Small Business Website Design: How to Hire Well

Hiring a small business website design service is a purchase most owners make rarely, under time pressure, with no easy way to judge quality in advance. The result is a market where price varies by an order of magnitude for work that looks similar in a proposal, and where the most consequential terms are the ones nobody reads.

This verdict stays on the hiring decision: how to specify the work, how to read what comes back, and what to settle before money moves. Our business website cost review carries the money side, and our website builder cost review covers the DIY alternative honestly, including where it is genuinely the better choice. Score your own brief as you read with the scope companion.

Start by deciding what the site is for

Almost every problem later in this process traces back to skipping this.

A website is not a deliverable, it is a tool for a specific job, and the job determines everything from the page list to who you should hire. The useful version of this exercise takes twenty minutes and produces one sentence.

Lead generation. The site exists to make strangers contact you. Then the enquiry path, the trust signals and the forms matter far more than the aesthetics, and you should be interviewing on conversion rather than on visual flair.

Credibility. The site exists because prospects who heard about you elsewhere will look you up. Then it needs to look competent and load fast, and it can be small. Many small businesses need only this, and are sold considerably more.

Selling directly. Then it is an ecommerce project, which is a different discipline with payment, tax, shipping and inventory questions attached, and the shortlist of who can build it narrows sharply.

Booking or scheduling. Then integration with a booking system is the core of the job, and the designer’s experience with that specific system matters more than their portfolio.

Information and support. The site reduces the volume of routine questions you answer by phone. Then structure and search matter more than design.

Write one sentence naming the primary job. It becomes the sentence you judge every proposal against, and it is remarkable how many proposals do not address it.

The four routes, honestly compared

Do it yourself on a builder. Cheapest in money, most expensive in your time. Produces a respectable result for a brochure site if you have any design sense, and a visibly templated one if you do not. Right for a great many small businesses, particularly early on.

Hire a freelancer. One person, direct communication, lower cost, faster to brief. Best value for a well-defined brochure or small business site. The risks are capacity, continuity and breadth: a freelance designer who is excellent at design may not write copy, may not do technical search work, and cannot be in two places at once.

Hire a small studio or agency. Several skills under one contract, process, redundancy if someone leaves. Costs more, and the person you talk to may not be the person building. Right when the project spans disciplines, when there is an integration, or when you want someone else to own the coordination.

Use a done-for-you package provider. A template-based service selling a small business site for a fixed monthly fee, often bundling hosting and maintenance. Genuinely convenient and genuinely cheap upfront. The thing to check with real care is ownership and exit: some of these arrangements mean you are renting a site rather than owning one, and leaving means starting again.

There is no ranking here. There is a fit, and the fit depends on the sentence you wrote in the previous section.

Write the brief before you make contact

A one-page brief transforms the quality of the responses you get, and it costs an hour.

Include the job the site is for, in the sentence you already wrote.

A page list, as best you can guess it. Five pages is a real answer, and knowing whether it is five or thirty changes the price more than almost anything else.

Who your customers are and what they need to see before they contact you.

Three sites you like and one word each on why. Three you dislike is equally useful and more honest.

What you already have: a domain, existing hosting, a logo, photography, written copy, brand colours. Be honest about what does not exist.

Whether anything needs to connect to another system: a booking tool, a payment processor, an email marketing platform, an accounting package. Each connection is a priced piece of work rather than a checkbox, so name them now instead of in week six. Our CRM buyer’s walkthrough and our email marketing pricing review cover the tools most often sitting on the other end of one.

Your budget range and your deadline. Withholding the budget wastes everyone’s time and produces proposals aimed at a number you cannot pay.

Who decides. A project with three stakeholders who have not agreed with each other is the most common cause of runaway revision rounds.

Send the same brief to everyone. This is the single highest-leverage thing in this article, for the same reason it is in any trade hire: comparable inputs produce comparable outputs.

