Tag: web design

Analytics platforms present hundreds of metrics. Almost all of them are irrelevant to a business with one location and a phone.

Here is the short list that matters, the list you can ignore, and one question that outperforms all of it.

The question that beats any dashboard

Ask every new customer how they heard about you. Write down the answer.

That is it. That is the highest value measurement available to a small business, it costs nothing, and it captures things no analytics tool can see.

Analytics cannot tell you that somebody was recommended by a neighbour, saw your van, or remembered a job you did on their street four years ago. For a lot of local businesses those channels dominate, and they are entirely invisible in any dashboard.

Do it for three months, consistently, on every enquiry. Then count. You will almost certainly find that your customers come from somewhere other than where you assumed, and that finding usually changes where the effort should go.

The failure mode is inconsistency. If you ask half the time, the data is useless. Put it on the form, add it to the intake script, make it a field in whatever you use to track jobs.

The numbers worth watching

Enquiries per month. Calls, forms, messages, walk ins. One number, tracked monthly. This is the closest thing to a headline metric a small business has, and most do not track it at all.

Where they came from. From the question above.

Conversion from enquiry to customer. Of the people who got in touch, how many bought. If this is low, your marketing is working and something after the enquiry is not, which is a completely different problem and one that spending more on marketing will make worse.

Customer value. What a customer is worth over the time they typically stay with you, not just the first transaction. You are the only person who can supply this and it is what makes every other number meaningful.

Website visitors, roughly, and how many are on phones. You do not need precision. You need to know whether it is fifty a month or five thousand, because that determines whether website changes can plausibly matter.

Your Core Web Vitals. Specifically Largest Contentful Paint on mobile, against Google’s published 2.5 second threshold. Check it quarterly in PageSpeed Insights. It is free and it takes a minute.

Review count and recency on Google. How many, and how recent. Recency matters more than most people assume.

That is the whole list. Seven items, most of them checkable in a few minutes a month.

What to ignore

Bounce rate. Frequently misunderstood. A visitor who arrives, reads your phone number and calls you registers as a bounce. That is a success recorded as a failure.

Time on page. Ambiguous in the same way. Long could mean engaged or confused.

Impressions. How many people theoretically could have seen something. Interesting to nobody with a business to run.

Follower count. Vanity, unless your business model is genuinely audience based. A thousand followers who will never buy from you are worth less than eleven local customers.

Likes. Discussed elsewhere on this site. Not customers.

Keyword rankings in isolation. Ranking first for something nobody searches is not an achievement. Enquiries are the measure.

How to measure a change honestly

When you fix something and want to know whether it worked, three rules.

Change one thing at a time, or accept that you will not know which change did it. This is frustrating and it is how measurement works.

Give it long enough. For most small businesses, two weeks of data is noise. A month is minimum, a quarter is better, and if your enquiry volume is small, longer still.

Compare like with like. Same period last year, not last month. A landscaper comparing March to June has learned about spring, not about their website. This single error invalidates more small business conclusions than anything else.

The simplest system that works

You do not need software. A spreadsheet with one row per month and these columns:

Enquiries. Customers. Where they came from, as a tally. Anything notable that changed.

Fill it in on the first of the month. It takes ten minutes. After a year you have something genuinely valuable that no analytics platform could have given you, because it includes the offline sources that dominate local business.

If you want website numbers as well, install something simple and check it monthly. Do not check it daily. Daily numbers on a small business site are almost entirely noise and watching them produces anxiety rather than insight.

The honest limitation

Some things cannot be measured cleanly and you should not pretend otherwise.

Brand perception. Word of mouth. The cumulative effect of looking like a real business for years. Whether the person who chose your competitor did so because of your website or because their brother in law was available.

Marketing measurement for a small business is directional, not precise. Anybody offering exact attribution at your scale is overselling, and decisions made on false precision are worse than decisions made on honest estimates.

What you can know is whether enquiries went up, where people say they came from, and whether the things you fixed are still fixed. That is enough to steer by.

Start here

If you do nothing else from this article: ask every new customer how they heard about you, and write it down, for three months without exceptions.

That single habit will teach you more about your marketing than any dashboard, it costs nothing, and it works for exactly the channels that matter most to a local business and that no software can see.

Two free tools worth connecting

You do not need paid analytics, but two free things are worth the twenty minutes it takes to set them up.

Google Search Console. Shows what people actually searched before your site appeared, how often they clicked, and which pages get seen. This is real query data from real people, which is better than any keyword research, and a startling number of businesses have never connected it.

Your Google Business Profile insights. Shows how people found you, whether they searched your business name or your category, and what they did next. Name searches mean your reputation is doing the work. Category searches mean you are being discovered by strangers. The ratio between them tells you something useful about which part of your marketing is functioning.

Both are free, both take about ten minutes to connect, and both answer questions no amount of guessing will.

The review you should do quarterly

Once every three months, sit down for half an hour with your spreadsheet and ask four questions.

Did enquiries go up or down compared to the same quarter last year? Same quarter, not last quarter.

Where did customers say they came from? Has the mix changed?

What did I change in the last three months, and did anything move?

What is the one thing I would fix next?

That is the whole review. It is enough to steer a small business and it does not require any expertise, any software, or any consultant.

Why precision is the wrong goal

A closing thought, because the pressure to measure everything is strong and it is often unhelpful at this scale.

Large companies invest heavily in attribution because a one percent improvement on enormous volume is worth real money. At your scale, the effort required to measure precisely usually exceeds the value of the precision.

What you need is enough information to make the next decision. Are enquiries trending up or down. Where do people say they find me. Did the thing I fixed appear to help.

That is a low bar and it is genuinely sufficient. Businesses that chase perfect measurement tend to spend more time in dashboards than with customers, which is a poor trade at any size and a terrible one at yours.

Ask people how they heard about you. Write it down. Everything else is refinement.

Setting a baseline before you change anything

If you are about to fix your website, start asking for reviews, or begin posting regularly, spend twenty minutes first recording where you are today.

Enquiries in the last three months. Current review count and the date of the most recent one. Your mobile PageSpeed score and LCP figure. Roughly how many people visit your site monthly.

Write those five things down with today’s date.

