Services / Enterprise App Development
Systems that run
your operations
Custom platforms for companies that outgrew spreadsheets and can't bend to off-the-shelf software. Built around your actual process, connected to the systems you already run, and rolled out to people who'll actually use it.
Free consultation. If an existing tool solves your problem for a fraction of the cost, we'll point you to it.
Overview
Sound familiar?
How operations quietly stop scaling
The company runs on a spreadsheet, and only one person truly understands it
The same data gets typed into three systems, by hand, every day
The tool you bought forces your process to fit it, and you still export to Excel
You pay per seat for modules nobody opens
The internal system works, but the person who built it left years ago
Monthly reports take three days to assemble and are outdated on arrival
Something went wrong last quarter and nobody can reconstruct who changed what
Access control is “everyone can see everything, please be careful”
Adding a warehouse, a country or a subsidiary means duplicating all of it again
Every one of these has the same root cause: the process grew, the tooling didn't.
What are we building?
Six shapes this usually takes
Enterprise projects rarely start from nothing — they start from a process that already works, badly. Tell us where the friction is and we'll tell you the smallest system that removes it.
Internal operations platform
Orders, projects, resources, production, inventory. One place instead of spreadsheets, email and three tools.
Client or partner portal
Self-service for the people you serve: orders, documents, status, invoices. Fewer phone calls, fewer emails.
Legacy modernisation
A system that still works but nobody can maintain. Migrated in phases, without stopping the business.
Integration layer
Your ERP, CRM, WMS and accounting talking to each other. No more manual re-entry between them.
Custom module on top of what you keep
You're not replacing SAP or Salesforce. You're adding the part it can't do.
Reporting and BI layer
One source of truth, dashboards your board trusts, reports that assemble themselves.
Not sure whether to build, buy or fix what you have? That's exactly what the first conversation is for — and we'll give you the answer even when it isn't “build”.
What you can hold us to
- Day 1
- You own the code and the infrastructure
- Fixed scope
- Paid discovery before any build commitment
- NDA + DPA
- Signed before we see your data
- Phased rollout
- Pilot team first, never a big-bang cutover
What solutions does your company need?
The building blocks
Most enterprise systems are assembled from the same set of parts. What differs is your process — that's the part we design from scratch.
Workflow and approval automation
The handoffs that currently live in email, made explicit, tracked and enforced.
- Process mapped once, then run the same way every time
- Approval chains with limits, deputies and escalation when nobody responds
- Conditional routing — different path for a 500 and a 500,000 order
- Task assignment with clear ownership and deadlines
- Steps that trigger actions in other systems automatically
- Nothing waits in someone's inbox unnoticed
Process flow · active step · owner avatar · SLA counter
One source of truth for your data
The end of “which version of this file is current”.
- Data model built around your real entities, not a generic CRM's
- Validation at entry, so bad data never gets in
- Search, filtering and bulk editing that handle real volumes
- Import, export and migration tools your team can run themselves
- Full change history on every record
Access control, SSO and audit trail
Who can see what, who did what, and proof of both.
- Roles and granular permissions down to field level
- Single sign-on with Microsoft Entra ID, Google Workspace or your identity provider
- Automatic access removal when someone leaves
- Immutable audit log: who changed what, when, and what it was before
- Session controls, two-factor authentication, IP restrictions where needed
Role × module permission matrix · audit entry with before/after
Reporting and dashboards
Numbers your board trusts, without anyone assembling them by hand.
- KPI dashboards per role — the warehouse sees different figures than the CFO
- Scheduled reports delivered by email on the first of the month
- Drill-down from a summary figure to the individual transaction
- Export to Excel, because someone will always want it
- Connection to Power BI or your existing BI stack if you have one
Dashboard · 4 KPIs · trend chart · drill-down
Integrations, including the difficult ones
Your systems talking to each other, including the ones that were never meant to.
