The Website Evidence Handbook

A plain-language guide to the checks Evident Sites uses to evaluate website speed, visibility, accessibility, safety, content and technical housekeeping.

Open Website Evidence Handbook surrounded by cards representing speed, search, accessibility, security, content and technical upkeep.

A website can look finished and still make customers wait, hesitate, get lost or leave.

This handbook explains the checks behind an Evident Sites Website Evidence Report. It is written for business owners, not developers. You will see what each check looks for, why it matters, what “good” generally looks like and what your business or visitors gain when it is handled well.

You do not need to memorize any of it. You do not need a perfect score. And a result marked “worth attention” does not automatically mean your website is broken, unsafe or costing you business. It means there is evidence worth understanding.

Checks: 107
Dimensions: 6
Handbook version: 2.0
Catalog reviewed: 2026-08-27

The catalog is versioned because the web does not stand still. If you are holding a report, check that its rule-set version matches this edition before comparing totals against a later run.

Before we get into the checks

What this handbook is

Most website advice falls into one of two camps. One tells you that everything is urgent — every score a crisis, every missing setting supposedly costing you customers. The other hands you a dashboard full of technical terms and leaves you to work out whether any of them matter.

This handbook takes a different approach. For each check, we ask four practical questions:

What are we actually checking?
Why might it matter to a visitor, search engine or business owner?
What does a healthy result generally look like?
What do you gain by getting it right?

The goal is not to frighten you into rebuilding a website. The goal is to give you enough context to make a sound decision.

What this handbook is not

This is not a promise of higher rankings, more leads or complete legal compliance. It is not a penetration test, a full manual accessibility audit or a substitute for legal advice.

Automated tools are useful because they are repeatable. They can identify measurable strengths and likely gaps. But they cannot understand every customer, every business goal or every real-world use of a website. That is why the report combines tool-based evidence with a plain-language review.

What each result means

Passing

The website met the rule or threshold used when the test was run. It does not mean the website can never fail, that every page is perfect or that nothing should ever change.

Worth attention

The result deserves a closer look. It may indicate friction, missed visibility, weak resilience or an opportunity to make the site clearer. Some items are inexpensive improvements; others involve real tradeoffs; a few may not be relevant to a particular website at all. The label is intentionally calmer than “failed” — the report exists to support a decision, not manufacture alarm.

Not tested

We could not complete that measurement during this audit run, or the source did not have enough information to return a responsible result — for example, Google may not have enough real-visitor data for a lower-traffic page. It is not a hidden failure and it is not treated as a pass. The report states the reason plainly so the check can be revisited.

Not applicable

The check does not reasonably apply to this website. A license-number check may not apply to an unlicensed profession; a font-loading check may not apply when a site uses no downloadable web fonts. Marking that honestly is more useful than forcing every website into the same checklist.

Why the checks are grouped two ways

The Website Evidence Report shows four buyer-facing scores near the front — First Impression, Calls & Leads, Mobile Experience and Search Readiness. Those answer questions a prospective customer might ask.

The underlying checks are organized into the six technical dimensions below. Those help explain what is happening underneath the experience.

1. Speed

Can a customer see and use the site without waiting?

A website appearing in stages on mobile, with evidence cards for loading speed, tap response and page stability.

Speed is not only a technical concern. It shapes the first few seconds of the customer experience.

A visitor may be standing in a basement with weak service, sitting in a truck between appointments or trying to find an emergency number late at night. A website that feels fine on an office computer may feel completely different on that visitor’s phone.

We check both overall performance and the specific conditions that commonly make a page feel slow or unstable.

When Google has enough recent field data, the report leads with what real Chrome visitors experienced over the preceding measurement window. That evidence is stronger for describing the current customer experience because it reflects many real visits, devices and connections. The simulated mobile test remains useful as a controlled diagnostic and as the primary evidence when field data is unavailable.

A field-data check marked “not tested” usually means Google did not have enough eligible traffic for that page or origin. It does not mean the page passed or failed.

The practical win: Faster pages. Less frustration. More visitors reaching the information or action they came for.

16 checks in this dimension.

SPD-01Phone speed score

What we check: We run the page under simulated mobile conditions and review its overall performance score.
Why it matters: Phones are often slower than desktop computers and may be using a weaker connection. Mobile testing is more likely to expose heavy images, blocking scripts and delayed content.
What good looks like: The main content appears promptly, the page remains stable and important interactions become available without a long wait.
What you gain: A more dependable first impression for the customers most likely to be browsing on the move.

SPD-02Desktop speed score

What we check: We run the same page under desktop conditions to see whether the site performs well with greater processing power and bandwidth.
Why it matters: A good desktop score helps separate general performance problems from issues that mainly affect mobile devices.
What good looks like: Pages appear quickly and respond smoothly without depending on unusually powerful hardware.
What you gain: A fast, professional experience for office users, commercial buyers and customers doing more detailed research.

SPD-03How long before your page appears

What we check: We measure how long it takes for the page’s largest important visible element—often the main image or headline area—to appear.
Why it matters: Visitors judge speed by what they can see, not by whether code is still loading in the background.
What good looks like: The main message or visual arrives early enough that the page feels ready.
What you gain: Visitors can understand where they landed before impatience turns into an exit.

SPD-04Page stability

What we check: We look for content that shifts position unexpectedly while the page loads.
Why it matters: A moving button or jumping paragraph can cause mis-clicks and makes the site feel unfinished.
What good looks like: Space is reserved for images and other elements so the layout stays put as resources arrive.
What you gain: A calmer page that is easier to read, tap and trust.

