Website maintenance costs in Australia, from prices you can check.

Price the site you already have, before you read our argument.

This is built for somebody who owns a website, is not planning to replace it, and wants to know whether what they pay to keep it running is reasonable. It asks 9 questions and returns a 12-month figure with every line itemised, each one marked as published, derived, anchored or assumed. Nothing is stored and there is no email field, so run it as often as you like and take the answer to whoever bills you.

On the answers it opens with, a WordPress site with no stated care plan, shared Australian hosting, a domain due for renewal and 1 to 3 changes a year billed by the hour, it returns $1,012 to $6,833 across 12 months. If you are pricing a build rather than upkeep, that is a different question and it has its own calculator. Both sit on the tools index, which explains which one answers which question.

GST moves every comparison here by 10%, and it moves it our way.

Of the 9 Australian care plan providers verified for this page, 1 states that its price includes GST, 5 state plus GST, and 3 say nothing at all. Our own $297 a month includes it. That single inconsistency is enough to decide which arrangement looks cheaper, and almost nobody writing about this subject mentions it.

Take a real case. A $199 a month plan quoted plus GST is $218.90 once the tax is on it. Set the raw $199 beside our $297 and the gap reads as $98. Set the true $218.90 beside it and the gap is $78.10. The naive comparison overstates the difference by $19.90 a month, and it overstates it in the direction that suits us. Repeat that across a category where 5 of 9 providers quote plus GST and the effect is not a rounding error, it is the argument.

So every figure whose provider states plus GST is grossed up before it goes anywhere near a comparison. The floor of the WordPress band is Let's Build a Website at $49 plus GST, carried as $53.90. The ceiling is DIGITALON Advanced at $399 plus GST, carried as $438.90.

The 3 providers who say nothing are treated differently, and this is the part that matters. Their figures are carried exactly as published and marked not stated. Adding a speculative 10% to a named business's price would be the same error pointing the other way, and it would be a fabricated number sitting next to a real company. If your own invoice does not state a GST position, that is a question for whoever issues it, and it is worth asking before you compare anything.

The 1 line this calculator removes instead of adding.

Answer that hosting sits inside your care plan and the hosting line is deleted, not set to 0. That distinction is visible on screen, because a line costed at 0 still reads as a line you might start paying for, and a deleted line reads as what it is: something you are already covered for.

It matters because bundling is rarer than people assume. Of the 9 verified care plans, exactly 1 contains hosting. DIGITALON bundles it and publishes what it reckons the inclusion is worth: $41, $61 and $299 a month depending on tier. Every other plan in the set expects you to be paying a host separately. So the most common real situation is a care plan plus a hosting bill, and the second most common is 1 plan quietly containing both while the owner keeps paying a legacy host as well.

Here is why this is not a small feature. Every other cost model in this category only ever accumulates. Lines go on, totals go up, and the reader ends up with a figure that double counts anything already bundled. An upkeep total that counts hosting twice is wrong by $222 to $950 a year, and it is wrong in the exact direction that makes replacing the website look like the sensible decision. Since we are the ones selling the replacement, a model that could only add would be a model with a thumb on the scale.

The same logic runs through the other cancelling answers. Say the domain is already paid for this year and it comes out of the 12 months ahead. Say your changes are inside the plan and the hourly line goes. Say you do the changes yourself and the line is removed rather than filled with an invented hourly rate for your time.

Why every price here was read from a provider page, not a search result.

Google's result title for Web Champion's care plans page reads "From $350". Open the page itself and the cheapest plan carrying a price is $2,400 a month.

That is a gap of roughly 7 times, attached to a named Australian business. Anyone assembling a price list from search results rather than from the pages themselves would have published that error, and it would have looked entirely credible sitting in a table. It is not a trick by the provider either: search engines build titles from whatever fragments they can find, and a "from" figure elsewhere on a site is exactly the kind of fragment they grab.

