DEV Community: John Moore

How I Built 7 Shopify Apps as a Solo Developer in Ireland

John Moore — Sun, 22 Mar 2026 21:19:58 +0000

I am not a 22-year-old fresh out of college with a laptop and a dream. I am 47, I have been running a jewellery retail business in Cork, Ireland for over two decades, and I am only now — after closing that business — turning back to software full time. This is the story of how I built seven production applications as a solo developer, and why it took running shops for 22 years to figure out what software actually needed to exist.

The Long Way Round

I studied Computer Science at University College Cork, graduating in 2001. My first job was at Sun Microsystems, where I spent three years working on enterprise systems. It was rigorous, disciplined work. I learned how large-scale software gets built properly.

Then in 2004, I took over the family jewellery business — Moores Jewellers, a multi-site retail operation across Cork. I ran it for 22 years. Four locations at our peak: Douglas Court, Mahon Point, Carrigaline, and Bandon. I served on the board of the Association of Fine Jewellers and held the role of Vice President.

During those 22 years, I never stopped thinking like an engineer. Every spreadsheet I maintained, every process I cobbled together with paper and manual effort, every problem I watched other retailers struggle with — I filed it away. When the time came to wind down the business in 2026 and launch JMS Dev Lab as a software development practice, I did not need to guess what to build. I had a list.

Why Seven Apps

The honest answer: because I kept finding problems that existing software did not solve, or solved badly, or solved at a price point that made no sense for small retailers.

Each app started the same way. I would hit a wall in my own business, look for a tool, find nothing adequate, and think: I could build this. After a while, the pattern became a pipeline. Here is what came out of it.

SmartCash — Cashflow Management

Every small retailer I know has the same problem: they can tell you their revenue, but they cannot tell you their actual cash position next month. SmartCash is a Shopify embedded app that does cashflow forecasting with AI-powered projections. It connects to a merchant’s Shopify data and turns sales history into forward-looking cash visibility. This was the app I wished I had during every January when jewellery retail goes quiet and the bills do not.

Jewel Value — Jewelry Valuations

Insurance valuations are bread and butter for any jeweller, but the process is archaic. Most shops still type certificates in Word documents. Jewel Value generates professional valuation certificates directly from a Shopify store, with support for 13 languages. I built it because I was tired of the formatting headaches and because jewellers everywhere have the same problem.

RepairDesk — Repair Ticket Management

Repairs are a significant revenue stream for jewellers and watchmakers, but tracking them is chaos in most shops. Handwritten tickets, no status updates for customers, items lost in the back room. RepairDesk is a repair ticket management system built specifically for this workflow. It tracks items from intake through to collection, with status updates and a clear audit trail.

StaffHub — Team Training

Training retail staff is repetitive and inconsistent. Every new hire gets a different version of the same information depending on who trains them. StaffHub is a team training and management app that standardises onboarding and ongoing development. I built it after watching the same training problems recur across multiple shop locations for years.

GrowthMap — Marketing Planner

Small retailers know they should be doing marketing but have no idea where to start. GrowthMap breaks marketing down into structured, actionable tasks — like Duolingo, but for marketing plans. It guides merchants through execution rather than just telling them to "do social media."

JewelryStudioManager — CRM

Custom jewellery work involves long conversations with customers, design iterations, approvals, and timelines. Managing all of that in email and notebooks is a recipe for missed details and unhappy clients. JewelryStudioManager is a CRM built specifically for jewellery studios handling custom and bespoke work.

Pitch Side — Football Coaching (The Odd One Out)

Not every app comes from retail. I coach underage football at Carrigaline AFC, and I found the same tooling gap there. Pitch Side is a free coaching companion for grassroots football — session planning, squad management, that sort of thing. It is built on Next.js with Firebase and Capacitor for mobile. No Shopify involvement, no subscription, completely free. I built it because grassroots sports clubs should not have to pay for basic organisational tools.

