No code vs custom code for internal tools: which should you use?
Use no-code tools like Airtable, Softr or Notion when the tool is simple, internal, and someone on your team enjoys owning it. Use custom code when the tool is central to client work, has complex rules, needs to connect several apps reliably, or has outgrown the no-code builder. We use both, and most firms should too.
| No-code (Airtable, Softr, Notion and similar) | Custom code from an agency or freelancer | Custom code on a Forge monthly plan | |
|---|---|---|---|
| Typical cost, US | Per seat; for example Airtable Team at $20 a user a month, billed annually | $8,000-18,000 for a portal, dashboard or onboarding system; $15,000-30,000 for a multi-part internal tool | $2,000 / $3,500 / $7,500 a month, first build included |
| Typical cost, UK | Per seat, priced in local currency | £6,000-14,000 for a portal or dashboard; £12,000-24,000 for a multi-part tool | £1,600 / £2,800 / £6,000 a month |
| Speed to start | Hours to days | 2-10 weeks depending on size, after scoping | Two-week ops audit, first build live in 2-6 weeks |
| Who maintains it | Whoever built it on your team | You, unless you buy support | We do, on the plan |
| Best for | Simple internal tools a keen team member will own | One defined build with an owner afterwards | Core tools that must keep working, rolled out to the team |
| Worst for | Client-facing or complex tools, or when the builder leaves | Small tools that change often | Simple tools a no-code app already handles |
The Airtable price is from Airtable, October 2026. Custom build ranges are typical, compiled from published rate guides, from our custom software cost guide. Our prices are from our pricing page.
We use both, and here is why
This is not a page arguing that no-code is bad. No-code tools have made it possible for non-technical people to build useful software, and for many jobs they are the right answer. We happily recommend them, and some of the tools we look after for clients started life in Airtable or Notion.
The question is never "no-code or code?" in general. It is "for this tool, at this firm, with these people, which one will still be working well in two years?"
What no-code is great at
Speed.
Someone on your team can have a working tracker by the afternoon.
Ownership.
The person who uses it can change it. No tickets, no waiting.
Low cost at small scale.
A few seats at $20 a month is far cheaper than any build.
Trying ideas.
No-code is the best way to find out whether a tool is even worth having before you pay for a proper one.
For a people business - a business that sells its people's expertise and time, such as an agency, accountant, law firm or consultancy - no-code is often the right first version of a client tracker, a content calendar or a simple intake form.
Where no-code runs into trouble
These are the patterns we see when firms come to us from no-code tools:
The builder leaves.
The one person who understood the base, the automations and the views moves on. Nobody else dares touch it.
Per-seat costs grow with the team.
At Airtable's Team price of $20 a user a month billed annually (Airtable, October 2026), 20 people is $4,800 a year and 30 people is $7,200, for one tool. Business is $45 a user a month. That is still often cheaper than custom, so do the sum before assuming otherwise.
Logic outgrows the builder.
Many linked tables, formulas and automations become hard to follow and easy to break.
Client-facing needs.
Branded portals, per-client permissions and a polished experience stretch no-code tools further than they were built for.
Several apps need to stay in step.
No-code tools connect to others through automations, which add another layer to look after.
If none of those apply to you, stay on no-code.
How to decide: a simple framework
Ask four questions about each tool.
- Who uses it? Internal only, or clients too? Client-facing tools need more polish and more care.
- What breaks if it stops? A team tracker going down is annoying. A client portal going down is a phone call.
- Who owns it? Is there a person on your team who wants to look after it, and would anyone else understand it if they left?
- How complex is it? A list with a few views is a no-code job. Many linked records, rules per client and connections to five other apps is a code job.
Mostly "internal, annoying, yes, simple" means no-code. Mostly "clients, a phone call, nobody, complex" means custom code.
Airtable vs a custom app
Airtable is a flexible spreadsheet-style app with views, forms and automations. It is excellent for internal tracking and quick tools. A custom app is better when the tool becomes the place your team runs client work, when permissions get complicated, or when you are paying for many seats to give people a simple screen. See our Airtable alternative page for the detailed comparison.
Softr vs custom
Softr builds portals and internal tools on top of your data with no code. It is a fast way to give clients or staff a login and a view. Custom makes sense when the portal needs to read from several apps at once, behave differently per client, or look and feel fully like your brand. More on our Softr alternative page.
Notion and HoneyBook
Notion is a great home for docs, wikis and light project tracking, and many firms should keep it there. It struggles as a client-facing system or a source of numbers. HoneyBook suits solo and small service businesses for proposals, contracts and payments; firms often outgrow it as the team and process grow. Our Notion alternative and HoneyBook alternative pages cover when to stay and when to move.