What belongs in the proposal

Now read what comes back, and read it for absence rather than for enthusiasm.

A page list and template count. Templates matter more than pages, because ten pages sharing one template is much less work than four pages with four distinct designs.

Who writes the copy. The most commonly unstated item and one of the largest. If it is you, the schedule is now in your hands. If it is them, it should be priced.

Who supplies and licenses images. Stock imagery has licence terms, and the licence should be yours.

Revision rounds, as a number. Unlimited revisions is not a benefit; it is an unpriced risk that eventually becomes a conflict.

Responsive behaviour. Explicitly, that it works on phones, since that is where most of your visitors will be.

Search fundamentals. Titles, descriptions, headings, clean URLs, a sitemap, reasonable performance. This is not the same as an ongoing search campaign, and nobody honest guarantees rankings.

Analytics and form delivery. Both should be installed and tested. A form that silently fails is the most expensive small bug in this field.

Accessibility basics. Sensible contrast, alt text, keyboard navigation, form labels. This is ordinary competent practice.

Launch and handover. What is transferred, in what form, and a training session or documentation.

Post-launch arrangements. What happens the day after, and what it costs.

Timeline with dependencies named. Specifically, which dates depend on you supplying something.

Ownership: the clause that matters most

This deserves its own section because it is where small businesses get genuinely trapped, and because the trap is invisible until you try to leave.

The domain. Should be registered in your business name, with your business as the registrant, ideally in an account you control. A designer registering it on your behalf in their own account is common and is a dependency you do not want. If it already happened, it is fixable, and the sooner the better.

The hosting. Should be an account in your name with your payment method, or at minimum an arrangement where you can take over without the site going dark. Bundled hosting is convenient; bundled hosting you cannot access is leverage.

The site files and database. Should be yours, with access credentials handed over at launch.

Design source files. Worth asking about explicitly, because a completed site does not always come with the editable design assets.

Third-party licences. Premium themes, plugins and fonts should be licensed to you. A site running on a licence registered to the designer stops receiving updates the day the relationship ends.

The exit clause. One plain sentence in the contract saying what happens to all of the above if either party walks away. Its absence is not neutral; it is a risk sitting entirely on your side.

Ask these six questions before signing. A good designer answers all of them without hesitation, because a good designer has no interest in holding a client hostage; it is bad business and reflects badly on them.

Judging a portfolio properly

Portfolios are curated by definition, so read them actively.

Visit the live sites, not the images. A design mockup tells you nothing about whether the thing works. Load them on your phone, on a slow connection, and see how they feel.

Look for businesses like yours. Not the same industry necessarily, but the same shape: similar size, similar job for the site.

Check whether the sites still exist. A portfolio of dead links tells you something about client retention.

Read the copy on those sites. If it is good across several, the designer probably contributes to it. If it is uniformly weak, copy is your problem to solve.

Notice the range. Five sites that look identical suggests a template applied repeatedly, which is not automatically bad and is worth knowing if you were expecting bespoke.

Ask for a reference you can speak to. Then ask that reference the useful question: what went wrong, and how was it handled.

Content: the thing that actually delays your project

Worth its own section because it derails more website projects than every technical problem combined.

Design cannot be finished around content that does not exist. A designer given placeholder text produces a layout that looks good with placeholder text, and then your real copy arrives at a different length and the layout has to change.

So decide, before the project starts, which of three routes you are taking.

You write it, on a deadline you commit to in writing, ideally before design begins. This is cheapest and it requires actual discipline.

They write it, priced as a line item. More expensive and usually better, because writing for the web is a skill and most business owners write about their business the way they talk about it internally rather than the way a customer needs to read it.

A separate copywriter writes it, briefed by you and coordinated with the designer.

The same applies to photography. Stock imagery is acceptable and often sensible; a real photograph of your actual premises, team or product is better, and arranging one takes lead time that belongs in the schedule rather than in a surprise.