This costs almost nothing and it is the difference between knowing whether something worked and guessing. The most common frustration I hear is somebody who invested in their presence and cannot tell whether it helped, and it always traces back to no baseline existing beforehand.

You cannot go back and measure the past. Twenty minutes now buys you the ability to answer the question in six months.

The one metric to watch if you only watch one

Enquiries per month, compared to the same month last year.

That single line, tracked consistently, will tell you more about whether your marketing is working than any dashboard, and it accounts for the seasonality that makes every other comparison misleading.

Everything else on this page is refinement on top of that one number.

---

Core Web Vitals thresholds, including the 2.5 second Largest Contentful Paint target measured at the 75th percentile, are published by Google at web.dev.

The Baymard Institute maintains a running meta analysis of cart abandonment research, collecting every credible study rather than running one and publishing a single figure. As of their September 2025 update, the average across 50 documented studies sits at 70.22 percent.

The number worth paying attention to is not really the average. It is the spread. The individual studies range from roughly 55 percent to 84 percent, and that range is the honest part, because it tells you abandonment varies enormously with what is sold, to whom, and how.

So roughly seven in ten people who add something to a cart do not buy it, and that has stayed broadly stable for years despite everything the industry has thrown at it. Some of that is unavoidable. People browse, compare, save things for later. But a meaningful share is caused by things on the page, and those are fixable.

Here is what a product page has to do.

1. Photographs that answer questions

Product photography is not decoration on an e-commerce page. It is the entire substitute for handling the object, and it is doing the job that a shop assistant and a shelf do in person.

Multiple angles, including the unflattering ones. People want to see the back, the underside, the seams. Hiding them creates suspicion, not mystery.

Scale. The single most common photography failure. A person cannot tell how big something is from an isolated shot on white. Show it in a hand, next to a familiar object, or being used. Returns caused by size surprises are expensive and almost entirely preventable here.

Detail shots. Texture, finish, the quality of construction. This is where you demonstrate that it is worth the price.

In context. The item in use, in a real setting. Isolated product shots on white are necessary and insufficient.

Consistent treatment. The same lighting and background across your catalogue. A store where every product was shot differently looks improvised.

2. A description that answers the actual question

Most product descriptions are written as marketing copy. They should be written as answers.

What is it made of. What size, precisely, in numbers. How does it behave over time. What is included. What is not included. How do I look after it.

Read your customer service messages and note what people ask before buying. Every one of those questions belongs on the page. If somebody had to ask, the page failed, and for every person who asked there were several who left instead.

Specificity outperforms enthusiasm consistently. “Eleven inches across, holds about a litre” does more than “generously sized.”

3. Price clarity, including the total

Baymard’s own checkout research points repeatedly to unexpected costs as a leading cause of abandonment, and it is worth taking seriously because it is one of the few causes entirely within your control.

The failure pattern is familiar: a customer builds a mental total through the process, then shipping and fees appear at the final step and the number jumps. That feels like a bait and switch even when nothing dishonest happened, and people leave over the feeling rather than the amount.

Show shipping cost, or the threshold for free shipping, on the product page. Show taxes if they apply. Show delivery timeframe. Every figure the customer will eventually see should be visible before they commit attention to checking out.

4. Stock and delivery, stated plainly

In stock or not. How long until it ships. How long until it arrives. Where you ship to.

Vagueness here is expensive because it is one of the last uncertainties before purchase. “Ships in 2 to 3 business days” is a small sentence that resolves a real hesitation.

5. Returns, said clearly and early

A clear returns policy is a conversion tool, not a legal footnote. It removes the main risk of buying something unseen.

Say it plainly and near the buy button. “30 day returns, we pay postage” does more work than any amount of persuasive copy, because it transfers risk away from the customer at the exact moment they are weighing it.

Burying it in a footer link works against you, because the anxiety is present whether or not the answer is easy to find.

6. Reviews on the product itself

Site wide testimonials do very little on a product page. Reviews of this specific item do a great deal.

If you sell clothing or anything with fit, reviews that mention sizing are worth more than everything else on the page combined. Prompt for that specifically when you ask.

A handful of genuine reviews including a lukewarm one is more persuasive than a wall of perfect scores, which reads as curated because it usually is.

7. A checkout that does not fight the customer

Baymard’s usability research consistently identifies forced account creation as a major abandonment cause. Somebody who is ready to give you money should not be required to invent a password first.

Offer guest checkout. Keep the form short and ask only for what you need to fulfil the order. Show progress if it runs to multiple steps. Do not clear the form when validation fails, which is a small failure that produces disproportionate rage.

8. It has to load

Everything above is irrelevant if the page does not appear. Product pages are especially vulnerable because they are image heavy, and image heavy pages are where oversized files do the most damage.

Google’s threshold is 2.5 seconds for main content, measured at the 75th percentile. Run your best selling product page through PageSpeed Insights on mobile and see where you stand. If images are named as the cause, they can almost always be a fraction of their size with no visible difference.

The diagnostic

Take your best selling product page and go through this list, marking each item present or absent.

Then buy something from your own store on a phone, all the way through to payment, as if you were a stranger. Notice every point where you hesitate or have to hunt for something. Those are the same points where customers leave, and you will find two or three you did not know about.

Then do the same on a competitor’s site and note what they answer that you do not.

Most stores are missing three or four items from this list. None of them require a rebuild, and the ones about clarity, shipping costs and returns are usually just writing.

The mobile version is the real version

Most of the traffic is on phones and most product pages are designed on a desktop, which produces predictable failures.

The buy button is below the fold and stays there. On a long product page, somebody scrolling through images and description reaches a point of decision and then has to hunt for the button. A sticky add to cart bar that stays visible while scrolling solves this and is standard on well built stores.

The images cannot be examined. On a phone, tapping an image should open it large and allow zooming. If it does not, you have removed the one thing that substitutes for handling the product.

The variant selectors are fiddly. Small dropdowns for size and colour, close together, are hard to use with a thumb. Larger tappable options are better and reduce mis-selection, which reduces returns.

The description is a wall. On a narrow screen, four paragraphs of unbroken text will not be read. Break it into short sections with headings so somebody can find the fact they want.

Buy something from your own store on a phone, all the way to payment. You will find at least two of these.