SPD-05Responsiveness while loading

What we check: We look for long tasks or heavy scripts that may keep the browser too busy to respond promptly.
Why it matters: A page can look visible and still ignore a tap, click or scroll because the browser is occupied.
What good looks like: The page remains responsive while work continues in the background.
What you gain: Less of the “I clicked it, but nothing happened” feeling that makes visitors repeat actions or leave.

SPD-06How long before anything appears

What we check: We measure when the first meaningful visual content appears on the page.
Why it matters: A blank screen can feel like an error even when the rest of the page is still loading normally.
What good looks like: The visitor receives an early visual signal that the page is working.
What you gain: A faster perceived experience, even before every component has finished loading.

SPD-07Speed Index

What we check: We review how quickly the visible portion of the page fills in over time.
Why it matters: Two pages can finish loading at the same moment while one feels much faster because useful content appears progressively.
What good looks like: The above-the-fold area becomes visually complete without a prolonged partial state.
What you gain: A site that feels ready sooner and creates less uncertainty.

SPD-08How long before the page can be used

What we check: We estimate when the page is sufficiently loaded and responsive for normal interaction.
Why it matters: Visible is not always usable. Visitors need navigation, buttons and forms to work when they reach for them.
What good looks like: Important actions become available close to the time the page first appears.
What you gain: Customers can call, navigate or request service without waiting for invisible work to finish.

SPD-09Files that hold up the page

What we check: We identify stylesheets, scripts and other resources that delay the browser from showing important content.
Why it matters: Some files must load first. Others can be reduced, deferred or reorganized so they do not block the initial view.
What good looks like: The page prioritizes the resources needed for the first screen and delays nonessential work.
What you gain: Faster visible content without removing useful functionality.

SPD-10Photo file format on the homepage deep test

What we check: Using Google’s homepage performance test, we look at whether large images in that deep test use formats that balance visual quality with efficient file size.
Why it matters: Images are often the heaviest part of a small-business website.
What good looks like: Photos use appropriately compressed, broadly supported modern formats when practical.
What you gain: Faster pages with less data usage and little or no visible loss in quality.

SPD-11Photos loading at once on the homepage deep test

What we check: Using the homepage performance test, we determine whether off-screen images load immediately or wait until the visitor approaches them.
Why it matters: Downloading every image before it is needed can compete with the first screen for bandwidth.
What good looks like: Important above-the-fold images load promptly while appropriate off-screen images are deferred.
What you gain: A faster opening experience, especially on image-heavy pages.

SPD-12Photo space reserved on the homepage deep test

What we check: Using the homepage performance test, we look for images whose dimensions are not defined before the files arrive.
Why it matters: Without reserved space, surrounding text and buttons can jump when an image loads.
What good looks like: The page knows each image’s intended shape and preserves the layout in advance.
What you gain: Better visual stability and fewer accidental taps.

SPD-13What real visitors actually wait for

What we check: When available, we review Google’s recent field measurement for when the largest important visible element appears for real Chrome visitors.
Why it matters: A controlled lab test describes one simulated visit. Field data reflects the range of devices, connections and conditions actual visitors encountered over time.
What good looks like: The experience falls within the current healthy range for the large-content loading metric across most eligible visits. When Google has insufficient data, the result is marked not tested rather than guessed.
What you gain: A more realistic view of what customers have been waiting for—not only what happened during one test run.

SPD-14How fast taps respond for real visitors

What we check: When field data exists, we review how quickly the page responds after real visitors tap, click or use the keyboard.
Why it matters: A page can appear complete while still feeling sluggish when someone tries to open a menu, choose an option or begin a form.
What good looks like: Interactions respond within the current healthy range for most eligible visits, without long periods in which the page appears to ignore the user.
What you gain: More confidence that important actions feel responsive under real-world conditions.

SPD-15Page stability for real visitors

What we check: When field data exists, we review how much unexpected layout movement real Chrome visitors experienced.
Why it matters: A controlled test may miss shifts caused by different screens, content, connection speeds or third-party elements.
What good looks like: The page remains visually stable for most eligible visits, with space reserved before images, embeds and other delayed elements arrive.
What you gain: Stronger evidence that customers are not regularly chasing moving buttons or losing their place while reading.

SPD-16Text visible while fonts load

What we check: When the site downloads custom fonts, we review whether readable fallback text can appear while those font files are still arriving.
Why it matters: A decorative typeface should not create a period in which important words are invisible.
What good looks like: Text appears promptly using a sensible fallback and changes cleanly when the intended font becomes available. If the site uses no downloadable web fonts, the check can be marked not applicable.
What you gain: Visitors can begin reading immediately, even on a slower connection or when a font service is delayed.

2. Being Found

Can search engines discover, understand and present the right pages?

Website pages connected through a sitemap to search, location and business-listing evidence.

Search readiness is not the same thing as a ranking guarantee.

Google considers many signals outside the website, including competition, location, reputation and relevance. These checks focus on the part a website owner can control: whether the site gives search engines clear access, useful page information and a coherent structure.

The practical win: Better eligibility to be found, clearer search listings and fewer technical barriers between your pages and potential customers.

22 checks in this dimension.

FND-01Page title

What we check: We confirm that the page has a descriptive title in the document metadata.
Why it matters: Search engines and browser tabs use the title to identify the page. It can also influence the title shown in search results.
What good looks like: A concise title that accurately describes the page and distinguishes it from other pages.
What you gain: Clearer search listings and easier recognition when several tabs are open.