The Tech Stack

All six Shopify apps are built with TypeScript, React, and Node.js. Beyond that, the specifics vary by app:

Frameworks: Next.js for most frontends, NestJS for some backends, Express for others
Databases: PostgreSQL with Prisma ORM for most apps, MongoDB for StaffHub, Supabase for GrowthMap, Firebase for Pitch Side
Shopify integration: App Bridge, Polaris UI components, Shopify Billing API, session tokens
Hosting: Railway for backends, Vercel and Cloudflare Pages for frontends
Architecture: Most apps use a monorepo structure (apps/shopify, apps/web, apps/backend, packages/shared) so the Shopify embedded version and a standalone web version can share code and data

I settled on TypeScript early and have not looked back. Strict mode everywhere. The type safety pays for itself when you are the only person reviewing your own code.

What I Learned

Domain expertise is an unfair advantage. I did not need to interview jewellers to understand their problems. I was the jeweller. I knew which workflows were painful because I had lived them for two decades. If you are building software for an industry, there is no substitute for having actually worked in it.

Solo does not mean sloppy. When there is no team to catch your mistakes, you build habits fast. Automated testing, strict linting, proper error handling, GDPR compliance — these are not optional extras when you are the only engineer, the only support person, and the only one who gets the email when something breaks at 2am.

Ship before it is perfect. Every one of these apps has rough edges I know about. The alternative was shipping nothing. I chose to get working software in front of real users and iterate from there.

Seven apps is probably too many. I will be honest about this. Building seven apps means spreading attention thin. Each app needs documentation, support, marketing, updates, and ongoing Shopify review compliance. If I were advising someone else, I would say pick two or three and go deep. But these apps exist because the problems exist, and I could not bring myself to leave them half-built.

The Shopify review process is thorough. Submitting apps to the Shopify App Store is not a rubber stamp. They check for GDPR webhook compliance, geographic requirements, UI consistency, and more. It is demanding, but it forces a quality standard that benefits merchants.

Where Things Stand

I am not going to dress this up with vanity metrics. I have no thousands of installs to report. These apps are newly submitted, the business is newly launched, and I am still winding down Moores Jewellers. What I have is seven working applications, each solving a real problem I experienced firsthand, built to production standard with proper architecture.

JMS Dev Lab is a sole trader software development practice based in Cork, Ireland. I build Shopify apps and custom software for small businesses. If you are a developer thinking about building for Shopify, or a retailer wondering whether your workflow problems have software solutions, I am happy to talk.

You can find everything at jmsdevlab.com.

John Moore is the founder of JMS Dev Lab, a software development practice in Cork, Ireland. He builds Shopify apps and custom business software. Previously, he spent 22 years running Moores Jewellers and 3 years as a software engineer at Sun Microsystems.

Internal Tools vs SaaS: When to Build Custom Software

John Moore — Sun, 15 Mar 2026 13:20:30 +0000

Every growing business hits the same crossroads. You need software to solve a problem, and you have two options: buy something off the shelf, or build something custom. Both have their place. The trick is knowing which one fits your situation.

When Off-the-Shelf SaaS Works

SaaS products are brilliant when your problem is generic. Email, accounting, project management, CRM basics — these are well-understood problems with mature solutions. Tools like Xero, Slack, and Trello exist because millions of businesses need roughly the same thing.

The rule of thumb: if the problem is common and your process is standard, buy.

When SaaS Starts to Hurt

The cracks appear when your business does something the software wasn't designed for. Warning signs:

You're paying for 5 tools to do one job. CRM + form builder + spreadsheet + Zapier glue = you've outgrown the buy approach.
You're working around the software, not with it. If your team has a list of "the system can't do that, so we do it manually," that's a red flag.
The tool dictates your process. Your business should define how you work, not your software vendor.
You're paying enterprise prices for small-business needs. Per-seat pricing spirals as you grow.
Data lives in silos. Customer data in one tool, orders in another, reporting in a third.

The Real Cost of "Cheap" SaaS

SaaS looks affordable on the surface. But add up subscription costs, workaround time, broken integrations, and features you've been requesting for two years — "cheap" SaaS can end up costing more than building something purpose-built.