So every figure on this page was read on the provider's own page, on 18 August 2026, and the URL is printed beside it. Check any of them. The point of publishing the link is that you do not have to take our word for the number, and if a provider changes a price the correction is 1 click away rather than an argument.

Every price this model rests on, with the URL it came from.

This is the whole evidence base: 17 prices, from 9 care plan providers, 3 Australian hosts, 2 domain registrars and 3 published hourly rates. Each row carries the wording as it appears on the provider's own page, what that provider says about GST, and a link. All read on 18 August 2026.

It is worth saying plainly what this table replaces. The pages currently ranking for this query publish a range and name nobody, which means their figure cannot be checked, cannot be dated and cannot be wrong in any way a reader could demonstrate. A range with no source behind it is not research, it is a confident sentence. Everything below can be opened in a new tab and argued with.

2 exclusions are worth stating rather than burying. DIGITALON's $999 a month tier is real, published and left out of the band as an outlier for a small business website. And every promotional rate in this dataset is ignored in favour of the standing one, because every promotion here runs the same way: it makes keeping the site look cheaper than it stays.

8 Australian providers that publish no price at all.

This list is the finding, not an afterthought. While researching the prices above we kept landing on pages that discuss what website maintenance costs in Australia at length, quote general industry tiers, and never once say what the business writing the page charges. Naming them is more useful to you than a 12th unsourced range would be, and it is checkable in both directions: open a URL and see for yourself.

None of this is an accusation. Declining to publish a price is a legitimate commercial decision, particularly for work that is genuinely scoped per client. But it has a consequence a buyer should understand: you cannot compare a business that publishes against a business that does not, and the ranges circulating in this category are largely built from the ones who do. Every "typical Australian maintenance cost" you read is a sample of the transparent minority.

The 2 registrars on that list deserve a separate mention, because a domain is a commodity product with no scoping involved. There is no per-client complexity to price. A registrar that will not show a renewal figure until you hold an account has made a decision about when you find out, and the moment you find out is the moment switching is inconvenient.

Website hosting costs in Australia, in 2 bands that should never be merged.

Shared Australian WordPress hosting runs $18.50 to $79.20 a month at standing rates. Genuinely managed Australian hosting starts at $69.95 and runs to $299.95. Those are different products sold to different buyers, and averaging them produces a number that describes nobody. The model keeps them apart and asks you which one you are on, with an honest third option for readers who do not know, which routes to the shared band because that is where most small business sites actually sit.

The real trap in hosting is not the headline rate, it is which rate you are reading. Every host in this set publishes at least 2 numbers, and the cheaper one always has a condition attached.

VentraIP advertise $9.25 and charge $18.50 from the second invoice onward. Digital Pacific will drop the Basic plan to $14.66 if you sign for 3 years, which is a commitment decision wearing a price tag. This model uses the standing month-to-month figure in every case, and it does so because the alternative flatters the option we are arguing against, which would make it easy for us and useless to you.

One more thing about this line, and it is the reason it sits second rather than first. Hosting is the item most owners shop hardest on, and across 12 months the entire shared band spans $222 to $950. The changes line below can move further than that on a single busy quarter. If you are trying to reduce what a website costs you to keep, hosting is rarely where the money is.

Domain name costs in Australia: $22.95 against $88.00 for the identical .com.au.

A .com.au is a commodity. The registry is the same, the product is the same, and there is nothing a registrar can do to make one better than another. Yet VentraIP publish $22.95 a year to renew and Melbourne IT publish $88.00 a year, which is 3.8 times as much. This is the single most useful line on the page for a reader who has never checked, because it is the easiest to change and the least likely to have been looked at since the site was set up.

Be careful with that 3.8x, because it is not quite like for like and we would rather say so than let it read cleaner than it is. VentraIP publish a registration price and a renewal price side by side. Melbourne IT publish an annual figure and state that renewal pricing becomes visible once you log in to an account, so $88.00 is the number they do publish rather than a confirmed renewal rate. The comparison stands, and the caveat is part of the finding: the registrar charging 3.8 times as much is also the one that will not show you what happens next year.