- ERP, CRM, WMS, accounting, HR, e-commerce
- Custom middleware where there's no usable API
- Legacy protocols nobody advertises support for: SOAP, SFTP, CSV drops, direct database reads
- Two-way sync with conflict rules, not overnight batch guesswork
- Retry logic, error queues and alerts when an integration fails — because it will
Documents and signatures
The paperwork layer, generated instead of typed.
- Contracts, offers, orders and protocols generated from templates
- Data filled from the system — no copy-paste, no wrong client name
- Versions, revisions and approval before sending
- Electronic signature integration
- Central storage with permissions and full-text search
Template → generated PDF → signed, with date
Notifications, alerts and escalation
The system tells people what needs attention, instead of waiting to be checked.
- Rules you define: stock below threshold, overdue payment, stalled approval
- In-app, email, SMS or Slack, depending on urgency
- Escalation upward when nothing happens
- Digest instead of noise, so alerts don't get filtered to a folder
Alert rule · three-level escalation path
Legacy migration without a shutdown
Replacing a working system is a different discipline from building a new one.
- Audit of the existing system, its data and its undocumented behaviour
- Data migration with reconciliation — counts and totals matched before and after
- Both systems running in parallel until the numbers agree
- Phased cutover by module or by team, with a rollback plan at each step
- The old system stays available read-only, so nobody loses history
gateway5/5
Security, compliance and reliability
The section your client's IT department will read first.
- GDPR by design: data minimisation, retention policies, subject access requests
- Encryption in transit and at rest, secrets managed properly
- Penetration testing before go-live, and after major releases
- EU hosting or your own infrastructure, whichever your policy requires
- Backups with a tested restore procedure — untested backups don't count
- Monitoring, alerting and a documented incident path
- NIS2 and sector-specific requirements where they apply to you
Compliance checklist with statuses · uptime monitoring panel
AI where it removes real work
Not a chatbot bolted on. Specific work taken off specific desks.
- Documents read automatically: invoices, orders, delivery notes, specifications
- Anomaly detection: a price, a quantity or a pattern that doesn't fit
- Questions answered in plain language: “which orders are late and why”
- Draft classification, routing and summarising of incoming requests
- Assistant that answers from your own procedures, for new staff
Turn a note into a post
Ready to publishGrateful for the opportunity to attend AI DevSummit 2026 in 🇩🇪 Berlin and co-host a session with our partner doubleuprompt and the people who made it click: Kamil & Michael. 🤝 Alongside supporting the event, I also recorded several interviews with industry experts (and welcomed the first guest on our podcast, Juliet 🎙️), gaining real insight into AI-driven software and the future of how we build. A great experience connecting with founders and seeing innovation in action. Excited for what's next! ❤️🔥 Huge thanks to the doubleuprompt crew 🤝 #AIDevSummit #SoftwareHouse #AI #Innovation #doubleucode
Adoption, training and handover
The part that decides whether any of the above matters.
- Pilot with one team before the whole company sees it
- Training tailored per role, not one generic session
- Documentation and short recordings your team keeps
- A named internal owner we prepare to take over
- Hypercare period after go-live, with someone reachable the same day
Rollout schedule: pilot → training → rollout → hypercare
Your process doesn't look like any of this? Good — that's usually the reason off-the-shelf software didn't work. Tell us how it actually runs.
Talk to usBoring technology, chosen on purpose
React, Node, TypeScript, PostgreSQL, Docker, Kubernetes, Azure and AWS — mainstream, well-documented and hireable. When we hand the system over, your team or your next vendor will recognise every part of it.
Selected work
Systems in daily use
Three implementations, three different starting points. Anonymised where clients require it — internal systems are rarely public.
Production, inventory and orders in one platform
Before: four plants, four ways of working, monthly figures reconciled by hand across a dozen spreadsheets. What we built: one platform covering production, stock and order intake, with permissions per plant.
Order management with WMS and ERP integration
Before: orders arriving by email and phone, re-keyed into two systems, with dispatch decided from a regional spreadsheet. What we built: one order pipeline synced both ways with the WMS and the ERP.