When Custom Makes Sense

Good candidates for custom builds:

Internal workflows unique to your operation. How you process orders or manage commissions might be different enough that no SaaS fits.
Client-facing portals. Branded, secure, integrated with your systems.
Replacing a tangle of spreadsheets and disconnected tools. One system, one source of truth.
Automating repetitive manual processes. Hours of copy-paste work every week? That's a prime automation candidate.

The Build Doesn't Have to Be Big

The best custom tools are focused. They solve one problem well. Replace the most painful spreadsheet, automate the most tedious process, or give your clients the one thing they keep asking for. Start small, prove the value, expand from there.

How to Decide

Ask yourself three questions:

Is my problem generic or specific? Generic → buy. Specific to your business → consider building.
Am I working around my tools? Significant workaround time means the "cheap" option is costing you more than you think.
Would a competitor gain an advantage by using the same tool? If your software is the same as everyone else's, your operations are too. Custom tools can be a genuine competitive edge.

Originally published at jmsdevlab.com

Buy vs Build: When Should a Small Business Build Custom Software?

John Moore — Sun, 15 Mar 2026 13:10:24 +0000

You need software to run part of your business. Maybe you've already tried a few tools and nothing quite fits. Maybe you're drowning in spreadsheets. Maybe a well-meaning friend said "just get a developer to build it."

The answer isn't always custom software. Sometimes it is. And sometimes the right answer is somewhere in between. Here's a practical framework for deciding.

The Three Options

Buy (off-the-shelf SaaS) — Subscribe to an existing product. Xero for accounting, HubSpot for CRM, Monday.com for project management.
Build-lite (no-code/low-code) — Use platforms like Airtable, Notion, Zapier, or Glide to assemble something yourself without writing code.
Build (custom development) — Hire a developer to create software tailored exactly to your business.

None of these is inherently better. The right choice depends on your problem, your budget, and where you are as a business.

When to BUY: Off-the-Shelf SaaS

Buy when:

The problem is well-understood and generic (accounting, email, basic CRM)
Your workflow matches what the tool was designed for
You have fewer than 10 users
You need something today, not in six weeks

If a well-known SaaS tool does 80% of what you need and the other 20% isn't critical, just use it. Pay the subscription. Move on.

When to BUILD-LITE: No-Code

Build-lite when:

You're prototyping and aren't sure what you need yet
Your team is small (fewer than 5 people)
The workflow is relatively simple
Someone on your team is comfortable with tools like Airtable

Think of no-code as a proving ground. Figure out what you need before investing in building it properly.

When to BUILD: Custom Development

Build when:

Your workflow is niche. No SaaS tool fits without heavy modification.
Per-seat costs are killing you. At 10+ users, SaaS subscriptions compound fast.
Data sensitivity matters. Full control for GDPR and data residency.
The tool IS your competitive advantage. Why use the same software as your competitors?
You've outgrown everything else. Five subscriptions, three spreadsheets, and a shared email inbox is not a system.

A Simple Decision Framework

Is this a common business problem? → Buy SaaS
Is my process still evolving? → Start with no-code
Can an existing tool handle 80%+ of what I need? → Buy it
Fewer than 5 users and simple workflows? → No-code is fine
Per-seat pricing becoming painful? (10+ users) → Custom makes sense
Is this process core to my competitive advantage? → Build custom
Have I outgrown SaaS and no-code? → Build custom

The Hybrid Approach

The smartest businesses buy for the common stuff and build for the parts that make them different. Use Xero for accounting. Use Shopify for your storefront. But that client portal, commission tracker, or internal dashboard? Build those.

3-Year Cost Comparison (10 users, workflow management)

Option	3-Year Total
SaaS (Buy)	£18,000–36,000
No-Code (Build-Lite)	£9,000–20,000
Custom (Build)	£5,500–21,000

SaaS has the lowest entry cost but highest long-term cost. Custom has the highest upfront cost but flattens out. No-code sits in between.

The "Missing Middle"

There's a gap in the market. You're too complex for off-the-shelf SaaS but too small for agencies charging £50,000+. That's exactly where solo developers and small studios operate. Focused, purpose-built tools without the overhead.