A useful discipline: write the content for your three most important pages before you sign anything. It clarifies the page list, it makes the brief sharper, and it tells you honestly whether you are going to do the rest.

Red flags, ranked

Making two quotes comparable

Nothing else in this verdict pays for itself as quickly as this step, and almost nobody performs it.

Normalising is arithmetic rather than negotiation, and it takes about twenty minutes. Write your own scope down first, in the points from the previous section, then adjust every quote until each one describes that same scope.

Start with your own total. Take the brief used in the worked example below: five pages, three distinct designs, lead generation, no online selling, an existing logo, no written copy, no professional photography, and a contact form that emails the office. A form that sends an email is not an integration, so nothing is added for it. That leaves the 100-point baseline, plus 35 points because the copy does not exist and somebody has to write it. Your real requirement is 135 points whether or not any quote in front of you says so.

Now pull every quote up to 135. A proposal that omits copywriting is a 100-point proposal. It is not cheaper than a 135-point one, it is 35 points smaller, and those 35 points get paid either in money to a writer or in your own evenings for a month. Add your own honest estimate of that back on before you compare anything. Do the same for photography, for an integration one side assumed and another did not, and for a page that one quote counted separately and another folded into an existing design.

Then separate the terms from the scope, because they behave differently. Revision rounds, domain and hosting registration, image licensing, what handover contains and what happens if either side walks away are not points of scope. They are risk allocation. They do not make the job bigger or smaller so much as decide who pays when something goes wrong. A cheaper proposal that leaves all of them undefined has not saved you money, it has moved a cost off the page and onto your side of it.

Two habits make this reliable. Send the same brief to everyone, so the quotes arrive comparable instead of being forced into comparability afterwards. And when a quote is silent on something, ask rather than assume, because silence is not inclusion and the gap surfaces at the worst available moment. It is the same discipline our how to choose HR software walkthrough applies to software shortlists: define the requirement once, then score every option against that single definition rather than against each other.

Only after all of this does price carry information. Two normalised numbers sitting close together are telling you the going rate for your actual scope, which is genuinely useful. Two that stay far apart after normalising are telling you something real about experience, capacity, or how much one party wants the work, and that is worth asking about directly rather than guessing at.

A worked example: two proposals for the same brief

A four-person trades business needs a site that generates enquiries. They write a one-page brief: five pages, lead generation, no ecommerce, existing logo, no written copy, no professional photography, a booking form that emails the office.

Proposal A is markedly cheaper. It lists five pages and a price. Copy is not mentioned. Images are described as sourced. Revisions are described as until you are happy. Hosting is included at a monthly fee, with the domain to be registered by the designer. No handover section. No post-launch arrangement.

Proposal B is higher. It lists five pages across three templates, names copywriting as an included line item with a word count, states that stock images will be licensed to the client, specifies two revision rounds with a rate for further rounds, includes analytics and form testing, lists the client as domain registrant and hosting account holder, includes a handover call and a written credentials document, and offers an optional monthly maintenance retainer with its contents itemised.

Proposal A is not cheaper; it is smaller. In the points used earlier, A has quoted the 100-point baseline while this business actually needs 135, because the copy does not exist yet and the brief says so plainly. Once copywriting is added, the ownership terms corrected, and the undefined revision scope priced, the two land much closer together. The remaining gap is real but modest, and it buys defined risk instead of undefined risk.

The business goes back to A with three questions: who writes the copy, whose name goes on the domain, and how many revision rounds are included. The revised proposal arrives close to B, and now the choice is between two comparable offers, which is the only situation in which price is a legitimate tiebreaker.

Common mistakes small businesses make here

Comparing prices before comparing scopes. The single most expensive error, and the most common.

Not writing a brief. Without one you get five proposals for five different websites and no way to judge them.

Letting the designer own the domain. Fixable, but only before it matters.

Buying a redesign when the problem is content. A site that does not generate enquiries frequently has a copy problem, not a design problem, and redesigning it produces a prettier version of the same silence.