Project and resource management platform
Before: sixty people planning capacity in personal spreadsheets, with utilisation known only at the end of the month. What we built: one place for projects, allocations and time, with live utilisation per team.
One backbone, every department. Finance, operations and the floor working from the same data instead of ten systems that disagree — one platform the whole company can run on.
Client feedback
What clients say a year later
—
Systems in production
—
Years in business
—
Of clients on ongoing development
—
Average rating
Month-end close used to take five days and two people. It takes one afternoon now, and the numbers match the warehouse for the first time.
Manufacturing
Production and inventory platform, 140 users
The previous vendor delivered exactly what the spec said, which turned out not to be what we needed. These guys spent two weeks on our floor first and rewrote the spec.
Logistics
Order management with WMS and ERP integration
We expected the hard part to be the software. It was getting sixty people to stop using their own spreadsheets — and that was the part they had a plan for.
Professional services
Project and resource management platform
How we work
From process audit to daily use
Enterprise projects fail in predictable places: unclear scope, unmigrated data and people who never adopted the thing. Our process is built around those three.
01
Free consultation
30 minutes. What the process looks like, where it breaks, what you've already tried. You leave with a direction and a rough range, whether or not we work together.
02
Paid discovery
2–4 weeks. We map the process with the people who actually do it, review your data and existing systems, and produce a scope, architecture, risk list and a real estimate. Fixed price, and it's yours to keep — including if you take it to another vendor.
03
Design and prototype
Clickable version of the core workflow, tested by your team before anything is built. Cheapest possible moment to discover we misunderstood something.
04
Build in two-week sprints
Working software on a staging environment from sprint one. Demo every second Friday, priorities reviewed each time.
05
Data migration
Legacy data mapped, cleaned and reconciled. Totals matched before and after, in writing.
06
Pilot with one team
A real group using it for real work while the old process still exists as a fallback.
07
Rollout and training
Module by module or team by team. Training per role, documentation, recordings.
08
Hypercare
An agreed period after go-live with fast response, then a transition to a support model or to your own team.
09
Ongoing development
Because a system that stops evolving starts being worked around.
Typical timeline: 3–4 months for a focused system, 6–9 for a platform with several integrations, roles and compliance requirements. Discovery gives you a real number instead of a range.
Working together
How we structure
the engagement
Four ways to work with us, depending on how defined the work is and what happens after launch. If you're not sure, we'll tell you which one we'd pick in your position.
Our Estimation Process
We listen
The process, the pain, the systems involved, who decides, what the deadline is driven by. 30 minutes.
We scope
In discovery, with your people. Output: architecture, phasing, risks, and a price you can put in a budget.
We commit
We walk your team and your IT through it, adjust, and phase it so the first value lands early rather than at the end.
Discovery, then fixed price
Best for
- A defined system with a clear process boundary
- Budget that has to be approved as one number
Scope becomes clear in discovery, then the build is priced against it.
Key benefits
- One agreed price for a defined scope
- Milestones, deliverables and acceptance criteria in the contract
- Change requests handled through a written, priced process — no arguments in month four
Time & materials
Best for
- A platform that will keep growing
- Multi-phase platforms and long modernisations
The direction is set, the details will change as users get their hands on it.
Key benefits
- Reprioritise every sprint without renegotiating the contract
- Transparent time log and a monthly burn report
- A monthly cap if your finance team needs one
Dedicated team
Best for
- When you need capacity, not a project
An extension of your organisation, managed by you.
Key benefits
- A named team that stays with you and learns your domain
- Scale up or down with notice
- Works inside your process, your tools, your standards
- Optional knowledge transfer to your own hires
Care & SLA
Best for
- After go-live
The system stays healthy and keeps changing, so nobody starts working around it.