FND-02The description under your Google listing

What we check: We look for a useful meta description that summarizes the page.
Why it matters: Search engines may use this text when building the descriptive snippet beneath a result, although they can choose other page content instead.
What good looks like: A specific, honest summary that helps the right searcher decide whether to click.
What you gain: A stronger opportunity to set expectations before the visitor reaches the site.

FND-03Google is allowed in

What we check: We look for rules that unintentionally block crawling or request that a page not appear in search.
Why it matters: A valuable page cannot perform in organic search if search engines are prevented from accessing or indexing it.
What good looks like: Public pages intended for discovery are accessible, while private or low-value areas are controlled intentionally.
What you gain: Fewer accidental visibility problems.

FND-04A secure connection

What we check: We confirm that the searchable version of each page uses a secure HTTPS address.
Why it matters: HTTPS protects data in transit and is the expected standard for modern websites.
What good looks like: The secure version loads consistently, and alternate versions resolve to the intended address.
What you gain: A cleaner canonical site, secure delivery and fewer browser warnings.

FND-05Mobile friendly

What we check: We review whether pages render and function at phone widths without forcing awkward zooming or horizontal scrolling.
Why it matters: Search engines evaluate mobile experiences, and many local-service searches happen on phones.
What good looks like: Text, navigation, forms and calls to action remain usable on a small screen.
What you gain: A site that can serve mobile searchers after they find it.

FND-06The headline on your page

What we check: We look for a clear primary heading that identifies the page’s main subject.
Why it matters: The main headline helps visitors and search systems understand what the page is about.
What good looks like: One prominent, descriptive headline aligned with the page’s actual content.
What you gain: Faster comprehension and clearer topic signals.

FND-07Descriptions on your photos

What we check: We look for useful text alternatives on meaningful images.
Why it matters: Alternative text supports people using screen readers and gives search systems context about image content.
What good looks like: Informative images have concise, accurate descriptions; decorative images are handled so they do not add noise.
What you gain: More accessible content and better-understood imagery.

FND-08A map of your pages for Google

What we check: We look for an accessible XML sitemap that identifies the public pages the site wants search engines to discover.
Why it matters: A sitemap can help search engines find and monitor pages, particularly on a new, large or frequently updated site.
What good looks like: The sitemap is present, readable and intentionally published. FND-21 separately verifies that the addresses it lists resolve cleanly.
What you gain: A clearer inventory for search engines and easier visibility monitoring.

FND-09Instructions for search engines

What we check: We review the site’s robots.txt file and related crawling directives.
Why it matters: A small configuration mistake can discourage crawling of important pages or resources.
What good looks like: The rules reflect an intentional crawling strategy and do not block necessary content.
What you gain: Fewer accidental technical barriers.

FND-10One official address per page

What we check: We look for a canonical URL that identifies the preferred address for the page.
Why it matters: The same content can sometimes be reached through multiple URLs. A canonical signal helps consolidate those versions.
What good looks like: Each indexable page points to the intended secure, public URL.
What you gain: Cleaner indexing signals and fewer duplicate-address ambiguities.

FND-11Business details Google can read

What we check: We look for relevant structured data that describes the business or page in a machine-readable format.
Why it matters: Structured data gives search systems explicit clues about names, services, locations and other entities.
What good looks like: Markup is relevant, accurate, visible on the page and consistent with current search documentation.
What you gain: Better machine understanding and potential eligibility for enhanced search presentation when supported.

FND-12Your business details, as Google reads them

What we check: We validate structured data and compare the extracted information with what the business actually says.
Why it matters: Markup can exist while still being incomplete, invalid or inconsistent.
What good looks like: Testing tools can parse the information without critical errors, and the values match the visible business details.
What you gain: More reliable machine-readable information and fewer misleading signals.

FND-13Headings in order

What we check: We review whether headings form a logical outline instead of being chosen only for their visual size.
Why it matters: Clear structure helps people scan the page and helps systems interpret the relationship between sections.
What good looks like: A meaningful primary heading followed by logically nested section headings.
What you gain: Pages that are easier to scan, understand and maintain.

FND-14How this page looks when shared

What we check: We inspect the title, description and image information used by social and messaging platforms when a link is shared.
Why it matters: An incomplete or accidental preview can make a legitimate link look unfinished.
What good looks like: Shared links show an intentional image, recognizable title and accurate description.
What you gain: A more credible introduction when customers or partners share your site.

FND-15How deep your pages sit

What we check: We count how many navigation steps or internal links are generally required to reach important pages.
Why it matters: Deeply buried pages are harder for visitors and crawlers to discover.
What good looks like: Important services, locations and conversion pages are reachable through a sensible information structure.
What you gain: Easier navigation and stronger internal discovery.

FND-16Pages nobody can click to

What we check: We identify public pages that exist in the crawl but receive no internal links from other pages.
Why it matters: Orphaned pages are difficult for visitors to discover and may receive weaker internal context.
What good looks like: Every useful public page has an intentional place in the site structure.
What you gain: Fewer forgotten pages and a more coherent website.

FND-17Every page titled differently

What we check: We compare titles across the site for duplicates.
Why it matters: Repeated titles make it harder to distinguish pages in search results, analytics and browser tabs.
What good looks like: Each important page has a title specific to its subject and purpose.
What you gain: Clearer page identity and better search-result differentiation.

FND-18Every page described differently

What we check: We compare meta descriptions across the site.
Why it matters: Copying the same description everywhere wastes an opportunity to explain why each page is relevant.
What good looks like: Important pages have unique, useful summaries; low-value pages are handled intentionally.
What you gain: More specific search snippets and cleaner content management.