What to fix first

If you are going through this list with limited time, the order that produces the most improvement per hour:

Add shipping cost and delivery timing to the product page. Unexpected costs at checkout are a leading documented cause of abandonment and this is entirely within your control.

Enable guest checkout if you do not already. Forced account creation is one of the most consistently identified abandonment causes in Baymard’s usability research.

Add a scale reference to your main product photos. One image showing the item in a hand or beside something familiar. This reduces both hesitation and returns.

State the returns policy near the buy button. One sentence.

Check the page speed on mobile. Product pages are image heavy and this is where oversized files do the most damage.

Those five are mostly writing and settings rather than development. None of them require a redesign, and together they address the causes that research consistently identifies rather than the ones that feel most pressing.

What abandonment is not telling you

A closing caution about the 70 percent figure, because it gets used badly.

A high abandonment rate is not automatically a problem to be solved. Carts get used as wishlists. People compare across three stores with items sitting in each. Somebody adds something to check the shipping cost with no intention of buying today.

None of that is a broken checkout. It is normal browsing behaviour, and no amount of optimisation will convert most of it.

What you are actually looking for is the portion of abandonment caused by something on your page: a surprise cost, a forced account, a missing answer, a page that would not load. Those are recoverable and they are the ones worth chasing.

The way to find them is not the abandonment percentage. It is going through your own checkout on a phone and noticing where you hesitate, and reading the questions customers ask you before they buy. Both of those point at specific fixable things. The headline number points at nothing in particular, which is why it is quoted so often and acted on so rarely.

---

Cart abandonment figures from the Baymard Institute’s meta analysis of 50 documented studies, updated September 2025: average 70.22 percent, individual studies ranging from roughly 55 to 84 percent. Core Web Vitals thresholds published by Google at web.dev. This article does not estimate what abandonment costs any specific store.

Your site is not performing. Somebody suggests a rebuild. It is a large number and a long project, and you have no way to judge whether it is necessary.

Here is how to work it out yourself, before anyone quotes you.

Start with the diagnosis, not the treatment

The mistake is deciding on a rebuild before establishing what is wrong. A rebuild is a treatment, and prescribing treatment before diagnosis is how people spend a lot of money and end up with a prettier version of the same problem.

So first, get specific about the symptom. “It’s not working” is not a diagnosis. These are:

People arrive and leave immediately. People cannot find what they need. People find what they need but do not make contact. Nobody arrives at all. It looks unprofessional next to competitors. It cannot be updated without help. It is broken on phones.

Each of those points somewhere different, and only some of them are about the website at all. If nobody arrives, your problem is upstream of the site and a rebuild will not solve it.

The case for fixing

Repairs are the right answer more often than the industry admits, because the industry is paid for rebuilds.

Fix if the structure is sound and the problems are specific. Slow loading, an untappable phone number, a confusing homepage sequence, missing evidence, outdated content. These are afternoon jobs on an existing site.

Fix if it works on phones. Mobile capability is the dividing line. A site that is genuinely responsive can usually be improved. A site built before mobile mattered, that only works on desktop, generally cannot be patched into shape.

Fix if you can edit it. If you or someone in your business can update text and images without paying a developer, that is a functioning asset worth improving.

Fix if you are under two years old. Early businesses change what they sell. Building an expensive site around a proposition that will shift in eight months is a common and avoidable waste.

The case for rebuilding

Rebuild if it does not work on phones. Not “looks a bit cramped.” Genuinely broken. Sideways scrolling, unreadable text, buttons that cannot be tapped. Retrofitting responsiveness onto a site that never had it usually costs more than starting again.

Rebuild if nobody can edit it. A site where every text change requires a developer will not be maintained, and an unmaintained site decays into a liability regardless of how good it looked at launch.

Rebuild if the platform is dead or unsafe. Abandoned software, unpatched, unsupported. This is a security issue and not really a marketing decision.

Rebuild if the business genuinely changed. Different services, different market, different name. When the underlying proposition changed, the structure usually has to.

Rebuild if you have tried fixing and measured no change. Note the word measured. Fixed it, gave it a fair period, compared like with like, and nothing moved.

The question that resolves most cases

Ask this: if every specific problem I can name were fixed on the current site, would I be satisfied?

If yes, do not rebuild. Fix the specific problems. You have just described a repair job and the answer is not a new site.

If no, and you cannot articulate what else is wrong, be careful. That usually means the dissatisfaction is aesthetic or emotional rather than functional. Sometimes that is legitimate, because looking dated has a real cost when you are quoting against competitors. But it is worth being honest that this is the reason, rather than dressing it up as performance, because it changes what you should be buying.

Getting an assessment you can trust

The person telling you what to do usually stands to be paid for the answer. That does not make them dishonest, but it is a real bias.

Ask for the diagnosis separately. “Before we discuss a rebuild, tell me what specifically is wrong with the current site.” A good answer is a list of concrete issues. A vague answer about it being outdated is a warning sign.

Ask what a repair would look like and cost. If they refuse to consider repairs at all, that is information about them rather than about your site.

Get the free evidence yourself first. Run PageSpeed Insights. Do the phone test at evening brightness with one thumb. Have somebody who does not know your business try to find your prices while you watch. That gives you an independent picture before anyone quotes.

Ask what happens to your search presence. A rebuild done carelessly can lose accumulated ranking, because URLs change and redirects get missed. Anybody competent will raise this before you do.

The middle option nobody offers

There is a third path that rarely gets proposed because it is less profitable: a targeted rework.

Keep the platform, keep the structure, replace the homepage and the two or three pages that matter, fix the speed, replace the stock photography with real work, and leave the rest alone.

This is frequently the correct answer for a small business. It addresses the pages that actually get seen, costs a fraction of a rebuild, and takes weeks instead of months. If your site is fundamentally sound but the important pages are weak, ask specifically for this and see what comes back.

An order of operations

If you are unsure, do it in this sequence and stop as soon as things improve.

First, fix what is free. Speed, tappable phone number, correct details, real photos, homepage sequencing. Days of work, little or no money.

Then measure. Give it a fair period and compare against the same period last year, not last month, especially if your business is seasonal.

Then rework the key pages if the fundamentals are fine but the important pages are not doing their job.

