Website Audits
Website Audit Checklist for Local Businesses: 34 Signals That Show a Real Opportunity
A useful website audit for a local business answers one question: what is stopping this site from bringing in customers, and can you prove it? The checklist below covers 34 signals in three groups (technical, conversion and commercial), plus a short visual review of screenshots. You can run all of it with a phone, a browser and PageSpeed Insights, and you can finish a single site in 15 to 20 minutes once you know the order.
It is written for people who sell websites, local SEO or conversion work to local businesses. Every signal comes with a free way to check it and a note on what it might mean for the business, so you leave with findings you can act on, not a long list of scores.
How to run this audit
Set aside a browser and a phone. Work in this order, because each step is faster than the one after it and often ends the audit early if you find something clear. For a shorter, time-boxed version, see how to run a free website audit before you pitch.
Use the site like a customer
Open it on your phone. Look for what the business does, how to call or book, and where it is. This covers most of the conversion group.
Check the basics that break trust
Reachability, HTTPS, the title in the tab, the footer year. These take under a minute.
Run PageSpeed Insights
Test the homepage on mobile, then desktop. Save the scores and the top issues it lists.
Read the listing
Open the business's Google listing for the rating, review count and contact details. This covers the commercial group.
Review the visuals
Take a mobile and a desktop screenshot and judge design, using the six questions in the visual section.
Write the finding
Pick the strongest one or two problems. Attach a screenshot, one plain sentence and the service it points to.
Here is how the 34 signals divide up.
- Technical (18)18
- Conversion (10)10
- Commercial (6)6
Source: SiteOpportunity signal model
The 18 technical signals
Technical signals describe how the site is built and how it performs. They matter for search, for speed and for basic trust. Check each with the free method shown.
| # | Signal | How to check it for free | Why it matters |
|---|---|---|---|
| 1 | Website reachable | Open the address on a phone and a laptop, with and without "www" | A site that will not load loses every visitor and every search visit |
| 2 | HTTPS enabled | Look for the padlock. Type the address with http:// and see if it redirects | Chrome has labelled plain HTTP sites "not secure" for years, which hurts trust before anyone reads a word |
| 3 | Excessive redirect chain | Open the browser's Network tab and watch how many hops before the page loads | Each hop adds delay. Three or more is worth noting |
| 4 | Mobile viewport tag | View the page source and search for name="viewport" | Without it, phones show a shrunken desktop page |
| 5 | Mobile performance score | PageSpeed Insights, mobile tab | The headline pointer for how the site feels on a phone |
| 6 | Accessibility score | PageSpeed Insights, Accessibility category | Shows automated accessibility problems, and only some of them |
| 7 | SEO score | PageSpeed Insights, SEO category | Flags missing basics such as titles, descriptions and crawlable links |
| 8 | Best-practices score | PageSpeed Insights, Best Practices category | Catches security and code-quality issues |
| 9 | Slow Largest Contentful Paint | PageSpeed Insights metric list | The main content takes too long to appear |
| 10 | High layout shift | PageSpeed Insights metric list | Content jumps while loading, so visitors tap the wrong thing |
| 11 | Excessive blocking time | PageSpeed Insights, Total Blocking Time | The page freezes while scripts run |
| 12 | Missing page title | Read the browser tab, or view the source for the title tag | The title is how the page appears in search and in tabs |
| 13 | Missing meta description | View the source for the meta description tag | Google may use it as the snippet, so a missing or generic one gives the business less control over how the listing reads |
| 14 | Broken or empty homepage content | Load the homepage and check for blank areas, error text or thin content | A site that looks broken or empty fails on first impression |
| 15 | Old copyright year | Read the footer | A weak but visible sign that nobody has touched the site |
| 16 | Outdated technology indicators | Look for obvious signs such as Flash content or a layout that breaks on a phone | Only report what you can prove |
| 17 | Very large page weight or slow resources | Network tab total transferred, or the "enormous network payloads" audit | Heavy pages hurt visitors on slower mobile connections |
| 18 | Mobile usability issue | Use the site on a phone: tiny text, buttons too close together, sideways scrolling | These frustrate visitors even when scores look fine |
Reading the speed numbers correctly
Speed is where audits go wrong, because tools show many numbers and it is tempting to quote them all. Keep it simple.
The three Core Web Vitals are Largest Contentful Paint (LCP), Interaction to Next Paint (INP) and Cumulative Layout Shift (CLS). Google's targets, assessed at the 75th percentile of real page loads, are LCP within 2.5 seconds, INP of 200 milliseconds or less, and CLS of 0.1 or less.
The overall Lighthouse performance score is a separate thing. Chrome's Lighthouse documentation puts 0 to 49 as poor, 50 to 89 as needing improvement and 90 to 100 as good, and builds the number from a weighted mix of lab metrics.
