Matt Turley, Continuum. Updated September 29, 2026.
Your business runs on a web app someone built years ago. PHP, MySQL, HTML tables. It works, but it looks dated, it is slow in places, and you worry about the day it breaks. You want to freshen it up, and someone has told you it needs a full rebuild.
If the app works and your business runs on it, refresh it: update the look, fix the slow and broken parts, and move it to a supported PHP version in stages while it keeps running. Rewrite only when the code can no longer be changed safely, and switch to off-the-shelf software when your process is standard and an existing tool already does the job.
| Refresh | Rewrite | Off-the-shelf SaaS | |
|---|---|---|---|
| What it means | Keep the app. New look, fixed weak spots, supported PHP, in stages. | Build a new app from scratch, then move your data and users over. | Stop running your own app. Move to a subscription tool. |
| Cost | At Continuum, a Ship Fix is $4,000 to $15,000, or ongoing work on a retainer. | At Continuum, from $15,000 for a first version. Clutch: most projects $10,000 to $49,999. | A monthly fee per plan or per user. Jobber, for example: $49 to $499/mo. |
| Timeline | First visible changes in weeks. The app stays live while it changes. | Six to twelve weeks for a first version, longer to match everything the old app does. | Days to weeks, mostly spent moving data and retraining staff. |
| Who it fits | The app works and holds rules your business depends on. | The code cannot be changed without breaking something else. | Your process is standard for your industry. |
| Main risk | Some old code limits what you can change quickly. | It runs long, and the old app still needs care the whole time. | You bend your process to the tool, and you do not own it. |
Sources: Clutch software development pricing (September 2026), Jobber pricing (month-to-month plans, September 2026). Continuum prices are from the pricing page.
An old app that still runs your business holds years of decisions: pricing rules, edge cases, the fix for the customer who always books twice. None of that is written down anywhere except the code. A refresh keeps all of it and changes what people see and what keeps breaking.
A refresh also ships in pieces. The booking screen gets a new look first. Then the slow report gets fixed. Then the app moves to a supported PHP version. Each step goes live on its own, so you see progress in weeks and never have a big switch-over day.
I have done this. TourPass came to me as somebody else's codebase. I redesigned and refactored it, and I have kept it shipping since early 2025.
Go in with open eyes. The new app must match everything the old one does before anyone can switch, and that list is always longer than it looks. Joel Spolsky called rewriting from scratch "the single worst strategic mistake" a software company can make, after Netscape went almost three years between major releases.
If your app mostly does what a standard tool does, like scheduling, invoicing, or bookings that work the same way everyone else's do, a subscription may cost less than keeping your own app alive. Check that you can export your data first, and price it over three years, not one. See off-the-shelf vs custom software for how to make that call.
According to php.net, every PHP version before 8.2 is past end of life and gets no security fixes. PHP 8.2 loses security support on December 31, 2026. If your app runs on PHP 7 or 5, the upgrade comes before any redesign. A new look on an unpatched server still leaves the door open.
Tell me what the app does and what bothers you about it. I will tell you which of the three it needs, including when the answer is a subscription tool. If you want it in writing first, a $299 Ship Check is a read of your codebase that says whether it is fine, fixable, or a rewrite.