A response to How Advanced Features Impact Website Project Pricing (March 2024)
In March 2024 I closed out most of my pricing series with a video on advanced features: eCommerce, event calendars, memberships, and integrations with outside platforms like a CRM. These are the parts of a website that do something, not just say something, and they have always been where budgets get nervous.
My advice back then came down to one move. Find the right off-the-shelf framework and configure it. For almost every advanced need, WordPress had a product ready to go, and I said so plainly:
“these kinds of advanced features can be set up without the need for advanced custom coding.”
That was true, and it was good advice. Custom code was the expensive path, so the smart play was to borrow someone else’s code, thousands of hours of it, for the price of a license.
The big engine for the small job
I did notice the downside at the time. Talking about WooCommerce, I admitted:
“WooCommerce is a pretty big engine to install”
My answer for simpler stores was a form with a payment add-on. That was a sensible middle step, but it was still the same logic. Pick the nearest existing product, then live with whatever it brings along. Sometimes that meant a giant commerce engine to sell a handful of things. Sometimes it meant an events plugin with forty settings when the client needed four.
Every one of those choices was a trade off. You got the feature cheaply, and in exchange you got the weight, the settings screens, the update risk, and the parts nobody would ever use.
Here is what the agentic revolution changes. AI agents now write and revise custom code under experienced direction, and they do it fast. When a client needs ticketed events that stop selling at capacity, we can build exactly that, inside their WordPress site, without hauling in a framework built for a thousand other use cases. The expensive part, the development itself, has collapsed.
Where custom code used to be the last resort
In the same video I described custom integration work as the exception:
“if a third party platform is unique or specialized we can always write a custom API.”
And I called that the moment where the “low code” part of low code / no code became necessary. Custom code was the fallback when nothing on the shelf fit.
Now it can be the first option we consider, not the last. Pushing form data into a client’s CRM, syncing members with an outside system, building a focused checkout: these are the kinds of jobs agents handle well, because the scope is clear and the result can be tested. The code is written for this site and nothing else, so it stays small.
That doesn’t mean we throw out every established product. When a client runs a real store with inventory, shipping rules, and tax across many regions, a mature commerce platform is still the right choice, and its ongoing support is worth paying for. The difference is that the choice is no longer forced by cost. We can pick the big framework because the job calls for it, not because building the right-sized thing would break the budget. That’s the same question I asked about the plugin ecosystem as a whole.
What still has to happen first
The heart of that 2024 video was about asking good questions. Will events be ticketed? How many each month? Does the client already have a deep investment in a CRM? I said the clearer you address these issues, the greater the opportunity to build trust.
None of that has changed. If anything it matters more. When building is fast, the risk moves to building the wrong thing quickly. Someone experienced still has to understand the business, define what the feature must do, decide what belongs in the website and what belongs in a connected service, and test the result before a single customer sees it. Agents follow a playbook. People decide what goes in it.
What has changed is what that conversation leads to. In 2024, scoping an advanced feature mostly meant matching needs to an existing product and accepting its compromises. Today it means deciding what the feature should be, and then building that. Advanced features still show up as their own line items, but they’re a fraction of what they used to cost.
We no longer have to choose between the feature we want and the feature we can afford.
¡Viva la Revolución!