What a Business Website Actually Costs to Run Every Month

Ask what a website costs and you will get a build quote. Ask what it costs to own and the room goes quiet. That gap is the whole subject of this page.

This is not a build-cost guide. If you are trying to budget for making a site in the first place, read how much a website costs a small business or our website cost page and come back. Everything below is about what leaves your bank account in month two and every month after it.

Build cost and running cost are two different budgets

They also trade against each other, which is the part nobody mentions during a sales conversation. A cheaper build very often means a more expensive stack underneath it: more paid plugins doing work a better build would have done natively, cheaper hosting that makes the site slow, and more hours of somebody untangling it later.

So the useful question when you are comparing quotes is not "what does this cost." It is "what does this cost me in month thirteen." Almost no proposal answers that, and asking is one of the fastest ways to tell a good vendor from a cheap one.

Who Google is quoting for this today

Pulled live from Google US, logged out, on 25 August 2026, on the two queries people actually type here. An AI Overview fires on both.

On "website hosting cost", the AI Overview cites three sources, in this order: Forbes, Wix, Hostinger. The organic results underneath run Reddit at number one, then GoDaddy, Forbes, Wix, YouTube, PCMag, Hostinger and Namecheap.

Read that list again. Four of those eight results are published by companies that sell hosting. Two more are publisher review sites that run affiliate programs. The single highest organic result is a Reddit thread, which is people asking strangers because they do not trust the vendor answers. That is a reasonable instinct.

On "how much does it cost to maintain a website", the AI Overview cites nine sources: GoDaddy, Network Solutions, Wix, CanSpace, Hostinger, Elementor, Krishang Technolab, WP Creative and Nexal IT Services. Organic runs Reddit first, then Network Solutions, GoDaddy, Wix, Rocket.net, Quora, Gravitate Design and Hostinger.

There is one difference between those two queries worth noticing. The hosting query is owned end to end by hosts and review publishers. The maintenance query is not: four of the nine sources Google cites in its overview are agencies and plugin vendors, not platforms. Forums take the top organic slot on both. And factoryjet.com appears in none of it, on either query, which is the honest state of play on the day this was published.

Every line that recurs

Here is the full set. Not one of these is optional in the sense of not existing. Some are optional in the sense of being over-bought.

The domain is small and it is the one that takes companies offline

Two rules and they are both boring. Turn on auto-renew. Make sure the registration is in the company name, not a former employee's and not the agency that built the site. We have watched a business lose its domain and its email at the same time because a contractor's card expired and the renewal notices went to an inbox nobody read.

The other thing to know is that the heavily discounted first-year offer renews at the standard rate. That standard rate is the price you will actually pay for the life of the business, and premium and specialty extensions renew higher than common ones.

Hosting is the most over-bought and the most under-bought line at once

It is also the widest range on the bill, running from genuinely free tiers on modern static and edge hosts up to enterprise infrastructure. What sets your number is traffic, whether the site takes payments, and how much of the operations work you are handing to the host.

The mistake here is one-directional and completely predictable: a business spends three months building a site and then puts it on the cheapest shared plan it can find, and then wonders why it feels slow. Hosting is the cheapest performance improvement available to most small sites and the last one anybody makes. If you want a number on what your current speed is costing you in lost conversions, the ecommerce speed calculator puts one on it, and Core Web Vitals in 2026 covers which metrics actually move.

SSL should be a zero on your bill

Let's Encrypt is a nonprofit certificate authority that issues free, automated TLS certificates, and checking their site on 25 August 2026 it describes itself exactly that way. Essentially every credible host and platform uses it or an equivalent. If a line item on your invoice says SSL and has a number next to it, ask what you are buying that your host does not already give away. Extended Validation certificates are real and do cost real money. Almost no small business needs one.

Business email is not the website, and it is always in the same budget

Google Workspace or Microsoft 365, priced per user per month. It gets forgotten in every website budget and then shows up on the card statement anyway. Five seats is a real recurring line, not a rounding error, and the seats of people who have left are one of the two or three things an audit reliably finds.

Apps and licenses are the line that quietly grows

This is where budgets actually break, and it breaks differently on each platform.

• WordPress: a premium theme, a page builder, and paid licenses for forms, security, backups, SEO and caching. Each renews annually. Letting one lapse means no updates, which is a security decision rather than a saving. If you are running this stack, WordPress development is the page for what that maintenance actually involves.
• Shopify: the plan is the floor, not the price. Reviews, email, subscriptions, upsells, inventory sync and page builders all bill separately, and merchants routinely pay more in apps than they pay Shopify.
• Custom builds: fewer licenses, but you now pay for the services the platform used to include, such as search, forms, image delivery, error monitoring and analytics.

There is no published figure for what your app stack costs, because your app stack is a set of decisions rather than a product. The only way to know is to export a year of card statements, list every recurring charge, and cancel what nobody has opened. Every business we audit has at least one of these. Most have several.

A backup you have never restored is not a backup

Some hosts include real backups. Many include a single overwritten copy and call it a backup. What you want is off-site, versioned, restorable, and tested at least once so you know the restore works. Uptime monitoring is free at the basic tier and there is no excuse not to have it, because finding out from a customer is the expensive way. We put actual numbers on that in the cost of website downtime.