FND-19Your Google listing, linked from the site

What we check: We look for an intentional link or map connection from the website to the business’s correct Google listing when local discovery is relevant.
Why it matters: The website and the listing are two major parts of a local business’s public presence. Connecting them helps customers move between official information, directions, reviews and the business’s own pages.
What good looks like: The site points to the correct current listing without sending visitors to a generic search result, outdated location or unrelated profile.
What you gain: A clearer path to directions, reviews and other location information customers often want before contacting a service business.

FND-20Your Google Business Profile

What we check: When the required data source is available, we verify that an appropriate Google Business Profile can be found and review key public details for basic consistency.
Why it matters: The profile can shape how a local business appears across Google Search and Maps, independently of the website itself.
What good looks like: The profile represents the correct business and location, points to the intended website and contains current customer-facing information. If the lookup cannot be completed, the report says not tested and explains why.
What you gain: Better visibility into whether the website and one of the business’s most important local-search assets are reinforcing the same story.

FND-21Your map for Google is accurate

What we check: We request the URLs listed in the XML sitemap and confirm that they resolve to current, intentional pages.
Why it matters: A sitemap can exist while still listing removed pages, errors, redirects or addresses that should no longer be treated as official.
What good looks like: Sitemap entries load successfully, use the preferred secure addresses and represent pages the business actually wants search engines to consider.
What you gain: A more trustworthy page inventory and fewer stale or confusing signals sent to search engines.

FND-22Links that bounce before landing

What we check: We follow internal links and identify unnecessary redirect steps between the clicked address and the final page.
Why it matters: Redirects are sometimes necessary, but avoidable chains add delay, complicate crawling and make future maintenance harder.
What good looks like: Current internal links point directly to their intended destinations, while legitimate old addresses use a single clean redirect when possible.
What you gain: Faster navigation, simpler maintenance and a cleaner route for visitors and search crawlers.

3. Works for Everyone

Can people with different abilities comfortably understand and use the site?

A website navigated by keyboard with visible focus, page landmarks, readable contrast and screen-reader support.

Accessibility is not a feature added for a small group of people. It is part of making a website perceivable, operable and understandable across different ways of seeing, hearing, moving and processing information.

Automated testing can identify many common issues. It cannot prove that a website is fully accessible. Keyboard review, screen-reader testing and human judgment still matter.

The practical win: Broader access. Less friction. A more usable experience for everyone.

13 checks in this dimension.

4. Safety & Trust

Does the site give visitors and browsers reasonable evidence that it is safe?

A protected website surrounded by certificate, domain-renewal and authenticated-email evidence.

The padlock is important, but it is only one part of website safety.

We also look at certificate health, malware signals, secure delivery and common browser protections. These checks identify practical hardening opportunities. They are not a penetration test and do not prove that a website can never be compromised.

The practical win: More confidence. Safer browsing. Less avoidable risk.

16 checks in this dimension.

TRU-01Mozilla’s browser-safety grade

What we check: We review a repeatable external assessment of common browser-facing security protections.
Why it matters: A site can use HTTPS and still omit headers that help browsers limit certain classes of misuse.
What good looks like: Strong protections are present without breaking legitimate site functionality.
What you gain: A clearer baseline for browser hardening and future improvement.

TRU-02The strength of your padlock

What we check: We assess the public HTTPS/TLS configuration using an external server test.
Why it matters: The presence of a padlock does not describe the full quality of the underlying protocol configuration.
What good looks like: Current protocols, trusted certificate chains and sensible configuration for the site’s environment.
What you gain: More dependable encrypted connections across supported devices.

TRU-03Certificate expiry

What we check: We confirm that the site’s certificate is valid and note when it expires.
Why it matters: An expired certificate can trigger a severe browser warning and block customers from proceeding.
What good looks like: The certificate is valid, matches the domain and has reliable renewal in place.
What you gain: Fewer preventable interruptions and trust-damaging warnings.

TRU-04Malware and blacklist check

What we check: We query public threat-intelligence sources for known malware, phishing or blacklist signals associated with the URL or domain.
Why it matters: Browsers, security tools and customers may avoid a site that has been flagged.
What good looks like: No current detections, with any prior issue investigated rather than ignored.
What you gain: Early awareness of reputation or compromise signals.

TRU-05A secure connection

What we check: We confirm that pages load through HTTPS and that the connection is not immediately rejected by the browser.
Why it matters: HTTPS helps protect data between the visitor and the website.
What good looks like: Secure connections are available sitewide and used by default.
What you gain: A standard layer of transport protection and visible browser confidence.

TRU-06Old links still land safely

What we check: We examine whether common alternate or older addresses redirect to the intended secure destination.
Why it matters: Bookmarks, directory listings and old search results may continue sending visitors to earlier URLs.
What good looks like: Permanent redirects lead cleanly to the correct HTTPS page without loops or unnecessary chains.
What you gain: Fewer dead ends and preserved value from existing links.

TRU-07Nothing insecure inside a secure page

What we check: We look for images, scripts or other resources requested over an insecure connection from an HTTPS page.
Why it matters: Mixed content can trigger warnings, be blocked by browsers or weaken the protection visitors expect.
What good looks like: All page resources load securely from trusted locations.
What you gain: Cleaner browser behavior and fewer hidden weak points.

TRU-08Server version kept private

What we check: We look for unnecessary server or framework version details exposed in response headers.
Why it matters: Version information can make reconnaissance easier without helping ordinary visitors.
What good looks like: The server shares only the information required for normal operation.
What you gain: A smaller public information footprint.

TRU-09Browsers told to stay secure

What we check: We look for a Strict-Transport-Security policy where appropriate.
Why it matters: This policy tells compatible browsers to use HTTPS for future visits and helps prevent accidental insecure access.
What good looks like: The policy is deployed carefully after the entire domain is ready for secure-only traffic.
What you gain: Stronger enforcement of the site’s secure connection.