Skipping the post-launch question. Then nothing happens after launch, which is a slow decline rather than a failure and so goes unnoticed.

Approving a design without seeing it with real content. Layouts that work with placeholder text sometimes do not survive contact with actual copy.

Adding scope verbally. Every change feels small. The invoice does not.

Treating unlimited revisions as a win. It is an unpriced risk, and the party bearing it will manage it somehow.

Not testing the forms after launch. Test them yourself, from a phone, on a different network, and check the email arrives. Do it again three months later.

Troubleshooting the relationship

The project has stalled. Establish honestly whose side the blocker is on, because it is more often the client’s content than anyone admits. If it genuinely is the designer, ask in writing for a revised schedule and a reason. A shared list of who owes what, with dates against each item, ends most of these arguments before they start, and our project management software cost review covers what a tool for that runs if a shared document stops being enough.

Revision rounds have run out and you are not happy. Reread the scope. If the request is genuinely within it, say so specifically, citing the document. If it is new, expect to pay, and get the price before the work.

You dislike the design fundamentally. Address it at the design stage, not after build. Be specific: which element, and against which point in the brief. “I don’t like it” cannot be acted on; “this does not read as trustworthy to the customers described in the brief” can.

You want to leave mid-project. Read the termination clause, pay for work genuinely completed, and get whatever exists handed over including domain and hosting access. This goes considerably better where the ownership terms were sorted at the start, which is the entire argument for sorting them at the start.

The site launched and enquiries have not increased. A new site is not a marketing campaign. Check that the forms work, that analytics is recording, and that anyone is actually finding the site, which is a visibility question rather than a design one. Our true cost of business software review covers the same tendency to expect a tool to do a job on its own.

The designer has disappeared. Document everything, check what you can access yourself, and if you hold the domain and hosting your position is recoverable. If you do not, that is the lesson this article most wants you to avoid learning firsthand.

What to agree before any money moves

A short list, all of which fits on one page.

The scope, as a page and template list.

Who writes the copy and who sources images, with licensing named.

The number of revision rounds and the rate beyond them.

The total price and a payment schedule with a final payment held until handover is complete.

Domain, hosting and file ownership, in your name.

A change process: nothing extra proceeds without a written agreement on the work and the price.

The timeline, with your dependencies named as dependencies.

What handover includes: credentials, documentation, training.

What happens after launch, and what it costs.

What happens if either side ends the engagement.

What a small business site actually needs, page by page

Scope creep starts with an inflated page list, so it helps to know what the lean version looks like. Most small business sites need five to seven pages, and a great many need five.

A home page that says what you do, who for, and where, above the fold, and gives one obvious next action. This is the page most often over-designed and under-written.

A services or products page, or one page per service if the services are genuinely distinct and people search for them separately. This is where the pages multiply legitimately, and it is a search decision as much as a design one.

An about page. Small business buyers read this more than owners expect, because they are deciding whether to trust a small supplier. Real names, real faces, real location.

A contact page with a form, a phone number, an email address and, where relevant, an address and hours. The form should be short. Every additional required field costs you enquiries.

Proof of some kind. Testimonials, case studies, accreditations, review counts, photographs of finished work. This is the highest-value page most small business sites omit.

Beyond those, add only what has a job. A blog is a commitment rather than a page, and an empty one dated two years ago is actively worse than none. A gallery is worthwhile if the work is visual and pointless if it is not. A pricing page is a strategic decision, not a design one, and worth thinking about separately.

Two things that are not pages but matter more than most pages: the site must load quickly on a phone on mobile data, and every page needs a clear next action. A beautiful site with no obvious way to make contact is a brochure left in a drawer.

What actually moves the price of the build

Two quotes for a small business website can differ by a factor of ten without either party being dishonest, because the phrase covers products that share a name and very little else. What separates them is scope, and scope is made of six levers you can name and count before anybody quotes you anything.