Then rebuild, if you have done the above and the constraint is genuinely structural.

Most businesses that follow this order never reach step four, and the ones that do reach it arrive with a clear brief and evidence about what was actually wrong, which produces a much better rebuild than starting from a vague dissatisfaction.

The honest summary

Rebuilds are sometimes necessary and frequently oversold. The tell is whether anyone diagnosed the problem before proposing the treatment.

Do the free diagnosis yourself. Write down the specific failures. Then ask whether fixing that list would satisfy you.

Most of the time it would, and most of the time nobody asked.

What a rebuild actually involves

If you do decide on one, knowing the shape of the project helps you judge quotes and spot omissions.

Discovery and structure. What pages exist, what each is for, what the site is trying to achieve. Skipping this is how you end up with a prettier version of the same problems.

Content. Somebody has to write the words and supply the photographs. This is the step that delays most rebuilds, because businesses assume the designer handles it and designers assume the business does. Agree explicitly who is doing it before you start.

Design and build. The visible part, and usually the shortest.

Migration and redirects. Every old URL needs to point to its new equivalent. Get this wrong and you lose accumulated search presence, which is a genuinely painful and entirely avoidable loss. Ask about it specifically.

Testing. On real phones, on cellular, by somebody who is not the person who built it.

Handover. Can you edit it? Do you own the domain, the hosting and the accounts? Get answers in writing.

That last point matters more than it sounds. A rebuild that leaves you unable to change your own text, or unable to access your own domain, has replaced one dependency with another.

The question to ask yourself in a year

Whichever route you take, the outcome worth aiming for is a site you can and will maintain.

The most common cause of a bad website is not a bad build. It is three years of nobody touching it. Services changed, prices changed, staff left, photos aged, and nothing got updated because updating required a phone call and an invoice.

So when comparing options, weight this heavily: after this project, can somebody in my business change the text and swap a photo without paying anybody?

If the answer is no, you are buying something that will decay on a schedule, and you should factor that into the comparison alongside the price.

The seasonal timing question

One practical consideration that gets ignored and occasionally costs a lot.

Do not rebuild during your busy season.

A rebuild needs your attention: content, decisions, feedback, testing. If it lands during the eight weeks when your phone does not stop, one of two things happens. Either the project stalls because you cannot respond, or it ships without your proper review because you were too busy to look.

There is also risk in changing your site right before the period when it matters most. Migrations occasionally go wrong, redirects occasionally get missed, and the worst time to discover that is the week your customers are actively searching.

Rebuild in your quiet season, launch before the busy one starts, and give yourself a few weeks of buffer to catch problems.

---

This article contains no statistics and no pricing claims. Where numbers appear elsewhere on this site, they are measured with free public tools or cited to a named source.

I look at a lot of small business websites. The same five problems account for most of the damage, they are largely invisible to the owner, and none of them require a rebuild.

Here they are in order of how much they cost.

1. It is slow, and you have never noticed

Why you have not noticed: You look at your site on a laptop, on your own wifi, with every file already cached from the last time. On that machine it is instant and has been for years. Your customer is on a phone, on cellular, from a cold start, and having a completely different experience.

What it costs: Every visitor who leaves before the page appears was, by definition, interested enough to click. They are the most qualified people you have and they are lost before you get to say anything. Whatever you spent to earn that click still cost you.

How to check: Google PageSpeed Insights, mobile score only. Look at the Largest Contentful Paint figure. Google’s published threshold for good is 2.5 seconds, measured at the 75th percentile of loads.

The usual cause: Oversized images. Somebody uploaded photos straight from a camera and the browser scales them down to fit a phone. The picture looks identical. The load time does not.

Time to fix: Resizing images is often an afternoon. Deeper problems are usually days, not months.

2. The phone number is not tappable

Why you have not noticed: You have never tried to call your own business from your own website.

What it costs: This one is stark. A person on a phone, ready to call, taps your number and nothing happens. They now have to select the text, copy it, switch apps, paste it. Some will. Many will go back to the search results where the competitor’s number is one tap.

You lost somebody at the exact moment of highest intent, which is the most expensive possible place to lose them.

How to check: Open your site on a phone and tap the number. If the call screen does not appear, it is broken.

The usual cause: The number is plain text, or part of an image, or the link was written incorrectly.

Time to fix: Minutes.

3. Nobody can tell what you do or where you are

Why you have not noticed: You know what you do. When you look at your homepage, your brain supplies the context automatically. A stranger has none.

What it costs: A visitor who cannot immediately place you leaves. They are not going to work at it.

How to check: Open your homepage on a phone and look only at what is visible before scrolling. Can you answer three questions from that alone: what does this business do, where is it, how do I contact it?

The usual cause: The information exists, further down, below a large hero image and several paragraphs about commitment to quality. It is a sequencing problem, not a content problem.

Time to fix: An afternoon, and usually no new writing.

4. The evidence is stock photography

Why you have not noticed: It looked professional when the site was built, and stock images are designed to look acceptable.

What it costs: Most people recognise stock photography, at least subliminally. A generic image of a smiling person in a hard hat signals one of two things: you had nothing of your own to show, or you did not think it mattered. Both undermine you, and both are usually untrue, because you almost certainly have done good work worth photographing.

You also lose the actual benefit. A photo of your real work is evidence. A stock photo is decoration and carries no information.

How to check: Look at every image on your site and ask whether it shows your work, your premises or your people. Anything that does not is costing you.

The usual cause: No archive of real photos, because nobody was capturing as they worked.

Time to fix: Immediate for the swap. The underlying fix is a habit: one before shot and one after shot on every job, as a rule rather than a decision.

5. Your details disagree with each other

Why you have not noticed: You would have to look at three things side by side, and there is no reason to.

What it costs: Nobody investigates an inconsistency. They register a small flicker of doubt and move on slightly less confident, and slightly less confident loses when two other options are on the same screen.

There is a second cost. Search engines use consistency of business name, address and phone across sources as a signal of legitimacy, so mismatches work against you in local search as well.

How to check: Open your website, your Google Business Profile and your main social bio at the same time. Compare phone, address, hours and business name character by character. “Street” and “St” count as different.