TRU-10Rules about what code may run

What we check: We review whether the site uses a Content Security Policy or an appropriate path toward one.
Why it matters: A CSP can limit where scripts and other resources are allowed to come from, helping reduce certain injection risks.
What good looks like: A tested policy reflects the site’s real dependencies without relying unnecessarily on unsafe allowances.
What you gain: Another layer of defense if unwanted code reaches a page.

TRU-11Nobody can frame your site as theirs

What we check: We look for browser rules controlling whether other sites may embed the page in a frame.
Why it matters: Unrestricted framing can contribute to clickjacking and misleading presentation.
What good looks like: Framing is limited using modern policy controls while preserving legitimate integrations.
What you gain: Better protection against another site presenting your interface as part of its own.

TRU-12Files treated as what they are

What we check: We look for protection against browsers guessing a resource type different from the one declared by the server.
Why it matters: Type guessing can create unexpected behavior and, in some situations, security risk.
What good looks like: Resources have correct content types and browsers are instructed not to reinterpret them unnecessarily.
What you gain: More predictable handling of files and responses.

TRU-13What you leak when visitors leave

What we check: We review the Referrer-Policy that controls how much address information may be shared when a visitor follows a link.
Why it matters: A full referring URL can sometimes reveal more context than the destination needs.
What good looks like: The policy balances useful attribution with appropriate privacy.
What you gain: Better control over information passed to other sites.

TRU-14Camera, microphone and location rules

What we check: We look for a Permissions Policy controlling access to sensitive browser capabilities where appropriate.
Why it matters: Most service-business websites do not need broad access to cameras, microphones or location sensors.
What good looks like: Capabilities are limited to what the site genuinely uses, without breaking required features.
What you gain: A more intentional browser-permission surface.

TRU-15Domain registration current

What we check: We review available public domain-registration data to confirm that the domain is registered and note the published expiration date when the registry provides one.
Why it matters: A domain is the address customers, search engines and email systems depend on. An accidental lapse can interrupt the website and business email at the same time.
What good looks like: The domain is active, its renewal responsibility is understood and automatic renewal or another dependable renewal process is in place. Some registration details may be private or unavailable and should not be guessed.
What you gain: Earlier warning of one of the most preventable—and disruptive—website ownership failures.

TRU-16Your email reaching inboxes

What we check: We review the public DNS records used to authenticate email sent from the business’s domain, including DMARC and related signals available to the audit.
Why it matters: These records help receiving mail systems evaluate whether a message is legitimately associated with the domain. Missing or conflicting settings can weaken protection against impersonation and may contribute to delivery problems.
What good looks like: Authentication records are valid, aligned with the systems that actually send mail and monitored before stricter enforcement is introduced.
What you gain: A stronger foundation for trustworthy business email and clearer control over unauthorized use of the domain. This check is evidence about configuration—not a guarantee that every message will reach an inbox.

5. What It Says

Does the content explain enough, clearly enough, to support a decision?

A clearly organized website page with reviews, editing marks, images and maintained external links.

A technically sound website can still underperform if the message is thin, vague or difficult to scan.

These checks look for signs that the content is complete, readable and maintained. A word count alone cannot determine quality. More words are not automatically better. The goal is enough useful information to answer the questions a real customer brings to the page.

The practical win: Clearer value. Stronger credibility. More confident inquiries.

20 checks in this dimension.

CNT-01Leftover placeholder text

What we check: We scan for common temporary copy such as lorem ipsum, sample labels and unfinished template language.
Why it matters: Placeholder text makes a live site feel abandoned and can confuse both visitors and search systems.
What good looks like: Every visible word is intentional and specific to the business.
What you gain: A more credible presentation and fewer embarrassing launch remnants.

CNT-02Broken links inside your site

What we check: We follow links from the reviewed pages to other pages and resources on the same website and identify destinations that return errors or cannot be reached. CNT-14 separately checks links that leave the site.
Why it matters: Broken links interrupt research and suggest that the site is not being maintained.
What good looks like: Important links lead to the intended live destination or are removed when no longer relevant.
What you gain: Fewer dead ends and a smoother path through the content.

CNT-03Broken photos

What we check: We look for image references that fail to load.
Why it matters: Missing photos leave empty areas, undermine trust and can remove important proof.
What good looks like: Images resolve correctly, include sensible fallbacks and are monitored after changes.
What you gain: A complete presentation with fewer visible maintenance problems.

CNT-04How hard your writing is to read

What we check: We use readability signals and editorial review to identify unusually dense sentences, jargon or complicated structure.
Why it matters: Customers often scan quickly and may be unfamiliar with industry language.
What good looks like: Direct sentences, familiar words, informative headings and technical terms explained when needed.
What you gain: More visitors understand the offer without feeling talked down to.

CNT-05How much your site actually says

What we check: We estimate the amount of meaningful, visible text across the sampled pages.
Why it matters: Very thin pages may not answer customer questions or give search systems enough context. Excess copy can be just as unhelpful.
What good looks like: Each page contains enough original information to fulfill its purpose without padding.
What you gain: Better-informed visitors and more complete topic coverage.

CNT-06Photo file sizes

What we check: We identify image files that are substantially larger than their displayed use requires.
Why it matters: Oversized images slow pages without necessarily looking better.
What good looks like: Images are resized and compressed for their actual display dimensions.
What you gain: Faster delivery while preserving useful visual evidence.