To keep that comparable without pretending to know prices we cannot know, treat a defined baseline engagement as 100 points of scope: five pages built on three distinct designs, an existing template adapted rather than a design drawn from scratch, copy and images supplied by you, no selling online, nothing connecting to another system, and an ordinary launch. That baseline is roughly what a great many small businesses actually buy. Everything below is an illustrative point cost measured against it, and points are not money, because what one point costs depends entirely on the route you picked. Our business website cost review carries the money.

Distinct designs, not pages. Each design beyond the baseline three adds around 15 points, because a new layout is a fresh round of design, build and testing rather than more of the same. Each additional page reusing a design you already have adds around 4. Ten pages on three designs is a small project. Four pages on four designs is not, and the page count hides that completely.

Custom design against an adapted template. Adapting a well-chosen template is the baseline. A design drawn from scratch, with the concept work, the review rounds and the bespoke build that follow it, adds around 50 points. That is the largest single lever available on a brochure site, and it earns its keep only where the design is doing real commercial work for you rather than satisfying a preference nobody outside your business will notice.

Who produces the content. You supplying finished copy is the baseline. Commissioned copy across those five pages adds around 35 points, and a photography day adds around 25. Both are the first things cut to make a quote look competitive, and both return later as either a delay or a second invoice, which is why they belong in the comparison from the start.

Selling online. Ecommerce is not a feature bolted to a brochure site, it is a different product with a catalogue, payment handling, tax rules, shipping logic and an order flow behind it. Call it 120 points, which more than doubles the baseline on its own. Read our ecommerce platform cost review and our merchant account explainer before committing to it, because the build is only the opening cost.

Integrations. Each system the site has to talk to adds around 40 points, and the first one adds the most, since it introduces a testing burden that did not previously exist. A booking tool, a payment processor, an email marketing platform, an accounting package: each is a separate piece of work with its own failure modes, and each needs somebody named as responsible when it breaks.

Migrating what already exists. Moving an existing site’s content across, and mapping its old addresses to the new ones so that nothing already working quietly stops, adds around 20 points on a small site and considerably more on a large one. It is invisible in a proposal that does not mention it, and expensive in the month after launch if nobody did it.

Maintenance sits outside this index entirely, because it is recurring rather than one-time. Mixing the two is exactly how a monthly package ends up looking cheaper than a build it was never comparable to.

One caution about all of these numbers. They are a scoring device for comparing proposals against each other, not a quote, and no designer will recognise them as their own pricing. Their value is that they make the shape of a job explicit before anyone attaches money to it, which is the only condition under which two prices mean the same thing.

Pricing models, and which suits you

How a designer charges tells you something about the project shape, and the models are not interchangeable.

Fixed price for a defined scope. The right default for a small business site. You know the cost, they know the work, and the scope document is what keeps both honest. Its weakness is that anything unforeseen becomes a change order, which is fine provided the change process is written down.

Hourly. Suits open-ended or ongoing work, and suits a client who genuinely does not know what they want yet. Its weakness for a first website is obvious: you cannot budget it, and the incentive structure is not aligned with efficiency. Where you do use it, ask for an estimate with a cap.

Monthly package. A fixed monthly fee bundling build, hosting and maintenance. Genuinely attractive cash flow for a small business, and the model to scrutinise hardest on ownership, because in some versions you are renting a site indefinitely and leaving means starting over. Ask directly: if I cancel after two years, what do I keep?

Value or outcome-based. Rare at this end of the market and generally a signal you are talking to someone selling marketing rather than a website.

The maintenance question, settled before launch

A website is not a painting. Left entirely alone it degrades, and the degradation is invisible for months.

Software gets outdated, which is a security question rather than an aesthetic one. Hosting bills come due. Contact forms break, often silently, when an email setting changes somewhere upstream. Certificates expire. Content goes out of date, and a site listing opening hours from two years ago costs you trust in a way no design flaw does.