The usual cause: Things got updated in one place and not the others. Old listings you forgot exist.

Time to fix: An hour, once you find them all. Search your own business name to turn up the listings you had forgotten.

The pattern underneath

Four of these five share a cause: you cannot see your own website the way a stranger does.

You are the worst possible tester of your own site because you know what everything means, where everything is, and what the business does. Every ambiguity that would stop a visitor is invisible to you, because your brain fills it in before you notice a gap existed.

Which is why the most useful thing on this entire list is not any individual fix. It is this: hand your phone to somebody who does not know your business, ask them to find out what you do and get in touch, and then watch without helping.

You will learn more in ninety seconds than from any amount of analysis. It is also mildly excruciating, which is why almost nobody does it.

What this adds up to

None of these five is a redesign. Together they are perhaps two days of work, most of it unskilled, and they address the failures that decide whether an interested person becomes an enquiry.

Business owners tend to assume that if their site is underperforming, they need a new one. Usually they need five specific things fixed on the one they have, and the reason they do not know that is that nothing about this problem announces itself.

Go and check the five. Write down which ones apply. That list is the actual size of the problem, and it is almost always smaller than the version in your head.

Three more worth checking

Not in the top five, but common enough to mention and all quick to fix.

Contact forms that go nowhere. Forms break silently. The address they send to gets closed, the plugin updates, messages start landing in spam. Send yourself a test message through your own form today, and then put a reminder in to do it every few months. Businesses have lost months of enquiries to this and only discovered it by accident.

A footer full of links to dead social accounts. You are advertising the abandoned page problem. Remove links to anything you no longer maintain.

No indication of service area. If you travel to customers, say where. A visitor who cannot tell whether you cover their town assumes you do not.

Why these five specifically

The five are not the most serious problems a website can have. They are the ones with the worst ratio of damage to effort.

Every one is invisible to the owner, cheap to fix, and positioned at a point where an interested person is deciding. That combination is what makes them expensive. A genuinely bad website is a known problem that somebody eventually addresses. These sit undetected for years in businesses that believe their website is fine.

The pattern to take away is not the specific list. It is that the failures costing you most are the ones you are structurally unable to see, which is why the outside test matters more than any checklist including this one.

Hand your phone to somebody who does not know your business. Watch without helping. Ninety seconds.

The order to fix them in

If all five apply, do them in this sequence, which is by effort rather than by impact.

Today, in under an hour: make the phone number tappable, correct any details that disagree across your website, Google profile and social bios.

This week, one afternoon: rearrange the homepage so what you do, where you are, and how to contact you are visible before scrolling. Replace the worst stock photos with real ones if you have any.

This month: resize the images and clear out unused scripts and widgets. If somebody else maintains your site, this is a specific request rather than a vague complaint about speed, which makes it far more likely to get done.

Ongoing: start capturing photographs of your work so the stock photography problem never returns.

The first two groups cost nothing but attention and they address the failures at the point of highest customer intent. Most businesses could clear them in a single Saturday and never do, because nothing about these problems announces itself.

---

Google’s Core Web Vitals thresholds, including the 2.5 second Largest Contentful Paint target measured at the 75th percentile, are published at web.dev. This article deliberately does not estimate what any of these failures costs a specific business, because that number depends on your traffic and your customer value, which only you have.

A small business website has one job. Take a person who is already somewhat interested and give them a reason and a way to make contact.

That is narrower than what most sites attempt, and the narrowness is the point. Sites that try to do everything usually fail at the one thing.

Here is what actually matters, roughly in order of impact.

1. It has to load

Nothing below matters if the page does not appear. Google’s published standard is that the main content should render within 2.5 seconds, measured at the 75th percentile of loads, and you can check yourself against it for free with PageSpeed Insights in about a minute.

Check the mobile score, not the desktop one. The desktop score is flattering and irrelevant because your customer is on a phone, often on cellular, often in a dim room in the evening.

The usual culprits are oversized images, scripts that block rendering, and accumulated widgets from services nobody uses anymore. These are days of work, not months, and they are almost always the highest return thing available.

2. The first screen has to answer three questions

Before any scrolling, on a phone, a visitor should be able to answer: what does this business do, where is it, and how do I contact it.

An enormous number of small business sites fail this, not because the information is missing but because it is below a large hero image and three paragraphs about commitment to quality.

The fix is usually rearrangement rather than writing. One plain sentence saying what you do and where. A tappable phone number. A button with the primary action on it. Everything else can wait for the scroll.

3. One obvious primary action

Decide what you most want a visitor to do. Call, book, request a quote, visit. One thing.

Then make that thing visually dominant and repeat it down the page. Sites offering five equally weighted options produce paralysis, and paralysis produces a back button.

The action should be a verb and it should be specific. “Get a quote” beats “Learn more.” “Book a cut” beats “Contact us.” Vague buttons perform worse because the visitor cannot predict what happens next, and unpredictability is friction.

4. Evidence, early

A stranger cannot assess your competence. They can assess evidence.

Photographs of your actual work, not stock. Reviews or testimonials with names attached. Before and after comparisons if your work transforms something. A specific description of a job you did.

Put some of this high on the page. The common pattern is to put testimonials at the bottom, which assumes people scroll all the way down, and most do not.

The strongest single element for most service businesses is a small set of real photos of real work. Stock photography actively costs you, because most people recognise it and it signals that you either had nothing to show or did not bother.

5. Prices, or at least a range

This is the one business owners resist most, so let me make the case.

The fear is that publishing prices lets competitors undercut you and scares off customers before you can explain the value.

The cost of hiding them is that a meaningful share of visitors will not enquire at all rather than ask. Not because they cannot afford you, but because asking creates an obligation, and people avoid obligations when they are still browsing. You never hear from them, so the cost is invisible while the benefit of hiding is easy to imagine.

You do not need a full price list. A range, a starting point, a typical project figure, or a worked example is usually enough. It lets people self select, which means fewer wasted quotes for you and fewer awkward conversations for them.

If you genuinely cannot publish figures because every job differs, publish the structure instead. Explain how pricing works, what drives it up or down, and what a typical range looks like. That answers the underlying question, which is not “what is the price” but “am I in the right place.”

