Most businesses need a website that explains what they do and collects enquiries. That is a well solved problem and it should not be expensive.
Some businesses reach a point where the website stops being a brochure and starts needing to do work. That transition is usually visible in how the team spends its days rather than in anything on the site.
What are the actual signs?
Five, and one is usually enough.
Somebody is retyping information between systems every day. An enquiry arrives on the website, gets copied into a spreadsheet, gets copied into accounting software. Every retype is time and a chance of error, and it grows linearly with the business.
Customers keep asking for information you have but cannot show them. Order status, documents, past invoices, service history. You have all of it. There is just no way for them to see it without emailing somebody.
A spreadsheet has become load bearing. One file, several people editing it, and everybody quietly afraid of it. This is the most reliable signal on the list.
You are paying per user for software you use ten percent of. Common with booking, inventory and project tools. The fee scales with headcount and the value does not.
The same question arrives twenty times a week. Where is my order. Do you cover this area. When are you next available. Each one is a person interrupted.
Does that mean I need custom software?
Not necessarily, and this is the question worth slowing down on.
Off the shelf software exists for almost everything and is cheaper, faster and better supported than anything built for you. The correct first move is always to check whether the problem is genuinely unusual, and it usually is not.
Custom work earns its place in two situations. When the way you operate is a real competitive advantage and standard software would force you to work like everyone else. Or when you need several existing systems to behave as one, which is integration rather than invention.
If neither applies, buy something.
What does “the website does work” actually look like?
Usually one of four things, and they are smaller than people imagine.
A customer login. Somewhere a client sees their own information without emailing anyone. Status, documents, history.
Booking and payment. Availability that reflects reality, taken online, without a phone call to confirm.
A connection between systems. The website enquiry appears in your CRM automatically, with its source attached, and nobody retypes anything.
An internal tool. Something your team uses that replaces the load bearing spreadsheet, with proper permissions and a history of who changed what.
None of these require rebuilding the website. They are additions, and they are usually built one at a time.
How do I know it is worth the money?
Count the hours honestly, then be conservative about the saving.
Take the retyping. If two people spend forty minutes a day on it, that is roughly twenty hours a month, permanently, growing as you grow. Put a cost on those hours. Compare it to a one time build plus modest running costs.
That comparison is usually decisive, and it is rarely done, because the twenty hours are spread thinly enough that nobody experiences them as a cost.
Be conservative though. Automation removes most of a task, not all of it, and there is always a period where things are slower while people adjust.
Should I do it all at once?
No, and this is where most custom projects go wrong.
The instinct is to specify everything, because it feels efficient to do one big build. What happens is a long project, a large invoice, and a system that reflects what you thought you needed nine months ago.
Build the single most painful thing first. Use it for a quarter. Almost always you learn that two of the things on the original list were unnecessary and one thing nobody mentioned is now obvious.
Smaller first builds also mean the relationship with whoever builds it gets tested cheaply.
What should I be careful about?
Ownership and dependency.
Custom software means depending on whoever built it. Before starting, settle who owns the code, where it is hosted, what happens if you want to move, and what documentation exists. Get this in writing at the beginning, when everybody is friendly.
The same question applies to whether another developer could pick it up. Software built in an unusual way by one person is cheap to build and expensive to inherit.
What would I do first?
Write down where the hours actually go for two weeks.
Not what you think, what happens. Most businesses discover the expensive problem is not the one they were planning to solve, and it is often much simpler to fix.
We build custom development only where it is genuinely warranted, and we will tell you when off the shelf software would serve you better. Tell us what you are working on.