CNT-07Photo file formats across your pages

What we check: Across the audited-page crawl, we review whether content images use efficient formats appropriate to the image type and browser support needs.
Why it matters: Format choice affects both visual quality and page weight.
What good looks like: Photos, illustrations and transparent graphics use formats suited to their purpose.
What you gain: Better-looking content without unnecessary download cost.

CNT-08Photos loading at once across your pages

What we check: Across the audited-page crawl, we look at whether content images outside the initial view are loaded before they are needed.
Why it matters: Long galleries and project pages can become heavy when every image downloads immediately.
What good looks like: Loading priority reflects where the image appears and how important it is to the first view.
What you gain: Rich visual content with less impact on the opening experience.

CNT-09Photo space reserved across your pages

What we check: Across the audited-page crawl, we look for content images that cause nearby text to shift as they load.
Why it matters: Movement interrupts reading and can make evidence such as reviews or project photos harder to follow.
What good looks like: Image proportions are known before the file arrives.
What you gain: A steadier reading experience.

CNT-10How much of the page is words

What we check: We compare useful visible text with templates, navigation, code and non-content elements.
Why it matters: A page can be technically large while saying very little about the service or customer’s problem.
What good looks like: The page’s primary subject receives enough specific, customer-relevant explanation.
What you gain: More substance and less template-shaped emptiness.

CNT-11Spelling

What we check: We scan visible content for likely spelling errors and review ambiguous terms in context.
Why it matters: Errors can distract visitors and weaken an otherwise professional presentation.
What good looks like: Business names, locations and industry terms are handled intentionally, with genuine mistakes corrected.
What you gain: Cleaner communication and fewer avoidable credibility hits.

CNT-12Grammar

What we check: We flag likely grammatical problems, fragments or agreement issues for human review.
Why it matters: Automated grammar tools are imperfect, but they can surface copy that deserves another look.
What good looks like: Sentences are clear and correct while preserving the business’s natural voice.
What you gain: Polished copy that still sounds like a real company.

CNT-13Reviews on your site

What we check: We look for customer reviews, testimonials or comparable first-party proof presented on the reviewed pages.
Why it matters: Visitors do not have to rely only on what the business says about itself. Specific, credible customer evidence can answer doubts about quality, reliability and fit.
What good looks like: Genuine, current proof is placed near the decisions it supports and is attributed appropriately. The site does not manufacture reviews or present vague praise as verified fact.
What you gain: More confidence at the moments when a visitor is deciding whether to call, request a quote or keep comparing providers.

CNT-14Links to other sites still work

What we check: We follow external links in the reviewed content and identify destinations that no longer respond or no longer lead where the page suggests.
Why it matters: Links to manufacturers, professional organizations, financing partners, government resources or social profiles can become stale without creating an obvious error on the website itself.
What good looks like: External links still reach the intended credible destination, use secure addresses when available and are removed or updated when the referenced resource changes.
What you gain: Fewer dead ends and more dependable supporting information for customers doing deeper research.

CNT-15How much your whole site says

What we check: We total the readable text across every page we crawl, and flag pages that carry very little of it.
Why it matters: A page with almost no text gives a visitor little to act on and a search engine little to understand. Thin pages are rarely a deliberate choice — they are usually a page that was started and never finished, or a design where the content lives inside images.
What good looks like: Each page answers the question it exists to answer, in words a person can read and a search engine can index. Substance comes from saying something useful, not from padding to reach a length.
What you gain: Visitors find answers without having to call to get them, and your pages have enough substance to be found in the first place.

CNT-16Pages for what buyers search

What we check: We compare the services buyers in your trade commonly search for against the pages and headings your site actually has.
Why it matters: People search for the specific job they need done, not for a category. A business that does the work but never names it in writing is invisible at exactly the moment someone is looking for it.
What good looks like: The services you want to be hired for each have a page or a clear heading that names them in the words customers use. You are not claiming work you do not do.
What you gain: You appear in the searches that matter for the jobs you actually want, rather than only for your business name.

CNT-17A headline that says what and where

What we check: We read the main headline and page title to see whether they state what the business does and where it works.
Why it matters: A visitor arriving from a search result decides in seconds whether they are in the right place. A headline that shows only a business name asks them to work that out, and many will not.
What good looks like: The first thing on the page makes the trade and the service area plain, in ordinary language, without becoming a list of keywords.
What you gain: Fewer visitors leave in the first few seconds because they could not tell whether you do what they need, near where they live.

CNT-18The towns you serve, in writing

What we check: We look for the towns and areas you serve, written in the page text rather than only implied by an address or a map.
Why it matters: Someone two towns over cannot tell whether you travel to them. Search engines have the same problem: an area that is never named in writing is an area you are not clearly associated with.
What good looks like: The places you actually serve are named where a customer would look for them, honestly. Listing towns you do not travel to produces calls you do not want.
What you gain: Fewer wasted enquiries from outside your area, and a clearer answer for the people inside it.

CNT-19How current your site feels

What we check: We look for dates on content — post dates, time tags, article metadata — that would let a visitor tell how recent the information is.
Why it matters: A site with no signal of recency invites the question of whether the business is still active. This is a matter of reassurance rather than of ranking, and it matters most to a visitor deciding whether it is worth calling.
What good looks like: Content that ought to be current carries a date, and that date is real. A page with nothing time-sensitive on it does not need one.
What you gain: A visitor can tell at a glance that there is somebody behind the website.