6. A path that works outside business hours

Most people look you up in the evening. If the only route forward is a phone call during business hours, you are asking somebody to hold an intention overnight and then place a call they do not especially want to make. A lot of interest dies in that gap.

Give an evening path. Online booking, a two field form, a text option. Anything that converts intent into a committed action at the moment the intent exists.

Keep forms short. Every additional field costs you completions. Name and one contact method is usually enough to start a conversation. You can ask the rest once they are talking to you.

7. Signals that you are real and current

A stranger is quietly checking whether this is a going concern. What helps: a physical address, a recognisable photo of your premises or vehicle, faces of actual people, recent work, a copyright year that is this year, details that match your Google profile and social accounts.

That last one is worth checking today. Mismatched phone numbers or hours across your own properties are common and quietly corrosive. Nobody investigates an inconsistency, they just feel marginally less sure, and marginally less sure loses to the competitor whose details agree with themselves.

8. Readability

Light grey text on white is the most common readability failure on small business sites and it is trivially fixable. Test at low brightness in a dim room, not on a bright monitor in daylight.

Short paragraphs. Real headings. Enough contrast. Text large enough to read without pinching.

What does not matter as much as people think

A slider or carousel on the homepage. Most visitors see the first slide and nothing else.

A long about page. Somebody who is ready to hire you rarely reads your history. A short version high on the homepage does more.

Perfect design. Clean and clear beats beautiful and confusing. I say this as somebody who designs for a living. A plain site that loads fast and answers the three questions will outperform a striking one that does not.

Being on every platform, linked from the footer. Links to abandoned social accounts are a liability, not an asset.

The diagnostic

Open your site on a phone, at evening brightness, on cellular, and give yourself thirty seconds and one thumb.

Can you tell what the business does? Can you find the price or a range? Can you take the primary action without zooming or scrolling sideways?

Then have somebody who does not know your business do the same thing while you watch without helping. That exercise is uncomfortable and it is worth more than any amount of analysis, because you will see exactly where they hesitate.

Most sites fail on two or three specific points, and those points are usually fixable in an afternoon. This is rarely a rebuild. It is usually a rearrangement of things that already exist.

The pages that actually matter

Small business sites often have eight or ten pages, and attention gets spread evenly across them. It should not be.

For most businesses, three pages carry nearly all the weight.

The homepage, because it receives most arrivals and most judgments.

The main service page, whichever one corresponds to the work you most want. If you offer six services, one of them is your best business, and that page deserves more care than the other five combined.

The contact or booking page, because it is where intent either converts or evaporates.

Everything else, including your about page, your full service list and any blog, is supporting material. Get the three right before touching anything else, and if you only have budget for a partial rework, spend it there.

Two failures specific to service businesses

No service area stated. If you travel to customers, say where you go, in plain words, using the place names people actually use. A visitor who cannot tell whether you cover their town will not enquire to find out. They will assume not.

No indication of what happens next. Somebody about to contact you is wondering what they are committing to. Will this be a sales pitch? Will someone turn up? Is a quote free?

One or two sentences resolves it. “Get in touch and we will arrange a visit to look at the job. Quotes are free and there is no obligation.” That sentence removes a real hesitation and costs nothing.

Testing it properly

Two exercises, in order of usefulness.

Watch somebody use it. Find a person who does not know your business, hand them your phone, and ask them to find out what you do and how much it costs. Say nothing. Do not help. Note every hesitation.

Ninety seconds of this is worth more than any amount of analysis, and it is mildly excruciating, which is why almost nobody does it.

Do the competitor pass. Open the two businesses that appear next to you in search results and run the same test on them. You will find at least one thing they answer that you do not, and it is usually price.

The realistic expectation

A small business website is not going to produce a flood of enquiries on its own. That is not what it does.

Its job is narrower: when somebody arrives who is already interested, do not lose them. Answer their questions, look credible, and make the next step obvious.

Sites that do that job well are usually plain. Fast, clear, honest about price, full of real photographs, with an obvious way to make contact. They are rarely the most beautiful sites in their category and they consistently outperform the ones that are.

If you are choosing between striking and clear, choose clear. You can be both, but only in that order.

---

Core Web Vitals thresholds, including the 2.5 second Largest Contentful Paint target measured at the 75th percentile, are published by Google at web.dev.

Somebody decides their business needs to look more professional, so they commission a logo. It arrives, it is fine, they put it on the website and the van, and roughly nothing changes.

This happens constantly and the disappointment is predictable, because a logo was never the thing that was going to fix it. A logo is one component of an identity system, and on its own it is close to inert.

Here is what the rest of the system is, and why it matters more than the mark.

What a logo actually does

A logo is a recognition device. Its job is to be identifiable at a glance, at small sizes, in one colour if necessary. That is a real job.

What it cannot do is make you look consistent, credible or coherent across everything a customer sees. Recognition requires having already seen you before. For a stranger encountering you for the first time, the logo is nearly meaningless. They have no memory to attach it to.

So the logo works for people who already know you. Everything else in the system works for people who do not, which is most of the people you need to reach.

What the rest of the system is

Colour. Not one colour. A defined set with rules about proportion. The most common identity failure I see is a business with a nice logo used against fifteen different background colours across their materials, because nobody ever said which ones were allowed.

Proportion matters more than selection. Two businesses can use the same three colours and look completely different depending on which one dominates. A palette where the dark neutral carries seventy percent of the surface and the bright colour is used sparingly reads as confident. The same colours in equal measure read as a children’s toy.

Type. Usually two typefaces, sometimes one, with defined sizes and weights for headings and body text. This does more heavy lifting for perceived quality than almost anything else and it is the element small businesses most often leave to whatever the website template supplied.

Spacing and layout rules. How much room things get. Whether elements align. Whether there is a consistent margin. This is invisible when done well and instantly noticeable when not, and it is a large part of why some materials look professional and others look homemade despite using identical assets.

Photography direction. What your photos look like. Bright or moody, staged or candid, tight or wide. Without a decision here, a business ends up with a website mixing stock photography, phone snaps and one professional shoot from four years ago, which reads as three different companies.

Voice. How you write. Covered elsewhere on this site, and it is genuinely part of identity, not a separate discipline.