Registration and renewal being different prices is normal across the industry and it is where most of the surprise lives. VentraIP charge $9.95 to register and $22.95 to renew, so the figure that attracted you is not the figure that recurs. Whatever you pay, check 2 things on your own account today: that the domain is registered to your business rather than to whoever built the site, and that auto-renew is on. A lapsed domain costs far more than $88.00 to sort out, and occasionally it cannot be sorted out at all.

The one-off charges that never appear in a plan comparison.

Comparing plans by their recurring figure is how everyone does it, and it quietly removes 4 published charges from the picture. Each of these sits on the same provider page as the plan it belongs to, which means the information was never hidden. It just does not survive being summarised into a table of monthly rates.

These are not in the calculator, deliberately. A setup fee you already paid 3 years ago is not part of your next 12 months, and adding a hacked-site recovery to everyone's total would be pricing a risk we cannot honestly estimate. They are here because they change how you read a plan: 2 arrangements at $99 a month are not the same arrangement if 1 of them started with a $99 fee and charges $349 to catch up a neglected site.

What an in-plan change allowance actually contains.

3 of the providers here publish an hourly rate and the spread is narrow: GoldenWeb from $120 an hour plus GST, DIGITALON at $149 plus GST, and Webgrow at $176 stated as including GST, falling to $149.60 above 3 hours a month. The model runs $120 to $176 an hour. That part is published and it is not controversial.

What is ours, and labelled as ours everywhere it appears, is the assumption that 1 change takes about 1 hour, and that a reader answering "9 or more" is modelled at 9 to 12 hours. Nobody publishes that conversion, so it cannot be sourced and we have not pretended otherwise. There is a hint in the data that it is generous to the alternative rather than to us: Webgrow publish that basic changes start from $88, which is well under an hour at their own rate.

The more useful question is what "changes included" means when a plan says it, because it means something different everywhere it is written. The published answers in this set are specific and worth quoting at your own provider. Approved Systems include 1 hour of support a month. GoldenWeb bill $120 plus GST for anything beyond the plan, dropping to $100 and $90 an hour on higher tiers. Let's Build a Website sell support in 30 minute blocks. Those are 3 genuinely different products sharing 1 phrase.

So the arithmetic on this line is straightforward once you count honestly. At $120 to $176 an hour, 4 changes a year is $480 to $704, which is already more than the entire shared hosting band. At 9 or more it passes $1,000 at the low end and $2,112 at the high end, at which point a plan with a real allowance in it starts making sense on cost alone. Count what you actually asked for last year rather than what you meant to ask for, and if the answer is nil because you stopped bothering, put that in the total as a cost rather than a saving.

What this model refuses to put a number on.

Every gap below could have been filled with a plausible figure nobody would have questioned. Each one is left empty instead, and named, because a total assembled from 4 sourced lines and 2 invented ones is not a sourced total, it is a sourced-looking one.

The 1 answer that switches the comparison on.

8 of the 9 questions price your upkeep. The ninth asks whether the website has produced an enquiry in the last 90 days, and it is the only one that changes what this page argues.

Answer yes, or that you are not sure, and we stay out of it. You get the total, the itemised lines, the subtraction where a line was already covered, and a plain instruction: if the site is bringing in work and it costs less than $3,564 to keep, keep it. That is a real recommendation rather than a rhetorical one. Most readers who run this will land there, and the arithmetic above is what put them there.

Answer no, and the page reframes rather than claims a win. The question stops being which option is cheaper, because at that point you are paying to maintain something that is not producing work, and the cheapest version of that is still the wrong spend. Whether the honest answer is us, a different provider, or paying somebody to work out why the site is invisible, is not something a calculator can settle.

If the tool tells you to keep what you have, keep what you have. We would rather publish a calculator that sends most of its readers away with a clear answer than one that manoeuvres all of them toward a form. The 36.7% figure above is not an admission, it is the product working.