CNT-20Pages that repeat themselves

What we check: We compare the text of the pages we crawl and identify pages that duplicate each other closely.
Why it matters: Near-identical pages usually come from copying one page to make another and changing only the town or service name. They give a visitor nothing new, and they leave a search engine choosing between pages that make the same case.
What good looks like: Every page earns its place by saying something the others do not — even when pages cover similar services or nearby areas.
What you gain: Each page can do its own job, instead of competing with a near-copy of itself.

6. Housekeeping

Are the behind-the-scenes essentials working cleanly and consistently?

A website maintenance toolkit surrounded by contact, analytics, privacy, license and error-recovery checks.

Housekeeping covers the small pieces that make a website easier to operate, measure, share and maintain.

Most of these items are not dramatic on their own. Together, they distinguish a carefully managed website from one that was published and forgotten.

The practical win: Cleaner foundations. Better measurement. Fewer surprises.

20 checks in this dimension.

HKG-01Renders on a phone

What we check: We confirm that the page adapts to a mobile viewport without major clipping, overlap or horizontal scrolling.
Why it matters: A page can load successfully while still being frustrating or impossible to use on a phone.
What good looks like: Content reflows naturally and critical actions remain visible and tappable.
What you gain: A more reliable baseline across common screen sizes.

HKG-02Google’s own quality score

What we check: We capture Google Lighthouse or PageSpeed categories used in the report and preserve the underlying results.
Why it matters: A repeatable third-party tool provides a useful baseline and makes before-and-after comparison possible.
What good looks like: Strong scores supported by healthy underlying metrics—not a number improved through cosmetic shortcuts.
What you gain: Independent evidence that can be rerun after changes.

HKG-03Errors behind the scenes

What we check: We look for significant browser-console errors during normal page loading and interaction.
Why it matters: Some errors are harmless noise. Others signal broken integrations, missing files or functionality that may fail only in certain conditions.
What good looks like: Critical errors are resolved and remaining warnings are understood.
What you gain: Fewer hidden problems and easier future troubleshooting.

HKG-04Built to a current standard

What we check: We review the document structure, declarations and compatibility signals expected of a modern web page.
Why it matters: Outdated or malformed foundations can create inconsistent behavior and complicate maintenance.
What good looks like: Current standards-based markup with intentional handling of any necessary legacy support.
What you gain: A cleaner foundation for browsers, tools and future developers.

HKG-05Accents and symbols display correctly

What we check: We confirm that the page declares and uses a character encoding capable of displaying names, punctuation and symbols reliably.
Why it matters: Incorrect encoding can turn apostrophes, dashes or accented names into garbled characters.
What good looks like: Text displays consistently across pages and devices.
What you gain: Fewer strange characters and a more polished presentation.

HKG-06The little icon on the browser tab

What we check: We look for a favicon and related site icons.
Why it matters: The icon helps visitors recognize the site among open tabs, bookmarks and device shortcuts.
What good looks like: A crisp, recognizable icon is provided in appropriate formats and sizes.
What you gain: Small but consistent brand recognition.

HKG-07What happens on a broken link

What we check: We intentionally request an invalid address and review the resulting 404 response.
Why it matters: Mistyped, moved and outdated links are inevitable. A generic or broken error page leaves visitors stranded.
What good looks like: The server returns the correct error status and the page offers a clear way back to useful content.
What you gain: Fewer dead ends and better recovery from mistakes.

HKG-08Your sitewide link-preview fallback

What we check: We review the sitewide defaults used when a platform creates a link preview.
Why it matters: Pages without their own settings often inherit global metadata, so the fallback needs to be intentional.
What good looks like: Default title, description and image information represent the business accurately.
What you gain: More consistent sharing across pages and platforms.

HKG-09Link previews on X

What we check: We look for the metadata commonly used to create X/Twitter card previews.
Why it matters: Some platforms use overlapping but not identical preview conventions.
What good looks like: Shared links have the information needed for an appropriate card when the platform supports it.
What you gain: Fewer incomplete or image-less social previews.

HKG-10Counting your visitors

What we check: We look for a functioning analytics implementation and obvious duplicate or incomplete tracking.
Why it matters: Without dependable measurement, it is difficult to understand which pages and marketing efforts produce useful activity.
What good looks like: The business collects the data it genuinely needs, tests key events and avoids unnecessary collection.
What you gain: Better evidence for future decisions.

HKG-11Connected to Google Search Console

What we check: We confirm the intended site property is connected or identify that connection as a setup item.
Why it matters: Search Console provides direct information about indexing, search appearance and technical issues.
What good looks like: The correct property is verified, accessible to the owner and reviewed periodically.
What you gain: A direct view into how Google sees the site.

HKG-12Forms that catch mistakes

What we check: We review whether forms identify required fields and provide understandable validation when information is missing or malformed.
Why it matters: Silent failure and vague error messages cost inquiries and frustrate visitors.
What good looks like: Errors are explained near the affected field, preserve entered information and do not rely on color alone.
What you gain: More successful form submissions and fewer customer mistakes.

HKG-13Cookie notice

What we check: We identify whether the site uses a cookie or consent notice and whether it matches the technologies observed.
Why it matters: Consent obligations depend on the site’s data practices, audience and applicable law. A banner is not automatically required everywhere, and a generic banner does not create compliance by itself.
What good looks like: The implementation reflects actual tracking behavior and guidance from qualified legal counsel where needed.
What you gain: A more honest and intentional privacy experience.

HKG-14Sitemap pointed to from robots.txt

What we check: We look for a sitemap reference in robots.txt when the site uses an XML sitemap.
Why it matters: The reference gives compliant crawlers a straightforward place to find the site’s page inventory.
What good looks like: The address is correct, secure and points to a valid current sitemap.
What you gain: Cleaner crawler guidance and easier maintenance.