Application rules. How all of the above assembles on a business card, a van, an invoice, a social post, a sign. The system is only useful if somebody can apply it without asking a designer each time.

The actual point of all this

Consistency is the product. Not beauty.

A customer who encounters your van, then your website, then your Instagram, then your invoice, is unconsciously checking whether these belong to the same organisation. When they clearly do, the business reads as established and deliberate. When they do not, it reads as improvised, and improvised reads as risky.

That judgment is being made below conscious attention. Nobody articulates “these typefaces are inconsistent.” They just come away with a vaguer feeling that this outfit seems a bit ad hoc.

This is why an expensive logo applied to an inconsistent system changes so little. The logo was never the variable.

Do you need this yet?

I would rather talk somebody out of this than sell it to them at the wrong time, so here is an honest assessment.

Probably not yet if: You are under two years old and still working out what you sell. Your website does not load properly on a phone. Your Google Business Profile is incomplete. You have no photographs of your own work. Your problem is that not enough people know you exist.

In all of those cases the money buys more elsewhere. An identity system makes an existing business look coherent. It does not create demand, and it will not rescue a presence with broken fundamentals underneath it.

Probably yes if: You are getting steady work and being taken less seriously than you should be. You are quoting against larger competitors and losing on perception rather than price. You have multiple people producing materials and everything looks different. You are about to do something that will put you in front of a lot of new people at once, like a launch or a large campaign.

The dividing line is roughly whether your constraint is awareness or credibility. Identity work addresses credibility. It does very little for awareness.

What good work costs and what it includes

Prices vary enormously and I am not going to invent a range for your market. What I can tell you is what should be included, so you can evaluate what you are being offered.

A proper identity engagement should produce the logo in multiple formats and configurations, a defined colour palette with usage proportions, typefaces with a working hierarchy, spacing rules, photography direction, application examples across the specific materials you actually use, and a document somebody else can follow.

If you are offered a logo file and nothing else, you are buying a mark, not a system. That may be all you need and it should be priced accordingly. Just be clear about which one you are purchasing, because the disappointment I described at the start comes almost entirely from buying the first and expecting the second.

The version you can do yourself

If this is not the right time to commission it, there is a meaningful amount available for free.

Pick two colours and one neutral, and write them down. Actual hex codes. Then use only those. This alone resolves most visible inconsistency.

Pick one typeface for headings and one for body, and never use others. Free options are plentiful and good.

Decide what your photos look like and stick to it. If your work photographs best in daylight against a plain background, do that every time.

Make one template for each thing you produce regularly. A social post, a quote document, an invoice. Build it once, reuse it.

That is not an identity system. It is a set of constraints, and constraints produce most of the consistency benefit at none of the cost. A business doing those four things looks considerably more deliberate than one improvising each time, and it is a Saturday afternoon of decisions.

The summary

You are not buying a logo. You are buying the reason everything you produce looks like it came from the same place.

If your work is good and your presence is functional and you are still being underestimated, this is the gap and it is worth closing properly.

If your site does not load on a phone and your Google profile is half empty, fix those first. No mark, however good, survives being applied to a business that appears not to be paying attention.

What to look for in a designer

If you have decided this is the right time, a few things worth checking that are more useful than looking at portfolios.

Do they ask about your business before showing you anything? Somebody who leads with pricing and turnaround, without asking what problem you are solving, is selling production rather than design.

Can they show you a system, not just marks? Ask to see the guidelines document from a previous project. If everything they show is logos in isolation, that is what you will get.

Do they ask where it will be applied? A designer who does not ask about your van, your signage, your invoices and your uniforms is going to hand you something that fails on at least one of them.

Do they push back on anything? Somebody who agrees with every idea you have is not doing the job. You are paying partly for judgment, and judgment sometimes disagrees with you.

The rollout nobody plans for

A point that gets missed and then costs money.

Once you have a new identity, everything carrying the old one is now inconsistent, and inconsistency is the exact problem you paid to solve. Vehicle livery, signage, uniforms, business cards, invoice templates, the website, every social profile, the email signature, any printed material still in a cupboard.

Some of that is expensive and some of it takes months. Plan the sequence before you start, decide what changes immediately and what waits for natural replacement, and budget for it.

A business running new branding on the website and old branding on the van for two years has spent money to look inconsistent in a new way. Either move fast across the visible surfaces or stage it deliberately, but decide which.

What identity does not fix

Worth saying plainly, because expectations here cause a lot of disappointment.

An identity system will not make people aware you exist. It will not generate enquiries on its own. It will not compensate for a slow website, an empty Google profile, or work that is not good.

What it does is remove a specific friction: the small, unspoken doubt a prospect feels when your materials look improvised. That friction is real and it costs real work, particularly when you are competing against larger outfits.

But it is one friction among several, and it is rarely the largest one for a business that has not yet fixed its fundamentals. Sequence matters more than quality here. The best identity in the world, applied to a site that does not load, is money spent on the wrong problem.

---

This article contains no statistics and no pricing claims. Where numbers appear elsewhere on this site, they are measured with free public tools or cited to a named source.

If you search for what a slow website costs, you will find a great many confident numbers. A one second delay reduces conversions by some precise percentage. Slow sites lose a specific share of visitors. Multiply by your revenue, and there is your loss, in bold, with a currency symbol.

I want to explain why I do not use those numbers, and then show you a more honest way to get at the same question.

Where those statistics come from

The widely quoted page speed conversion figures almost all trace back to studies run by very large companies on very large sites. Retailers with millions of monthly visitors. Travel booking platforms. Media companies. These are real studies and the researchers were not doing anything dishonest.

The problem is what happens to the findings afterward.

A number derived from a retailer processing an enormous volume of transactions gets lifted out of context and applied to a six person landscaping company. The traffic is different by orders of magnitude. The customer’s decision process is different. The purchase is different. The competitive set is different. On a huge site, a fractional percentage change in conversion is a meaningful figure because it is calculated across a vast number of sessions. On a site receiving forty visitors a week, that same percentage is a rounding error applied to a rounding error.

So when someone tells a small business owner that their slow site is costing them a specific amount per month, that figure was produced by taking research about a company nothing like theirs and performing arithmetic on it. It looks like measurement. It is closer to astrology with a spreadsheet.

