Website Audits
What Is a Website Opportunity Score and How Do You Use It to Prioritise Leads?
A website opportunity score is a number from 0 to 100 that estimates how big and how actionable the case is for improving a business's website. A higher score means a stronger opportunity for someone who sells web design, local SEO or conversion work. In SiteOpportunity it always comes with a second, separate number, the Confidence score, which says how much you can trust the first one.
This guide explains what goes into each score, what the labels mean, how to read the two together, and how to use them to decide who to contact first. It also covers what a score cannot tell you, because that is where most mistakes happen.
What the Opportunity Score measures
The score answers one question: how much observable, fixable room for improvement is there on this business's website, and is the business worth helping? It adds up four parts, and together they total 100 points.
- Technical gap40%
- Conversion gap25%
- Visual modernisation gap20%
- Commercial fit15%
Source: SiteOpportunity scoring model
Each part looks at something different.
| Part | Points | What feeds it | Plain-English question |
|---|---|---|---|
| Technical gap | 40 | 18 checks: reachability, HTTPS, redirects, mobile viewport, PageSpeed scores, layout shift, blocking time, titles and descriptions, broken pages, outdated indicators, page weight, mobile usability | Is the site built and running in a way that hurts visitors or search? |
| Conversion gap | 25 | 10 checks: visible call to action, phone number, contact or booking link, lead form, service explanation, trust signals, testimonials, location, navigation, above-the-fold message | Can a ready-to-buy visitor actually become a customer? |
| Visual modernisation gap | 20 | A review of desktop and mobile screenshots across six dimensions: hierarchy, typography, colour consistency, mobile presentation, design age and conversion clarity | Does the design help or hurt trust and clarity? |
| Commercial fit | 15 | 6 checks: review count, rating, category, public contact details, signs of an active business, and how weak the site is compared with the reputation | Is this a business worth investing in and reachable? |
The technical part carries the most weight because problems there are measurable and usually easy to prove. For plain-English examples of what these problems look like on real sites, see 15 website problems local businesses ignore. If you want the full list of checks and free ways to run them by hand, read the website audit checklist for local businesses.
Why commercial fit counts
Commercial fit is not about the website at all, and that is deliberate. A business with excellent reviews and a weak website may be a better prospect than one with no reviews and no visible activity. The reviews prove demand. The website is the leak. Without a commercial-fit part, a score would reward any bad site, including sites for businesses that have no customers to lose.
How the visual part is scored
The visual part is the only one that starts from a model's reading of screenshots. The six dimensions are each scored as a quality level, then turned into gap points by the same fixed formula as everything else. The model does not choose the final score. If the visual review cannot run, for example because a screenshot fails, the system uses a middle value for that part instead of assuming the best or worst case, and the Confidence score drops to reflect the missing evidence.
What the score labels mean
Scores come with labels so that you do not have to interpret a raw number. The labels describe the size of the opportunity, never the quality of the business.
| Score | Label | What to do with it |
|---|---|---|
| 75 to 100 | Priority Opportunity | Contact first. Lead with the strongest evidenced finding |
| 60 to 74 | High Opportunity | A good second wave. Worth a specific, personal message |
| 40 to 59 | Potential Opportunity | Worth a look if commercial fit is strong or the fix is small and clear |
| 0 to 39 | Low Opportunity | Skip it or park it. The site is probably doing its job well enough |
The labels are deliberately neutral. There are no words like "ugly" or "failed" anywhere in them, and you should not use such words in outreach either. The score describes an opportunity for you. It is not a verdict on a business.
What the Confidence score measures
Confidence is a separate 0 to 100 score that measures data completeness and reliability. It does not measure website quality. A perfectly good website can have high Confidence, and so can a broken one. Confidence only asks: how much of the analysis did we actually manage to see?
It is built from evidence checks. Each of these six checks adds 15 points when it succeeds, for 90 in total.
| Evidence check | What it confirms |
|---|---|
| Website reachable | The site loaded when tested |
| Desktop screenshot captured | The desktop view was seen |
| Mobile screenshot captured | The mobile view was seen |
| HTML extracted | The page content could be read |
| PageSpeed audit completed | The speed and quality tests ran |
| Visual analysis validated | The screenshot review passed its checks |
The last 10 points come from business-data completeness: whether the analysis had the business's rating, review count, phone number and category.
| Confidence | Label | How to treat it |
|---|---|---|
| 80 to 100 | High Confidence | Findings are well supported. You can be specific |
| 60 to 79 | Good Confidence | Solid, but skim the evidence before you use it |
| 40 to 59 | Limited Confidence | Treat findings as leads to verify yourself |
| 0 to 39 | Low Confidence | Do not assert findings. Use a research-first message |
SiteOpportunity builds this behaviour into its outreach drafts. Below 60, it shows a warning and softens wording. Below 40, it does not write assertive claims and prepares a research-first opener instead. The point of Confidence is that missing evidence must never look like a strong opportunity.
