You ever have one of those months where the same topic comes up again and again and again? For example, you're talking to your buddy Doug about when a Wix or Shopify project is ready for a bespoke development. Then, you're at a "meetup" where the topic comes up, again, except this time, in the reverse. And, again, last Tuesday, on a professional call with colleagues where the discussion is brought up. Again...
It all started with a simple question, "when does a client need to go fully custom versus pursuing Shopify/Wix/Squarespace for an ecommerce solution". Kind of a tough question to not get biased over... I mean, creating custom solutions is how I feed my family. How could I not be biased? Still, I'm going to try to answer all of this, without too much bias, mostly just to flesh out my thoughts and ideas. But, if it helps others too, the more the merrier. That said, to be clear, I'm 100% taking the tack of "what are the negatives with a hosted solution", so, a little bias I guess.
Dammit. Happened fast...
Definitions
First, let's define some terms just so future me doesn't have any problems understanding what the hell I wrote.
Bespoke
By this, I mean a custom solution written by a web developer or team and hosted through a web host. Depending on the intended functionality, a bespoke website can cost from the mid hundreds into the millions with no upper limit. Mostly though, in my experience, they hover in the mid/low four to five figures, for teams of developers working together building a purpose built solution (again, depending on the client needs). With bespoke, there is no limit whatsoever to what can be achieved nor how expensive it can become so long as the technology of the day allows it.
Third Party Solution
This refers to a myriad of companies whom specialize in providing cookie cutter, usually online built, website solutions. These solutions range from simple brochure style web sites to full ecommerce solutions with quite a few bells and whistles. These include services provided by Wix, Squarespace, GoDaddy, and Shopify, where a customer can build their site using an online "builder" and ala carte their way into features and solutions.
Who This Is For
This is really only going to be of use to developers and client whom could potentially collaborate. If anyone is thinking about kicking the tires on a random idea, or building their companies first brochure site, all by their lonesome, there'll be very little of use. Go with a hosted CMS and site builder. 100%. Agencies will happily take your money and help you pursue your dreams, but statistically speaking, if you're in that position you have a lot of mistakes to make first. Best not spend money to do it.
Conversely, if a project has anything to do with data, subscriptions, ecommerce, membership, or, well, any complexity of logic, then I'm taking to you.
Dynamic Costs
One argument I leaned heavily into was the base cost for a base solution. Woops. I was wrong. All 3 of the main players appear to have a base cost slightly more than third party hosting. Though, by "slightly more" I am referring to orders of magnitude, but when the base hosting cost for a simple website is $5 a month, $15 a month, for a business venture, isn't too bad, all things considered (especially with the thousands of dollars involved with a bespoke solution).
Even payment processing fees are more in-line with Stripe than anything absurd or dramatic. All hovering around 2.9% with a +$0.30 tacked on. So, that too is a bust.
Where a hosted solution really starts to crack though are with apps and add-ons/add-ins; nearly all apps, across all platforms, incur additional monthly costs. For example, depending on the app, Shopify can increase into the thousands per month. For example, the Shopify Dynamic Pricing AI by Intelis costs an additional $49 a month (for basic) up to $399 a month for their Business tier. Hell, the PageFly Landing Page Builder can cost up to $99 a month! And there are hundreds of apps to choose from just with Shopify Apps, so costs can easily get head tilting and weird.
Customization and Flexibility
A real negative for hosted solutions though is a lack of flexibility. And, of course self hosted is limited on functionality as well. After all, a hosted solution is a lowest common denominator system built to appeal to the widest ranges of customers. Solutions have to fit the broadest number of potential customers. So you're locked into their templates, their functionality, their UI elements, their backend logic, and their philosophy. Whether you like it or not, you only have access to the templates, themes, layouts, colors, fonts, and on and on, that the hosted solution provides. Want something else? It's either "good luck" or "money, please!" for the most braindead of a problem.
Now, options on functionality and template and UI obviously is something very much unique to each platform. Shopify has their ecosystem, Wix theirs, and Squarespace has theirs. Each require special consideration and, most concerning, lock-in with their vendors. Where, again, monthly increases in costs come into play.
Workflow Questions
Controversial take: being able to "get things done effectively" matters. How is data exported? How is Search Engine Optimization (SEO) handled? How are changes staged and reviewed before going live? Whathaveyou...
Again, with bespoke, sky's the limit. Wanna update Amazon Products inventory when a product is sold, while dispatching emails and SMS messages, and have the system pop up a unicorn farting rainbows? Coolio. Done and done. But, again, you'll pay for it. But only the once.
Data is a universal item. Usually, some variety of Excel, CSV, JSON, or XML, but also a complete database is often an options (which is my preference). SEO is easily implementable at best, and, at worst, brute forcible, with plenty of nuance for elegance in between. And version control, with our fancy CI/CD layers at play, so disaster can be averted like it's 2025 and a concrete understanding of a platform is possible.
There isn't a single hosted CMS that will fit every business model or workflow out of the box. No matter what, someone, somewhere, is gonna have to adapt to match their functionality. Battles will definitely be picked. Stress will definitely increase. Money will be spent. Maybe not on a platform, or a developer, or agency, but somewhere. To someone. Time to adapt. Time to figure it out. Time to improve.
Which, to be fair, isn't always a a net negative; it's very much a personal question of what matters most.
Wrapping Up
At the end of the day, none of this is about “right” or “wrong.” It’s about alignment. Hosted platforms absolutely have their place, especially for early-stage ideas, simple catalogs, or teams that don’t have the appetite (or budget) for custom engineering. But the moment a business needs to push beyond the guardrails, the cracks become real: workflow bottlenecks, data friction, recurring app fees, limited customization, and that subtle-but-constant feeling of fighting the platform rather than building the vision.
That’s the inflection point. And it will happen with every growing business eventually.
So maybe the real question isn’t “Should we go bespoke or hosted?” but “When will the limitations start costing more than the convenience?” Because that’s the moment when custom development stops being a luxury and starts being the most cost-efficient path forward.