HKG-15Language consistently declared by the template

What we check: We confirm that the page template consistently outputs the intended document language.
Why it matters: Language declaration supports assistive technology, translation tools and consistent browser processing.
What good looks like: The value is valid and matches the actual primary language of each page.
What you gain: More reliable behavior across tools and templates.

HKG-16Your phone number in the footer

What we check: We look for a consistent, usable business phone number in the sitewide footer when phone contact is part of the business model.
Why it matters: Visitors who reach the bottom of a page should not need to hunt for the next step.
What good looks like: The correct number is visible and consistent across the site. HKG-18 separately tests whether visible numbers behave as tap-to-call links on phones.
What you gain: One more dependable path from research to contact.

HKG-17Privacy policy

What we check: We look for an accessible privacy policy and compare its presence with the site’s observable forms, analytics and data collection.
Why it matters: Visitors deserve understandable information about how their data is handled. Exact legal obligations vary by jurisdiction and business practice.
What good looks like: A current policy based on the business’s real practices, prepared or reviewed with appropriate legal guidance.
What you gain: Greater transparency and a clearer privacy foundation.

HKG-18Phone number you can tap

What we check: When phone contact is relevant, we confirm that visible phone numbers use a working tap-to-call link on mobile devices.
Why it matters: A mobile visitor should not have to memorize, copy or retype the number shown on the same screen.
What good looks like: The displayed number is correct, clearly recognizable and connected to the matching telephone link wherever it functions as a contact action.
What you gain: A shorter path between intent and a phone call, especially for customers searching while away from a desk.

HKG-19A way to contact you

What we check: We look for at least one clear, working contact path appropriate to the business, such as a phone number, inquiry form, booking action or published email address.
Why it matters: A persuasive website cannot produce an inquiry if a ready visitor cannot find or complete the next step.
What good looks like: Contact options are visible where decisions occur, lead to a working destination and set an honest expectation about what happens next.
What you gain: Fewer interested visitors reaching the end of their research without a practical way to respond.

HKG-20License number on display

What we check: For businesses and locations where a professional or trade license is relevant, we look for the required or expected license identifier in an appropriate visible location.
Why it matters: A license number can be both a regulatory disclosure and a trust signal, but the exact obligation depends on the trade, jurisdiction and use of the page.
What good looks like: The number is current, accurate and displayed consistently where customers or regulators would reasonably expect it. When no public license disclosure applies, the check is marked not applicable.
What you gain: Clearer evidence that the business is legitimate and fewer avoidable disclosure gaps.

How the checks work together

No single score tells the entire story.

A fast website can still be confusing. A well-written site can still be inaccessible. A secure connection cannot rescue a page that gives visitors no reason to trust the company. Strong search markup cannot compensate for a service page that says almost nothing.

The checks are useful because they expose different kinds of friction:

Visibility friction: Can the right person discover the page?
Understanding friction: Can they quickly tell what the business does?
Trust friction: Do the site and company feel credible?
Access friction: Can different people use the content and controls?
Action friction: Can a ready customer call, book or request a quote?
Operational friction: Can the site be measured, maintained and improved?

Six evidence families combine into the customer journey from discovery and understanding to trust and action.

The goal is not to chase 100s in isolation. It is to build a website whose pieces support one another.

What changes after a rebuild?

The strongest use of the Website Evidence Report is comparison. Before work begins, the report records the existing site’s scores, findings and check-level results. After the new site is built, the same repeatable checks run again.

That creates a before-and-after record:

What was preserved
What was improved
What remains a tradeoff or future opportunity
Whether the new site meets the agreed technical baseline

The same website and evidence report before and after focused improvements, with remaining tradeoffs still documented.

Not every number will move in a straight line. Tools change, network conditions vary and third-party services can affect results. That is why the report preserves the date, tested pages, tool versions and relevant conditions.

The point is not to manufacture a perfect after-score. The point is to prove that the work addressed the evidence it was designed to address.

What Evident Sites handles

When Evident Sites builds a website, these checks are not handed to the client as a technical to-do list. The relevant work is incorporated into strategy, design, content, development, launch and ongoing maintenance. That can include:

Structuring pages around real customer questions
Making phone and inquiry paths obvious
Building responsive layouts for phones and desktops
Optimizing images and loading behavior
Creating meaningful page titles, descriptions and headings
Implementing accessible labels, landmarks and navigation patterns
Configuring secure delivery and appropriate browser protections
Connecting measurement and search tools
Testing forms, errors, redirects and sitewide details
Rerunning the evidence checks before launch

Some decisions still belong to the business owner. Content accuracy, service details, privacy practices, legal language and business claims require owner participation and, where appropriate, professional legal advice. That division of responsibility should be clear before a project begins.

A website does not need to be perfect to be useful

A local service website is not a laboratory experiment. It is a working part of a real business.

Sometimes the right decision is a focused repair. Sometimes the existing site has a strong foundation and needs better content. Sometimes the design has reached the point where rebuilding is more sensible than stacking another patch on top.

The Evidence Report is meant to help distinguish those situations. If the site already supports the business well, the evidence should say so. If it does not, you should be able to see why.

Want to see what the evidence says about your website?

Get a Website Evidence Report: a plain-language review of what is working, what deserves attention and whether a focused refresh or rebuild is likely to be the better investment.

No scare tactics. No obligation. No technical homework.

Request your Website Evidence Report

Sources and standards referenced

These checks are grounded in published guidance rather than house opinion. Where a threshold is an Evident Sites judgement rather than a standard, the report says so.

A tool score is not a universal law. Nothing here promises rankings, legal compliance, complete accessibility or complete security.