Custom software or off-the-shelf software: when each one makes sense

In almost every small or midsize company there comes a moment when someone says: "we need a system". The first impulse is almost always to buy one that already exists, because it looks fast, it has a free trial and someone else is using it. That impulse is usually right. The problem appears when you buy without asking which part of the operation is the one that makes the company different.

This article is for directors, owners and operations managers of companies with 20 to 200 employees who have to decide whether to pay for a license or commission something custom.

A common problem versus a workflow that is an advantage

There are two kinds of processes inside any company. The first kind is common: payroll, invoicing, basic accounting, customer management. Almost every company does something similar, and there are very mature products to handle them. Buying is the sensible choice.

The second kind is what makes your company charge, deliver or serve in a way others do not. It may be the way you price each project, the way you assign crews or the follow-up you give your repeat customers. That is where your advantage lies, and that is where a generic product tends to get in the way.

A smart decision starts by separating the two. If everything you need is common, there is no reason to build anything. If the process that matters most to you is different, custom software is worth a look.

Cost over three and five years, not today's cost

The most frequent mistake when comparing is to look only at the price of the first month. A per-user license looks cheap until you multiply it by the number of people, by the months and by the years the relationship with the provider will last.

A simple way to calculate it:

• Take the monthly per-user cost of the license you are evaluating.
• Multiply it by the number of users who will use it, including those who join later.
• Multiply it by 36 months for a three-year horizon, or by 60 for five.
• Add the costs of implementation, training and the modules you do not need today but that the provider will ask you to buy at some point.

Compare that total with the cost of a fixed-price custom development with a written scope, plus whatever maintenance you decide to contract. License prices change often, so it is worth confirming them with the provider before deciding.

Signs that off-the-shelf software forces you to improvise

Off-the-shelf software can be the right decision and still create problems. These are the signs that the product does not fit your operation:

• People export information to spreadsheets to do something the system does not do.
• The process changes to fit the screen. An extra step gets done because the system asks for it, even though it adds nothing.
• A report the director needs takes three manual steps and a review by someone who knows the trick.
• There are two systems for the same thing because the first one does not cover part of the work and nobody dares to retire it.

When one of these signs appears, you do not always have to change systems. Sometimes adjusting the process is enough. But if several appear, the cost of improvising already exceeds the cost of the license, and it is worth reviewing with numbers in hand.

Ownership: who keeps the code and the data

This is the question most often skipped and the one that weighs most later. With a rented license, you have access as long as you pay. If you stop paying, or if the provider changes its price or disappears, you lose the tool and, with it, the way you had organized your operation.

Before buying, ask three things. Who owns your company's data. In what format you can export all of it. And what happens to your history if you decide to leave in two years. The answers to these questions weigh as much as the features shown in the demo.

With custom development, ownership must be written down from the start: the code and the data belong to the company that hires. That is not always the case, so it is worth checking in the contract and not in the presentation.

The practical rule: buy and adapt first

Our recommendation for most small and midsize companies is simple. Buy and adapt first. Use what already exists for common processes, even if it is not perfect, and save the development budget for a clear differentiator.

Custom software makes sense when three conditions are met at the same time: the process is different from your competitors', that process makes you money or loses you customers when it fails, and no existing product covers it without workarounds. If one of the three is missing, buying is most likely still the better option.

And if after reviewing your case we conclude that there is no clear differentiator, we will tell you. Honest judgment is also part of what you pay for.

What to ask for in a quote so you can compare

If you are going to compare options, ask every provider for the same information so the proposals can be compared:

1. Exact scope: which processes it covers and which are left out.
2. Total cost over three years, including implementation, licenses, support and training.
3. What support includes, during which hours and with what response time.
4. Who owns the code and the data, and in what format they are exported.
5. Time to have something working, and what gets delivered first.
6. What the proposal does not include, in writing.

A proposal that does not answer one of these six questions is not ready to be compared with the others, no matter how attractive the price is.

Next step

If you have a process where you are torn between buying and building, start by splitting it into common and differentiating. With that list in hand, the decision usually becomes clear.

If you would like to review it with us, tell us about your case at https://miscloud.tech/contacto?lang=en. We reply with a tailored proposal, with clear scope and cost.