I am not being pedantic for its own sake. I care about this because business owners make real decisions based on these numbers, and when the number turns out to be fiction, the reasonable response is to distrust the whole subject, including the parts that are true.

What is actually established

Strip out the invented precision and a solid core remains.

Google publishes thresholds for what constitutes a fast page, based on research into human attention and on data about what real sites achieve. Their standard is that the main content of a page should render within 2.5 seconds, measured at the 75th percentile of loads. That is a real, checkable, publicly documented standard, and you can measure your own site against it for free in about a minute.

Google also uses page experience as a factor in ranking. Not the dominant factor, and not one that outweighs actually being relevant to the search, but a real one.

And directionally, the finding that people abandon slow pages is consistent across essentially every study that has looked, at every scale. The magnitude varies enormously by context. The direction does not.

That is what is known. It is less dramatic than the usual pitch and it is enough.

The e-commerce case, where numbers do exist

If you sell online, there is one genuinely well documented figure worth knowing, and it is worth knowing precisely because of how it is constructed.

The Baymard Institute maintains a running meta analysis of shopping cart abandonment research. Rather than running one study and publishing one number, they collect every credible study they can find and average them. As of their September 2025 update, that average across 50 documented studies sits at 70.22 percent.

The number itself is interesting. What makes it more useful is that Baymard also publishes the spread. The individual studies range from roughly 55 percent to 84 percent. That is an enormous range, and it is the honest part. It tells you that cart abandonment varies hugely depending on what is being sold, to whom, and how, and that any single figure presented without that range is hiding something.

So if you run a store, roughly seven in ten people who put something in a cart do not complete the purchase, and that has been broadly stable for years despite everything the industry has thrown at it. Speed is one of several reasons. It is not the only one, and Baymard’s own research on checkout usability points to a range of causes, many of them about unexpected costs and forced account creation rather than performance.

How to get a number that is actually yours

Here is the approach I would use if I wanted to know what speed is costing a specific business, without importing anyone else’s research.

Establish where you stand. Run your key pages through Google PageSpeed Insights. Record the mobile scores and the LCP figures. This takes minutes and costs nothing. Now you have a baseline that is about your site rather than an average.

Find out how many people are affected. If you have any analytics, look at mobile sessions specifically. If you do not have analytics, this is the moment to add something simple. Even a rough count of monthly mobile visitors changes this from a hypothetical into arithmetic.

Know what a customer is worth to you. Not revenue, and not per transaction if you have repeat business. What is the value of one new customer over the time they typically stay with you. You are the only person who can supply this and it is the number that makes everything else meaningful.

Fix the speed. The common fixes are unglamorous and fast: resize oversized images, defer scripts that are blocking rendering, remove widgets nobody uses. For most small business sites this is days of work, not months.

Measure again, after enough time has passed. This is the step people skip and it is the only one that produces a real answer. Compare the same period, ideally against the same season the year before, because a landscaping company comparing March to June has learned nothing about page speed.

At the end of that you have a number that belongs to your business. It might be large. It might be small. Either answer is worth having, and either is worth more than a borrowed percentage.

Why to fix it even if the number turns out small

Suppose you go through the exercise and the measurable revenue difference is modest. Is the work still worth doing?

Usually yes, for reasons that do not appear in a conversion calculation.

The speed fixes are cheap and permanent. Resizing images is a one time job that keeps paying. Unlike advertising, you do not have to keep buying it.

Everyone sees it. Speed affects every visitor, including the ones who arrived through a referral, the ones checking you out after meeting you, and the ones who already decided to buy and just want to find your phone number. It is infrastructure, not a campaign.

The people you lose are your best prospects. Every visitor to your site got there through interest. They are further along than a stranger. Losing them to a loading screen is losing the most qualified people you have.

And the fix rarely stands alone. In practice, whoever is compressing your images also notices that your phone number is not tappable, and that a widget from a cancelled service is still loading, and that the hero photo is four years old. The speed audit is a general inspection that happens to have a number attached.

The honest summary

A slow site costs you something. I do not know what, and neither does anybody quoting a percentage at you.

What I can tell you without hedging is that the standard is public, at 2.5 seconds, that you can check yourself against it for free right now, and that the fixes are almost always cheaper and faster than the people selling them imply.

Start with the measurement. If you come out under the threshold, congratulations, genuinely, go and spend your attention on something else. If you come out well over it, you have found a real problem with a known fix and no mystery attached.

That is a better position than most businesses are in, which is not knowing either way.

The cost that is easiest to overlook

There is one cost of a slow site that does not appear in any conversion calculation, and for small businesses it may be the largest.

Slow sites do not get maintained.

When loading your own site is a mildly unpleasant experience, you stop visiting it. When you stop visiting it, you stop noticing that the phone number is old, the top service is no longer offered, the photos are from four years ago and there is a typo in the heading.

I have seen this repeatedly. The speed problem and the neglect problem arrive together, and the neglect problem usually does more damage. A fast site tends to stay current because its owner can bear to look at it.

So even setting aside every visitor you lose to loading time, fixing the speed changes your own relationship with the thing, and that has a compounding effect nobody measures.

What to ask whoever built it

If somebody else maintains your site, three questions produce useful answers and do not require you to understand the replies in technical detail.

“What is our mobile PageSpeed score and our LCP?” A competent answer includes actual figures. An answer explaining that scores do not matter is a warning sign, because whether or not the score is a perfect proxy, the underlying measurements are real and they are what your customers experience.

“Are our images sized for how they display?” This is the most common cause and the easiest to confirm. If the answer is vague, it is worth checking yourself in the PageSpeed report, which names the specific files.

“What is loading that we do not use?” Old tracking scripts, widgets from cancelled services, plugins nobody remembers installing. This audit is quick and frequently removes a surprising amount.

None of these are gotcha questions. Somebody doing a good job will have ready answers and will probably be pleased you asked.

---

Core Web Vitals thresholds are published by Google at web.dev. Cart abandonment figures are from the Baymard Institute’s meta analysis of 50 studies, updated September 2025, average 70.22 percent with a documented range of roughly 55 to 84 percent. This article deliberately does not estimate what a slow site costs any specific business.