Originally published at jmsdevlab.com

No-Code vs Custom Software: When Free Tools Stop Being Free

John Moore — Sun, 15 Mar 2026 13:09:05 +0000

No-code platforms are genuinely impressive. Airtable, Bubble, Glide, Retool — they let non-technical people build real applications without writing a line of code. For prototyping, for solo founders, for simple internal tools, they can be exactly the right choice.

But there's a pattern we see over and over again: a business adopts a no-code platform, builds something useful, the team grows, the requirements get more complex — and suddenly the "affordable" tool costs more than custom software ever would have.

This isn't an anti-no-code article. It's an honest look at the numbers, so you can make the right decision for your business.

The Real Cost of No-Code Platforms

No-code pricing looks reasonable when you first sign up. But most platforms charge per seat, per builder, or both — and those costs compound fast as your team grows.

Airtable — Free for basic use, then $20–$45 per seat per month
Bubble — $29/mo for Starter, up to $349/mo for Team
Glide — $199/mo for the Business plan
Retool — Free for small use, then $10–$50 per builder per month

The Per-Seat Trap

Per-seat pricing is the reason no-code costs catch people off guard. At 20 users on Airtable Business ($45/seat/mo), you're paying nearly $11,000 a year — for what started as a "cheap" spreadsheet replacement.

Custom software, by contrast, costs the same whether you have 5 users or 500. You pay once to build it, and you're done.

The Hidden Costs Nobody Mentions

Consultant fees — Most businesses doing anything beyond the basics hire a no-code consultant at $40–$200 per hour. Surprisingly close to what a custom developer charges — except the consultant's work is locked inside a platform you don't own.

Platform lock-in — Everything you build on Airtable belongs to Airtable. If the platform raises prices, changes its API, or shuts down, you start over.

Migration costs — When you outgrow a no-code platform, the migration isn't free. Your business logic needs to be reverse-engineered and rebuilt.

3-Year Cost Comparison

Spreadsheet Replacement (10 users)

Approach	3-Year Total
Airtable Business	$16,200
Custom software	$5,200

Client Portal (15 users)

Approach	3-Year Total
Bubble Team	$12,564
Custom software	$7,800

Staff Training Tracker (20 users)

Approach	3-Year Total
Airtable Business	$32,400
Custom software	$7,100

The Crossover Point

Under 5 users — No-code is often cheaper
5–10 users — Depends on complexity
10–20 users — Custom frequently wins on total cost
20+ users — Custom almost always wins

GDPR and Data Sovereignty

Airtable, Bubble, Glide, and Retool are all US-based, which means your data is subject to the US CLOUD Act. For European businesses handling customer data under GDPR, this creates a genuine compliance risk. Custom software can be hosted on EU infrastructure with full data sovereignty.

When No-Code IS the Right Choice

Prototyping and validation — Testing an idea before committing
Solo founders and tiny teams — Per-seat pricing isn't a problem
Simple, stable requirements — Basic forms, simple databases
Temporary or experimental tools — Short-term projects

The Bottom Line

No-code platforms are brilliant tools with a specific sweet spot: small teams, simple requirements, fast iteration. Outside that sweet spot, they become expensive, limiting, and risky.

Before you commit to a no-code platform for a business-critical tool, run the numbers. Project your team size over three years. Then compare that to a fixed-price custom build with flat hosting costs and no per-seat fees.

Originally published at jmsdevlab.com

How Much Does Custom Software Actually Cost? A Straight Answer

John Moore — Sun, 15 Mar 2026 13:02:53 +0000

If you've ever searched "how much does custom software cost," you've probably found a lot of articles that say the same thing: it depends.

That's not helpful. You're a business owner trying to work out whether custom software is even realistic for your budget. You need numbers, not waffle.

So here are real numbers. These are based on the kind of projects we actually build at JMS Dev Lab — tools for small and medium businesses in Ireland and the UK, not Silicon Valley startups raising millions.

