clicks. On-site analytics can show whether visitors reach contact, booking, directions, pricing, or product pages. The business system should confirm completed leads or orders. Preserve date ranges, filters, consent effects, tagging changes, and known outages. Follow trends across comparable periods rather than forcing unlike counters to reconcile.
Measure speed with field and lab evidence
The Web Vitals guidance identifies Largest Contentful Paint, Interaction to Next Paint, and Cumulative Layout Shift as the current Core Web Vitals and recommends evaluating the 75th percentile of page loads. It also distinguishes field data, which represents real users, from lab testing, which supports controlled diagnosis.
Google's Core Web Vitals documentation describes good targets as LCP within 2.5 seconds, INP below 200 milliseconds, and CLS below 0.1. Treat these as experience guardrails, not a substitute for task completion. Test important templates and devices, then use field data when enough exists. Record cache state, network, device, test location, page version, and third-party scripts so a comparison can be repeated.
Check reliability, accuracy, and accessibility
Monitor the public URL, HTTPS, DNS, certificate expiry, and key task endpoints from outside the hosting account. Verify structured contact details, title and description, canonical URL, robots rules, sitemap, and error pages. Check that the site remains usable when an optional analytics or advertising script fails.
Review headings, labels, alternative text, keyboard order, visible focus, contrast, zoom, errors, status announcements, and reduced-motion behavior. Include a screen reader and keyboard-only path for the highest-value tasks. Test phone, tablet, and desktop sizes actually used by visitors. Accessibility failures often reveal general task failures too, such as an unlabeled booking control or error message that vanishes before it can be read.
Connect website activity to business outcomes
Define qualified outcomes before reporting them. Examples include answered calls, confirmed appointments, valid quote requests, completed purchases, repeat visits to a useful reference page, or directions followed by a verified visit where measurement is lawful. Remove obvious tests, spam, duplicates, employee activity, and cancelled transactions according to a written rule.
Use a funnel with modest stages: relevant search visibility, landing-page visit, successful task start, confirmed completion, and business follow-up. Report counts and rates together. A higher conversion rate can hide falling qualified traffic, while more traffic can hide a broken booking flow. Do not infer that a ranking change caused revenue without a design that can support that conclusion.
Run controlled reviews and keep the evidence
Capture a baseline before changing content, navigation, hosting, ads, or forms. Change one bounded element when practical, then re-run the same tasks and compare equivalent periods. Guardrails should include lead quality, failure rate, accessibility, privacy, load experience, and staff handling time. Note seasonality, promotions, closures, outages, and tracking releases.
Maintain a short website record with task definitions, test results, analytics definitions, field and lab performance, search trends, incidents, customer feedback, changes, owners, and next review date. The defensible conclusion is specific: the tested tasks worked under the recorded conditions and produced the observed outcomes. That is stronger than declaring the site successful because it was online or because one dashboard number increased.
Sources
- Using Search Console and Google Analytics data for SEO — Google Search Central
- Web Vitals — web.dev
- Understanding Core Web Vitals and Google search results — Google Search Central