Reading the two scores together
A single number would hide the most important thing: how sure you can be. Two scores let you sort prospects into four groups.
| High Confidence | Low Confidence | |
|---|---|---|
| High Opportunity | Contact now with a specific finding | Verify by hand first, then decide |
| Low Opportunity | Skip. The site is fine and the evidence is solid | Ignore, or re-run the analysis later |
Here are three fictional businesses and how you might treat each.
Harbor View Chiropractic
78
Opportunity
88
Confidence
- No visible booking button on mobile
- Phone number is plain text, not tappable
- Strong review base but no reviews shown on the site
Harbor View is a priority. The evidence is strong, the findings are specific, and you can write a confident, concrete message.
Pine Ridge Auto Repair
71
Opportunity
45
Confidence
- Mobile screenshot failed to load
- Speed test did not complete
- Strong rating and review count
Pine Ridge looks promising, but half of the evidence is missing. Open the site yourself before you say anything to the owner. If you confirm the problems by hand, it becomes a strong prospect. If you cannot, it is a research-first email at best.
Lakeside Bakery
22
Opportunity
90
Confidence
- Site loads fast and works on mobile
- Clear contact details
- Modest review count
Lakeside has solid evidence that its website works. Do not spend time on it.
How to use the scores to prioritise a list
Sort by Opportunity
Start with the highest scores. In a big market, look at the Priority and High groups first.
Filter by Confidence
Set aside anything with Limited or Low Confidence for a manual check, so you never pitch on weak evidence.
Read the findings
Open each prospect and read the evidence. The score gets you to the business, and the findings give you the message.
Check the recommended service
Each result suggests a primary service, such as mobile optimisation or a booking build. Make sure it is something you sell.
Pick a batch you can finish
Choose the number of prospects you can contact well this week, and start there. Come back to the rest.
Two habits make the difference. First, do not quote a score to a prospect. "Your website scored 34" is a verdict. "I couldn't find a way to book on my phone" is an observation they can check. Second, use the score to choose whom to contact and the findings to choose what to say. If you want to turn a finding into a pitch, how to turn website problems into a sales pitch shows how, and the guide to cold email templates for local businesses covers the email itself.
What a score cannot tell you
Scores save time, but they do not replace judgment.
- Budget and intent. A high score shows a fixable gap. It does not show a willingness to pay for it.
- Who decides. The best prospect on paper is no use if you cannot reach the person who decides.
- Timing. Sites change. An analysis is a snapshot, and one credit covers a fresh one. Valid cached results within the recent window are reused without using another fresh-analysis credit, so an older result may not reflect a change made this week.
- Taste. Two people can look at the same design and disagree. That is why the labels describe opportunity rather than beauty.
A rough version you can do by hand
You do not need software to think in these terms. For a small list, rate each part yourself, using the same maximums.
| Part | Maximum | How to rate it |
|---|---|---|
| Technical gap | 40 | Near 0 if the site is fast, secure and clean. Around 20 with a few clear problems. Near 40 if it is slow, insecure or broken on phones |
| Conversion gap | 25 | Near 0 if calling and booking are easy. Around 12 if the path exists but is weak. Near 25 if a visitor cannot easily act |
| Visual gap | 20 | Near 0 if the design is clear and trustworthy on both devices. Near 20 if it hurts clarity and trust |
| Commercial fit | 15 | Near 0 for a business with no reviews or activity. Near 15 for a well-reviewed, active business with public contact details |
Add the four and compare the total with the labels above. It will not match an automated score, but it will rank a short list sensibly. For where the underlying checks come from, see how to find local businesses that need a new website.
Where SiteOpportunity fits
SiteOpportunity computes both scores for every business in a search, shows the findings and screenshots behind them, and recommends a primary service based on fixed rules. The scoring uses deterministic rules, and each result keeps its sources so that you can see why it scored the way it did. You review the evidence and decide what to do. Nothing is sent automatically. The features page shows what a result looks like, and the FAQ covers the most common questions about scores and credits.
Want to see Opportunity and Confidence side by side on a real market? Run a free search
Questions about the scores
Is a high Opportunity Score a guarantee that a business will buy?
No. The score estimates how big the website-improvement case is. It does not measure budget, intent or willingness to reply, and it is not a guarantee of clients, rankings or revenue.
What is a good Opportunity Score?
Scores of 60 and above are labelled High or Priority Opportunity, so they are worth contacting first. Scores from 40 to 59 are worth a look when the fix is clear. Below 40, the website is probably doing its job.
Why is my Confidence score low?
Confidence drops when evidence is missing, for example a screenshot that did not load, a speed test that did not finish, or thin business data. A low score means verify by hand, not that the website is bad.
Does AI decide the score?
No. The final numbers come from fixed rules. A model reads the screenshots and rates six visual dimensions, and those ratings go through the same formula as the other parts.
Should I show a prospect their score?
It is better not to. A number sounds like a verdict. Describe one specific thing you observed, such as no visible booking button on mobile, so the owner can check it for themselves.