Key takeaways
- Most comparisons give you two options. There is a third: building your own system on Airtable or Softr, which costs less than custom development and fits better than packaged software.
- What decides between them is how many people need access, what it costs as that number grows, and how much control you need over the interface. Those are the three we run on every project.
- Off-the-shelf is the right answer more often than a development agency will tell you. If your process looks like everyone else’s process in your industry, somebody has already built it.
- The case for building is not that packaged software is bad. It is that some services have no packaged option, because the service itself is unusual.
- Most failed builds we are asked to rescue failed on the builder, not the platform.
We build on all three of these, so this is not an argument for the one we sell.
How is custom software different from packaged software?
Packaged software is built once and sold to everybody, so it carries every feature its whole market asked for and expects you to work the way it works. Custom software is built for one company, from scratch, so it holds your process exactly and nothing else. The first is a subscription. The second is a project you own.
That is the comparison every article on this subject makes, and it is accurate. It is also missing an option that did not exist ten years ago.
Is there anything between off-the-shelf and custom software?
Yes. You can build your own system on a platform like Airtable or Softr, which sits between the two: cheaper and faster than writing software from scratch, and shaped to your process in a way a packaged product will not be. It has its own ceiling, and knowing where that ceiling sits is most of the decision.
| Off-the-shelf | Built on a platform | Custom development | |
|---|---|---|---|
| What it is | A finished product you subscribe to | Your own system built on Airtable, Softr or similar, without a development team | Software written from scratch on AWS, Next.js, React |
| Fits your process | You adapt to it | Close, within the platform’s limits | Exactly |
| Cost shape | Per user, forever | Per user, forever, plus the build | Larger once, then maintenance |
| Control of the interface | What the vendor allows | What the platform allows | Total |
| Time until people are using it | Immediate | Short build | Longest of the three |
| Who controls the roadmap | The vendor decides what it does next | You, inside the platform’s own limits | You |
| Talking to your other systems | Whatever the vendor supports | Whatever the platform connects to | Built for the stack you have |
| Who maintains it | The vendor | You, or whoever built it | You, or whoever built it |
| Breaks down when | Your process is genuinely unusual | User counts grow, or the interface has to do something the platform will not | Nothing, which is also why it costs what it does |
The middle column is where most of the service firms we work with end up, and it is absent from almost every comparison written on this subject, because those comparisons are written by custom development agencies.
A few of them do name a middle option, and it is a different one: modular products like WordPress or Odoo, where you extend a finished platform with plugins. That is worth knowing about, and it is not the same as building on Airtable or Softr, where the structure is yours from the first table rather than inherited from somebody’s product. Airtable holds the data and the rules, Softr puts a branded interface on top of it, and what the two of them produce belongs to you in shape even though the platform is rented.
One argument you will read everywhere is worth treating carefully. Custom software is sold on competitive advantage, on the idea that a competitor cannot buy what you built. That is true, and it matters only when the software is part of how you win. For most service firms the advantage is in the work, and the system exists so the work happens reliably.
When is off-the-shelf software the right choice?
When your process looks like the standard process for your industry. Accounting, payroll, email, scheduling, basic CRM: these are solved, thousands of companies need the same thing, and the packaged product has had years of other people’s edge cases fixed in it. Buying it is faster and cheaper than anything you could build.
The honest version of this is that off-the-shelf wins more often than an agency that sells custom development will tell you. We say no to builds for this reason regularly.
What does “off-the-shelf software” mean?
It means a product designed for a market rather than for one company. That is the whole definition, and it explains both of its properties: it is cheap because the cost is spread across every customer, and it is rigid for the same reason, because it cannot bend to one of them.
What if there is no software built for my industry?
Some services have no packaged option at all, and this is the situation people most often arrive with.
A company that produces nexus letters for military veterans has no software category. A real estate law firm that wants to run transactions its own way is offered legal practice software with a hundred features it will never open, priced for all hundred. In both cases the search for a tool ends without one, and the choice becomes build or keep using spreadsheets.
Professional services hit this more than most industries, because the service is the differentiator. If what you sell is standardised enough to be bought off a shelf, so is the software that runs it.
How do I decide between buying and building?
How many people need access, what it costs as that number grows, and how much control you need over the interface. We run those in that order on every project. Every comparison on this subject lists the same pros and cons, and almost none of them tell you how to weigh them.
How many people need access. This one moves faster than anything else. Platform tools price per user, and a system that is comfortable at fifteen people is a different conversation at eighty. The number of editors, not the number of employees, is the figure to run.
What it costs as that number grows. Not what it costs today. A subscription that looks cheap in year one is a permanent line item that scales with headcount, and a build is a larger number once followed by a smaller one. The comparison only means something across several years.
How much control you need over the interface. This is the one people underestimate, and it is where platform tools stop. If the screen your clients see has to look and behave a specific way, platforms will take you most of the way and then refuse the last part. Custom development has no ceiling here, which is often the only reason to pay for it.
Before you commit to any of them
The answer is usually in your process, not the tools
We map how a service firm actually works before recommending anything, and the mapping often shows that the problem is a workflow rather than a missing system. That part is the same whichever direction the build goes.
See how we map operationsWhat does moving off spreadsheets actually look like?
A real estate law firm we work with was running transactions on Google Sheets and email. Every deal lived in a spreadsheet, every update lived in somebody’s inbox, and the status of a file depended on who you asked.
They now run transaction management on Airtable. Same work, one place, and the status of a deal is a field rather than a conversation. That is the middle column: not a packaged legal product, not a custom build, a system shaped to how they already worked.
The signal that sent them looking is the one worth watching for. If your team exports data out of your main system into spreadsheets so they can do the actual work, the main system has stopped being the main system.
Why do custom builds fail?
Most of the rescues we are asked to do have the same cause: the person who built it had not built enough of them.
Airtable in particular is easy to start with and hard to be good at. Anyone can make a table and an automation in an afternoon. Data modelling, automation design and interface work are the parts that take years, and they are also the parts that decide whether the thing survives its second year. A base built without them works until it has real data in it.
The second cause is narrower and almost as common: somebody competent with the platform who has never touched the integration layer around it, so the system is fine and nothing else in the business can talk to it.
Frequently asked questions
What is the difference between off-the-shelf and proprietary software?
Off-the-shelf software is licensed to many companies and owned by the vendor. Proprietary software here means a system built for and owned by one company, either by its own team or by a firm it hired. The difference that matters day to day is who decides what it does next.
What are examples of off-the-shelf software?
QuickBooks and Xero for accounting, Salesforce and HubSpot for CRM, Monday and Asana for projects, ADP and Workday for HR, Clio and MyCase for law firms. Every sector has its own. If it has a pricing page with tiers on it and a free trial, it is off-the-shelf software.
Is building on a platform the same as custom software?
No. A system built on Airtable or Softr is yours in shape but not in code: you are inside somebody else’s product, with its limits and its per-user pricing. It is faster and cheaper than custom development, and it has a ceiling that custom development does not.
How do I know when we have outgrown off-the-shelf software?
The usual sign is the export. When people pull data out of the system into spreadsheets to do real work, or when a meaningful part of the process lives somewhere the software does not know about, the tool has stopped holding the process.
Is custom software worth it for a small business?
Sometimes, and less often than it is sold. For most small service firms the platform route reaches the same outcome for less, and custom development earns its cost when the interface or the scale genuinely rules the platform out.
If you are weighing this for your own firm, the longer version of how we build these is on the business operating system page, and what we do with Airtable covers the platform route specifically.