So decide who handles it. Three legitimate answers.

A retainer with the designer. Convenient and often good value. Get in writing what it covers: software updates, backups with a stated retention, uptime monitoring, a number of content-change hours per month, and a response time for something being broken. A retainer without those specifics is a subscription rather than a service.

A separate maintenance provider. Works well where the build was a one-off and you want continuity independent of the original designer.

Yourself. Entirely viable on a builder platform and on a well-set-up content management system, provided the handover included training. Put a recurring calendar reminder in place, because the failure mode here is not inability but forgetting.

One test worth running quarterly regardless of who maintains the site, and it takes two minutes: submit your own contact form from a phone on mobile data and confirm the email arrives. The number of small businesses losing enquiries to a form that stopped working is genuinely large, and nobody complains to you about a form that failed. They just go elsewhere.

When a redesign is the wrong purchase

A meaningful share of people looking for a website design service already have a website, and are unhappy with it. Before spending, it is worth establishing what is actually wrong, because the diagnosis often points somewhere cheaper.

If nobody visits the site, that is a visibility problem, not a design problem. A redesign changes nothing about it. Check analytics first: if traffic is near zero, the money belongs somewhere other than design.

If people visit and do not contact you, look at the copy and the enquiry path before the aesthetics. Is it obvious what you do within five seconds? Is there an obvious next action on every page? Does the form work, and is it short? Those three are frequently the whole problem, and all three are cheap to fix.

If the site is slow or broken on phones, that is a technical fix and possibly a hosting change, both of which are much cheaper than a rebuild.

If the content is out of date, that is a content job.

If the site genuinely looks like it was built fifteen years ago, or you cannot edit it, or it is built on something unsupported, then a rebuild is warranted and the case is easy to make.

The reason to work through this honestly is that a redesign is the most expensive available answer, and it is the one most often recommended by people who sell redesigns. Arriving at a conversation having already checked your analytics puts you in a much stronger position, and occasionally saves you the entire cost.

Questions to ask on the first call

Twelve questions, none of which require you to know anything technical, and all of which reveal something.

Who will actually do the work, and will I be speaking to them?

Can I see three live sites you have built for businesses of roughly my size?

Who writes the copy, and if it is me, what deadline do you need it by?

How many revision rounds are included, and what happens after that?

Will the domain and hosting be in my name?

What exactly is handed over at the end?

What happens to the site if we stop working together?

What is not included that clients commonly expect to be?

What is your typical timeline, and what usually causes it to slip?

What happens after launch, and what does that cost?

Have you built anything that connects to the systems I use?

What would you want from me to make this project go well?

The last one is the most revealing. Someone who answers it specifically, with content deadlines and decision-maker clarity and access to your existing accounts, has run real projects. Someone who says nothing much is needed from you has either not run many or is telling you what they think you want to hear.

The bottom line

Hiring a small business website design service goes well when you do three unglamorous things: decide what the site is actually for, write the brief once and send the same one to everyone, and settle ownership before you settle price.

Almost every horror story in this field comes from skipping one of those. Sites that never got finished because the content never arrived. Businesses that could not move their own domain. Redesigns bought to fix a copywriting problem. Proposals compared on price that were never comparable.

Score your own brief with the scope companion before the first call, and take the gaps it names into the conversation as questions rather than discovering them in month three.

None of it requires technical knowledge. It requires treating the purchase as a purchase, which is exactly the discipline you would apply to any other supplier, and which people abandon surprisingly often when the deliverable is a website.

VetLoft publishes this buying verdict to help small business owners hire a website design service with confidence, not as legal, contractual, or professional advice. Every weighting and comparison above is illustrative, and contract terms, domain registration practice, licensing, and consumer protections vary by provider and jurisdiction, so read any agreement in full and take qualified advice where the value or complexity of the engagement warrants it. Verify ownership of your domain, hosting, and site files yourself rather than relying on assurances.