.webp)
%201.png)
Every CMS comparison you will find is a feature matrix. Platform A has multilingual, platform B has a better editor, platform C has an app store. You read it, you agree with all of it, and you are no closer to a decision.
Feature matrices do not decide anything because features are not what breaks. What breaks is the gap between who was promised control and who actually has it.
Here are the four questions that decide the platform. Answer them honestly and the shortlist writes itself.
1. Who Edits, and How Often?
Not who owns the site. Who opens it on a Tuesday to change a headline.
If the answer is a marketer who will make small changes weekly, you need a visual editor and you need publishing to be a non-event. If the answer is a developer who batches changes monthly, you have far more freedom and should optimize for something else.
The expensive mistake is buying for the org chart instead of the calendar. Teams pick a developer-first platform because they have a developer, then discover that every landing page for every campaign goes through a sprint queue. The platform did not fail. The workflow did.
Test It Before You Commit
Ask for a trial where your actual marketer changes a headline, swaps an image, and publishes. Time it. If it takes more than five minutes or requires anyone else, you have your answer.
2. Who Ships, and How Fast?
Campaign velocity is a platform decision that nobody frames as one.
Count how many net-new pages your team shipped in the last quarter. Then ask how many they wanted to ship. The gap is usually not a headcount problem, it is a friction problem, and the friction lives in the publishing path.
A platform where a new landing page is a designer plus two hours produces a different marketing organization than one where it is a ticket plus two weeks. Over a year that difference compounds into dozens of campaigns that either happened or did not.
3. What Breaks at Scale, and Does Anyone Own It?
Every platform is pleasant at ten pages. The question is what happens at four hundred.
Three things reliably break, and they break in the same order:
Structure
Content types that were flexible become inconsistent. Six people have created six ways to make a case study. Nothing enforces the shape.
Performance
Pages accumulate embeds, scripts, and third-party tags nobody removes, and the site gets slower every quarter without any single decision causing it.
Governance
Nobody knows which pages are current. Old campaign pages stay published. Duplicate service pages compete with each other in search results, and traffic splits between them.
Notice that none of these are fixed by choosing a better platform. They are fixed by choosing a platform whose structure is hard to abuse, and by naming an owner. If your evaluation does not include who governs content after launch, you are choosing where the mess will accumulate, not whether.
4. What Does It Cost Three Years In?
License cost is the number in the proposal and the least important number in the decision.
The real three-year cost is license plus the cost of every change. A platform that is 30% cheaper and requires a developer for routine edits is not cheaper. Estimate it crudely: how many content changes per month, times how long each takes, times who does it. That number usually dwarfs the license.
The second hidden cost is exit. Ask, before signing: if we leave in three years, what comes with us? Content in a structured, exportable format is portable. Content baked into proprietary page builders is not. You are not planning to leave, and that is exactly why this is the moment to ask.
The Dimension That Did Not Exist Three Years Ago
There is now a fifth question, and most CMS evaluations have not caught up to it: can the platform be read by the systems that answer questions about you?
A growing share of buying research never reaches your site. Someone asks ChatGPT, Claude, Perplexity, or Google's AI Overview which vendors do a thing, and gets an answer synthesized from pages those systems could parse. Whether your pages are in that answer depends on things a CMS makes easy or hard: clean semantic markup, structured data, content that answers a question in the page rather than behind an interaction, and server-rendered text rather than content assembled by script after load.
This is not a future consideration. If your category has a "best X for Y" question, that answer is already being generated, and it is being generated with or without you.
Add One Question to Your Evaluation
View source on a page built with the platform. If the substance of the page is in the HTML, you are fine. If the HTML is a shell and the content arrives later, you are betting on a crawler being patient.
How to Run This in Two Weeks
Days 1 to 3
Answer questions 1 and 2 with numbers, not opinions. Count edits and pages shipped last quarter.
Days 4 to 7
Shortlist two platforms. No more. Three is a way to avoid deciding.
Days 8 to 11
Build the same real page on both. Not a demo page. A page you actually need, with your content and your approval process.
Days 12 to 14
Have the marketer who will own it make three changes on each, unaided. Whoever they pick, pick.
The framework is deliberately boring. Platform decisions go wrong through enthusiasm, not through lack of analysis.
We help marketing teams choose and implement the platform that matches how they actually work, then build on it. If you are mid-evaluation, see how we build custom sites or how we approach low-code delivery.
.png)
.png)
Join the 1% of Data-Driven Founders
Stop letting technical debt limit your revenue. Get a 30-minute strategic diagnostic of your digital infrastructure. No fluff, just strategy.
.gif)
.gif)