Key benefits
- Response and resolution times in writing, by severity
- Monitoring, backups with tested restores, security patching
- Fixed development hours each month for changes
- A named person who knows your system
Working with procurement, IT or a compliance team? We're used to it. NDA, DPA, security questionnaires, penetration test reports, insurance certificates, source-code escrow and vendor onboarding forms — send them over and we'll work through them.
doubleucode · Vendor onboarding & procurement · Call: +48 667 083 196
Frequently Asked
Questions
Can't find the answer to your question? Just send me an email:
Ask me a questionShould we build this or buy something off the shelf?
Buy, whenever a standard product fits. Building makes sense when your process is a genuine competitive difference, when the per-seat cost of a SaaS across hundreds of users exceeds the build, or when you've already tried a product and your team quietly went back to Excel. We'll give you the honest answer in the first conversation, including the version where we don't get the project.
How much does a system like this cost?
A focused internal system with clear workflows typically starts around $8k. A platform with several integrations, role hierarchies, reporting and compliance requirements runs into the low six figures. Discovery is a fixed fee agreed up front and produces a real number rather than a range. Worth comparing against what you currently pay in licences and manual hours.
How long does it take?
Roughly 3–4 months for a focused system, 6–9 for a larger platform. You'll see working software in the first month and a pilot team using it well before full rollout — you won't wait until the end to find out whether it works.
We already have SAP, Dynamics, Salesforce. Do we have to replace them?
Almost never, and we'd usually advise against it. The common pattern is that your ERP handles what it's good at while a custom layer handles the process it can't model, with the two kept in sync. Replacing a working ERP is one of the most expensive decisions a company can make.
Can you integrate with a system that has no API?
Usually yes. We've worked with SOAP services, scheduled CSV and SFTP exchanges, direct database reads and screen-level automation where nothing else existed. If it truly can't be reached, we'll tell you early and propose a workaround rather than discovering it in month three.
What happens to our data in the old system?
It gets migrated, reconciled and verified — record counts and financial totals matched before and after, documented. The legacy system usually stays available read-only for a period so nobody loses history, and we don't switch anything off until the numbers agree.
Who owns the code, and what if you disappear?
You own everything: source code, documentation, infrastructure, data. It sits in your repository and your cloud account from day one, not ours. We use mainstream technology specifically so another team can pick it up, we can set up source-code escrow if your procurement requires it, and handover documentation is a deliverable rather than a favour.
Can our own IT team maintain it afterwards?
Yes, and some clients plan for exactly that. We can build with your standards and your stack, involve your developers during the project, and run a structured handover with documentation and training. You're not obliged to keep paying us to stay in control of your own system.
Cloud or our own servers?
Whichever your policy requires. EU cloud regions by default, your own data centre or private cloud where regulation, a client contract or an internal rule demands it. We'll tell you honestly what on-premise costs you in flexibility and operational overhead.
How do you handle GDPR and sector regulation?
Built in from the architecture: data minimisation, retention rules, subject access and deletion, encryption, full audit logging. We sign a DPA before touching production data, maintain a record of subprocessors, and work with your compliance team on anything sector-specific — NIS2, financial or healthcare requirements. We'll say plainly where something needs a specialist auditor rather than a developer.
Will it work with our single sign-on?
Yes — Microsoft Entra ID, Google Workspace, Okta, on-premise Active Directory. Your staff use the accounts they already have, and access ends automatically when someone leaves.
How do you stop our people going back to spreadsheets?
By involving them in discovery, so the system reflects how the work actually happens rather than how a manager described it. Then a pilot team, role-specific training, and a hypercare period where problems get fixed within days. When people abandon a new system, it's almost always because it made one daily task slower than before — that's what we look for in the pilot.
What if requirements change mid-project?
They will. On time & materials you simply reprioritise the next sprint. On fixed price, changes go through a written, priced change process — you always know the cost before agreeing, and we don't treat scope changes as an accusation.
Can we start small?
Yes, and usually you should. One process, one team, one measurable result — then expand once it's proven. It's also the cheapest way to find out whether we're the right vendor for you.