Source: Chrome for Developers, Lighthouse performance scoring documentation
Two things follow from this. First, blocking time, the measure of how long a page is frozen by scripts, carries the biggest single share of the score, so a script-heavy site can score poorly even if it looks fast. Second, PageSpeed Insights shows field data (real Chrome users over 28 days) only when a site has enough traffic to appear in that dataset. Many small local sites do not, so you will often be reading lab data from a simulated mid-range phone. Say that when you report it: "In a mobile lab test, the page took X seconds to show its main content."
Does speed really matter for sales? A Google and Deloitte study of 37 brands and more than 30 million sessions found that a 0.1 second mobile speed improvement across the whole journey lifted retail conversions by 8.4% and average order value by 9.2%, and the lead-generation brands saw a 21.6% improvement in progression toward form submissions.
The 10 conversion signals
Conversion signals ask a plain question: can a visitor who wants to become a customer actually do it? These are the fastest checks and the ones business owners understand at once.
| # | Signal | How to check it | What a failure looks like |
|---|---|---|---|
| 1 | Primary call to action visible | Open the homepage on a phone without scrolling | No clear "Call", "Book" or "Get a quote" button above the fold |
| 2 | Phone number visible | Look in the header and footer. Tap it | Number buried, or plain text that does not start a call |
| 3 | Contact or booking link available | Find a "Contact" or "Book" link in the menu | You have to hunt for a way to get in touch |
| 4 | Lead form available | Look for a short form on the homepage or contact page | No form, or a long form that asks for too much |
| 5 | Service explanation visible | Can you tell what services are offered in a few seconds? | Vague copy that never says what the business does |
| 6 | Trust signals visible | Look for licences, certifications, awards, guarantees, real photos | Nothing that says "this is a real, reliable business" |
| 7 | Testimonials or review proof | Look for customer quotes or a link to reviews | Great ratings elsewhere, none shown on the site |
| 8 | Location information visible | Look for an address, map or service area | Visitors cannot tell whether the business serves their area |
| 9 | Clear navigation | Try to reach services, about and contact in two taps | Menu is confusing, crowded or broken on mobile |
| 10 | Above-the-fold value proposition | Read only the first screen, then ask what the business offers and why choose it | Generic headline such as "Welcome to our website" |
Use the five-second test on the first screen: look for five seconds, close the tab, then write down what the business does and what you would do next. If you cannot answer, a first-time visitor cannot either.
Not every failure carries the same weight. A missing phone number on a plumber's mobile site is a serious problem, because many customers search and call. A missing testimonials section on a business with plenty of other trust signals is minor.
The 6 commercial signals
Commercial signals do not judge the website at all. They judge the business behind it and whether an improvement would be worth its while.
| # | Signal | Where to find it | Why it matters |
|---|---|---|---|
| 1 | Google review count | The business's Google listing | A large review base means real customers and real demand |
| 2 | Google rating | Same listing | A healthy rating means a business worth investing in |
| 3 | Business category | The listing's category | Some categories rely on the website more than others |
| 4 | Public business contact details | Listing and website | You need a public, business-facing way to reach them |
| 5 | Active-looking local business | Recent reviews, photos, posts, opening hours | Avoids pitching a business that has quietly closed |
| 6 | Website weakness versus reputation | Compare the reviews with the audit | The gap between the two is often the pitch |
The last signal matters most. A business with excellent reviews and a weak website may be a better prospect than a business with no reviews and no visible activity, because the reviews prove demand that the website is not capturing. That is the principle behind the method for finding prospects.
The visual review: six questions
Numbers cannot tell you whether a page looks trustworthy. For that, take one desktop and one mobile screenshot and answer six questions about what you can see.
- Visual hierarchy. Does the eye go to the most important thing first, or does everything compete?
- Typography. Is the text readable at normal size on a phone, with enough contrast and spacing?
- Colour consistency. Do the colours look deliberate and consistent, or patched together?
- Mobile presentation. Does the layout adapt to the phone, or is it a squeezed desktop page?
- Design age. Do the layout and imagery look several years old in a way that affects trust?
- Conversion clarity. Is the next step obvious from the design alone?
The key rule is to separate observation from inference. "The header phone number is 11 pixels tall on mobile" is an observation. "The site looks unprofessional" is an inference, and an unhelpful one. Report what you can see and describe the effect on the visitor. Never call a business's site ugly. It insults the owner you are trying to help, and it is not evidence.
Accessibility deserves a special mention in this review, because owners rarely think about it and it affects real visitors. The WebAIM Million analysis of the top million home pages found detectable WCAG 2 failures on 95.9% of them. These are the six most common issues.
Source: WebAIM Million, latest report. Share of home pages with each failure.
Because these problems are so common, they are easy to explain without blaming anyone, and low contrast text is quick to spot by eye. Automated tools find only part of the accessibility problem, so describe a score as a starting point and not as proof of compliance.