Why Most Developers Won't Give You a Straight Answer

Most agencies and freelancers dodge the pricing question because they want to get you on a call first. Others genuinely can't give you a number because they charge by the hour and have no idea how long something will take.

Both of those should worry you.

Hourly billing means you carry all the risk. If the project takes longer than expected — and they almost always do — you pay more. The developer has no incentive to be efficient.

We work differently. We quote a fixed price before any work begins. You know the total cost upfront. If we underestimate, that's our problem, not yours.

Real Price Ranges for Small Business Software

Here's what custom software typically costs for the kind of projects small businesses actually need. All prices are in euro and include design, development, and deployment.

Spreadsheet Replacement: €3,000 – €6,000

You've got a spreadsheet that's grown into a monster. Multiple tabs, complex formulas, several people editing it, constant errors. You need it turned into a proper system with a clean interface, validation, and multi-user access.

This is the most common starting point. It's also the project with the fastest payback — if your team is losing even a few hours a week to spreadsheet headaches, a €4,000 tool pays for itself within months.

Client Portal or Dashboard: €6,000 – €12,000

Your customers or clients need to log in and see their information — order status, project progress, documents, invoices. Or your team needs an internal dashboard that pulls data from multiple sources into one clear view.

This costs more because it involves user authentication, roles and permissions, and usually some kind of integration with your existing systems.

Custom Business Application: €12,000 – €25,000

A full application built around your specific workflow. This might be a booking system, a job management tool, a commission tracker, a repair ticketing system — something that doesn't exist off the shelf because your business does things differently.

What Affects the Price

Those ranges are wide because not all projects are the same. Here's what pushes a project towards the higher end:

Complexity of the logic. A simple data entry form is cheap. A system that calculates pricing based on 15 variables, applies business rules, and generates documents is not.
Integrations. Connecting to Shopify, Xero, Google Workspace, or your existing tools adds time.
Number of user roles. A tool for one admin is simpler than a system with customers, staff, and managers all seeing different things.
Custom reporting. Charts, exports, automated emails, or scheduled reports add work.
Mobile access. If your team needs to use the tool on phones and tablets, the interface needs additional design and testing.

The Hidden Costs People Forget

Hosting: €5 – €50/month

Your software needs to live somewhere. For most small business tools, cloud hosting costs between €5 and €50 a month depending on usage.

Maintenance: 10–15% of build cost per year

Software isn't "done" once it's deployed. Browsers update, security patches need applying, and you'll inevitably want changes. Budget roughly 10–15% of the original build cost per year.

Custom vs No-Code vs SaaS: The 3-Year Comparison

No-code platforms charge per user, per month. At 10 or more users, the maths changes dramatically:

No-code platform (10 users): €200–500/month = €7,200–18,000 over 3 years. And you don't own it.
Custom software (10 users): €8,000 build + €2,400 maintenance over 3 years = €10,400 total. You own it outright.

At 10+ users, custom software is typically cheaper over three years — and you end up with something that fits your business exactly.

Funding: The LEO Grow Digital Voucher

If your business is based in Ireland, the Local Enterprise Office (LEO) Grow Digital Voucher covers 50% of the cost of digital projects, up to €5,000.

That means a €10,000 custom build could cost you €5,000 out of pocket. A €6,000 spreadsheet replacement could drop to €3,000.

When NOT to Go Custom

Custom software isn't always the right answer:

Your needs are generic. If you need basic invoicing, use Xero. If you need a simple CRM, try HubSpot's free tier.
You have fewer than 3 users. At very small scale, SaaS tools are usually cheaper.
You're not sure what you need yet. Start with a spreadsheet, learn what works, and come back when you've outgrown it.
You need it tomorrow. Custom software takes weeks, not days.

How to Get Started

If this sounds realistic for your situation:

Free conversation. Tell us what problem you're trying to solve. No forms, no sales pitch.
Fixed quote. Clear scope, fixed price. No hourly rates, no surprises.
Build and deliver. We build it, you test it, you own the code outright.

Most projects go from first conversation to live software in 4–8 weeks.

Originally published at jmsdevlab.com