Payment processing is a deduction, so nobody scrutinizes it

It is a percentage of revenue rather than a subscription, which means it never arrives as an invoice and never gets reviewed. On most stores it is the single largest recurring cost of running the website, larger than hosting, apps and licenses put together. The next section does that arithmetic properly.

Maintenance is the real number and it is usually invisible

Updates, plugin conflicts, broken forms, content changes, security patching, speed regressions, and the annual "why is the contact form not sending." There are only three ways this gets handled: you do it, you pay someone to do it, or nobody does it.

That third option is what most businesses actually choose, and it is the most expensive one. A site nobody updates accumulates outdated dependencies, a checkout that broke on one browser, and content that stopped matching the business two pivots ago. Then it gets replaced. The SEO version of that decay is worse than the security version and slower to notice, which we covered in why website maintenance matters for SEO.

The one part of this we can price exactly

Almost every number in this category is a range somebody made up. Shopify is the exception, because it publishes plan pricing and card rates openly. These figures were checked on shopify.com/pricing on 25 August 2026 and are for the US.

Published online card rates start at 2.9% plus 30 cents on Basic, 2.7% plus 30 cents on Grow, and 2.25% plus 30 cents on Plus. Two add-ons are worth knowing because they are the clearest proof that the plan is a floor: POS Pro is an extra $89 a month per location, and marketplace order sync is free up to 50 orders a month, then charges 1% capped at $99 a month.

Prices on any third-party platform change, so treat 25 August 2026 as the date these were true and check the plan page before you budget from them. If you are pricing the build rather than the running cost, what a Shopify website costs is the separate conversation.

A worked example: a store doing $20,000 a month

Take a small Shopify store on Grow, billed monthly, doing $20,000 a month in card sales through Shopify Payments. Every figure in the first two rows below is either published or arithmetic on a published figure. The rest genuinely cannot be priced without seeing the store.

Look at what dominates. It is not Shopify. Processing is about five times the plan fee, and it is the only line on that table that grows automatically as the business succeeds. If you have never looked at your effective processing rate, that is the highest-value hour on this whole page.

When a plan upgrade actually pays for itself

This is the one calculation the category never does, and it is simple division. The per-order 30 cents is the same on Basic and Grow, so it cancels out and you can ignore it.

• Billed monthly: Grow costs $66 more than Basic ($105 against $39). The published rate drops 0.2 percentage points, from 2.9% to 2.7%. Divide $66 by 0.002 and the upgrade breaks even at about $33,000 a month in card sales.
• Billed yearly: the gap narrows to $50 ($79 against $29), so the crossover falls to about $25,000 a month.

Run the same division on your own step up, using whatever the plan page says on the day you do it. Two caveats: this assumes Shopify Payments at the starting published rate, and your real rate can differ by card type and region. It is still a far better basis for the decision than a feature comparison.

The four costs that only appear in year two

• Renewal pricing. Hosting introductory rates end. Domains renew at the standard rate rather than the promo. Annual licenses renew at list or lose their launch discount. None of it is hidden, and almost nobody forecasts it, because year one was quoted and year two never was.
• Platform version upgrades. A major PHP release, a theme architecture change, a framework major version. Sometimes trivial, sometimes a project. It arrives on a schedule you do not control, which is the part that makes it a budget problem rather than a technical one.
• Subscription creep. Every marketing experiment adds a tool. Very few get canceled. This compounds silently, and the third-year total is always a surprise to the person who individually approved each small monthly charge.
• Content drift.Prices, team, services and case studies all go stale, and a stale site both converts worse and ranks worse. The fix is a few hours a month of somebody's attention, and it is the one line that never appears on any budget anywhere.

What you can cut, and what you cannot

Cut: paid SSL where your host issues free certificates, page builders you no longer use, duplicate SEO plugins doing the same job, apps nobody has opened in ninety days, email seats for people who left, and hosting you sized for traffic that never arrived.

Do not cut: off-site backups, security updates, uptime monitoring, domain auto-renew, or the maintenance hours. Each of those has a failure mode that costs more than a full year of the thing you canceled.

Worth spending more on:better hosting, because speed is conversion, and a real maintenance arrangement, because the alternative is a rebuild on somebody else's timetable.

Two questions to ask before you sign anything

"What will this cost me in month thirteen?" Get the answer in writing. Introductory pricing is the most common surprise in this entire category and it is completely avoidable by asking once.

"Who owns the domain, the hosting account and the code?" If the answer is your agency, you do not own your website, you rent it. Anyone reputable will put all three in your name without being asked, and will not treat the question as an insult.

What to do this week

• Export twelve months of card statements and highlight every recurring charge attached to the website. Most people have never seen this list in one place.
• Cancel anything nobody has opened in ninety days. Start with apps, then email seats.
• Look up your effective payment processing rate and multiply it by last month's revenue. Compare that to your hosting bill.
• Confirm the domain and hosting account are registered to your business, and that auto-renew is on.
• Restore one backup somewhere safe, to prove the backup is real.

If that exercise shows the site costs more to keep alive than it earns, that is not a maintenance problem and a bigger retainer will not fix it. A website redesign is the honest next conversation, and what a redesign costs a US small business covers what changes that number. If you are still at the earlier question of what a site costs to make in the first place, website cost is our page on it.