A worked example
Here is how the audit looks on a fictional business, Willow Creek HVAC, a heating and cooling company with a strong review base.
| Signal | What you observe | Evidence | Severity |
|---|---|---|---|
| Phone number visible (conversion) | Number appears only in the footer, as plain text | Mobile screenshot, address bar visible | High |
| Primary call to action visible (conversion) | No button above the fold on mobile | Same screenshot | High |
| Mobile performance score (technical) | Score in the 30s in a mobile lab test | PageSpeed Insights screenshot | Medium |
| Missing meta description (technical) | Search listing shows a random text snippet | Google search screenshot | Low |
| Google review count and rating (commercial) | Hundreds of reviews and a high rating | Listing screenshot | Supports the pitch |
| Website weakness versus reputation (commercial) | Strong reputation, weak mobile site | Combination of the above | Supports the pitch |
Now turn it into one finding. Do not send the whole table.
That is one problem, in plain words, tied to a business result, with a screenshot behind it. The next step, turning it into a short email, is covered in the guide to writing a cold email about a website.
Turn findings into services
Each strong finding maps to something you can offer. Use this as a quick reference.
| If you find | Offer |
|---|---|
| No visible booking or lead path | Booking or lead-form implementation, conversion-focused landing page |
| Slow mobile performance | Website speed optimisation |
| Broken layout on phones, tiny tap targets | Mobile website optimisation |
| Missing titles and descriptions, weak local presence | Local SEO improvement |
| Great reviews absent from the site | Trust and credibility upgrade |
| Serious accessibility failures | Accessibility improvement |
| Dated look, weak hierarchy, inconsistent colours | Branding and visual refresh |
| Unclear service and value messaging | Content and messaging improvement |
| Problems in all of the above | Full website redesign |
Lead with the smallest fix that solves the clearest problem. 15 website problems local businesses ignore shows what several of these findings look like in practice. A full redesign is a bigger conversation, and often a smaller first project earns the trust for it.
The 15-minute audit, as a checklist
Copy this and use it on every prospect
- Open the site on a phone. Can you see what the business does, and how to call or book, without zooming?
- Tap the phone number. Does it start a call?
- Check for the padlock and that http:// redirects to https://
- Note the title in the browser tab and the description in the search result
- Read the footer copyright year
- Run PageSpeed Insights on mobile. Save the score, the LCP time and the top two issues
- Scroll the homepage once looking for a form, trust signals and testimonials
- Open the Google listing. Note the rating, review count and recent activity
- Take one mobile and one desktop screenshot
- Write one finding: what you saw, the effect on visitors, the service it points to
Mistakes that make audits useless
- Sending the whole report. Busy owners read one line. Pick the strongest finding and lead with it.
- Mixing lab and real-user data. Say which you are quoting.
- Quoting a score with no context. "Score of 34" means little. "The main content takes several seconds to appear on a phone" means a lot.
- Treating automated checks as final. Tools miss things and occasionally flag things that are fine. Look at the page yourself.
- Judging taste as fact. "Ugly" is an opinion. "Text below readable size on mobile" is a finding.
- Auditing behind a login. Stick to public pages. Do not try to get around access controls.
Doing the audit at scale
By hand, this takes 15 to 20 minutes per business. For one niche and one city, that is fine, and a good way to learn what matters. Across many cities, it turns into a full-time job, and your checks drift from site to site.
SiteOpportunity runs the same signal groups for you. It measures the deterministic signals, reviews desktop and mobile screenshots for the visual questions, and shows the evidence behind each result. It also gives a separate Confidence score that reflects how complete the evidence was. If a screenshot failed or a test did not run, confidence drops rather than the tool guessing. See the feature overview for what each result contains, or the FAQ for how Opportunity and Confidence differ.
Want to see the same checks on a whole market at once? Try a free search
Questions about running the audit
Do I need paid tools to audit a local business website?
No. A phone, a laptop, PageSpeed Insights and the business's Google listing cover nearly every signal here. Paid tools save time when you audit many sites, but the free route gives you the same evidence.
Why do PageSpeed scores change between runs?
Lab tests run on simulated devices and networks, so results vary a little each time. Run the test two or three times and report the typical result, or a range, rather than a single number.
What if PageSpeed Insights shows no real-user data?
That is normal for small sites. Field data needs enough Chrome traffic to appear, so you will see lab data only. Say so in your notes and treat the score as a pointer.
Which signals should I mention first in outreach?
Lead with a conversion problem you can show on a phone, such as no visible booking path or a phone number that does not tap. Those are easy for an owner to see and clearly linked to lost customers.
Is it acceptable to audit a business's website without asking?
Looking at a public page is like visiting it as a customer. Stay on public pages, do not bypass logins or blocks, and be honest in outreach about what you saw and how.