What moving from no-code to code looks like
If a tool does outgrow no-code, the move is less dramatic than people fear. The no-code version has already done the hardest part: it shows exactly how your team works, which fields matter and which views people actually open. That is the brief.
A sensible move happens in steps. First, the new tool reads from the same information, so nobody has to rekey anything. Then a small group of the team uses it alongside the old one for a week or two, and says what is missing. Then everyone switches, with a short walkthrough so nobody is left guessing. The old base stays read-only for a while, in case anyone needs to look something up.
What you should not do is switch everything overnight, or let the person who built the no-code version feel their work was wasted. It was not. They designed the tool; the code just makes it sturdier. Bringing them into the rollout is usually the fastest way to get the rest of the team using it.
Pick no-code if...
- The tool is internal and simple.
- Someone on your team wants to own it, and others could pick it up.
- Seats are few, or the per-seat cost is still lower than a build.
- You are testing whether the tool is worth having at all.
Pick custom code from an agency or freelancer if...
- You have one clearly defined tool to build.
- Someone can look after it once it is handed over.
- The scope will not change much.
Pick custom code on a monthly plan if...
- The tool is core to how you serve clients, or connects several apps.
- Nobody on your team wants to own it technically.
- You want it rolled out properly, with the team trained, so people actually use it.
We work in four steps: scope, build, roll out, improve. A two-week ops audit ($2,500 / £2,000) maps your apps and counts the hours your team loses; it is credited to your first month. Monthly plans are $2,000, $3,500 or $7,500 (£1,600 / £2,800 / £6,000), with your first build included and a 3-month minimum. The software runs on your own accounts, so you own it. More on our custom software for small business page.
Our recommendation: how we'd decide for a 20-person firm
- Start with no-code for anything new and internal. It is the cheapest way to learn what you actually need.
- Keep it there while it works. If the tool is used, understood by more than one person and not breaking, there is no reason to move.
- Move to code when the framework says so. Client-facing, critical, ownerless or complex - any two of those is a strong signal.
- Use what you built as the specification. A no-code tool your team has used for a year is the best brief a developer can get.
- Do the seat sum. Compare three years of seats against a build plus upkeep before deciding. Sometimes the seats win.
Plainly: if your team is happy in Airtable or Notion, stay there. We would rather you did than pay us for something you do not need.
Worked example: a client portal, no-code vs code
Take one real job: a client portal, one branded home per client where every deliverable, brief and sign-off lives. Our custom software cost guide puts the typical market cost of a client portal at $8,000-18,000 in the US or £6,000-14,000 in the UK, taking 2-6 weeks (typical, compiled from published rate guides). At Fame, producers stopped hunting through Drive folders and answering "where is the latest version?" emails.
With no-code
Airtable holding the deliverables and a portal builder like Softr on top can give clients a login and a list within days. Seat costs for the team using Airtable Team would be $4,800 a year at 20 people ($20 x 20 x 12). For a simple portal, this may be all you need. It gets harder when files live in Drive, status lives in your project tool, and sign-offs need to flow back to both.
With custom code from an agency or freelancer
A quote in the $8,000-18,000 (£6,000-14,000) range, built in 2-6 weeks, reading straight from the tools your team already works in. Upkeep is extra.
With Forge
The ops audit is credited to month one, and the portal is the first build, included. On the $2,000 / £1,600 plan, the 3-month minimum is $6,000 / £4,800 in total, including rollout to your team and clients, and we look after it after that. If a no-code portal does the job, start there and come back if it stops fitting.
Across all of Fame's tools, the team got back 5,460 hours as of 14 October 2026.
Questions, answered plainly
What are the downsides of using a no-code platform?
Tools depend on the person who built them, per-seat costs grow with the team, complex logic becomes hard to follow, and client-facing polish is limited. For simple internal tools, these rarely matter.
Does no-code actually work?
Yes, for the right jobs. Simple internal trackers, forms and first versions of tools work very well in no-code. It struggles with complex, client-facing or critical tools.
What are the downsides of using Airtable?
Per-seat pricing ($20 a user a month on Team, billed annually) adds up as a team grows, large bases get hard to follow, and it depends on whoever built it. It remains a strong choice for internal tracking.
Is there anything better than Airtable?
It depends on the job. For docs and wikis, Notion; for client portals on top of your data, a builder like Softr; for core, complex or client-facing tools, a custom app built around how your firm works.
What are some examples of no-code tools?
Airtable for spreadsheet-style tracking, Softr for portals, Notion for docs and light projects, and Zapier, Make or n8n for automations between apps.