Skip to content
AppForGap

Custom software studio

Software for the gapbetween the work andthe people deciding.

What is actually happening and what management can see are rarely the same thing, and most of what a process costs you lives in the distance between them. We find what is worth changing, then build it: reporting, internal tools, integrations, AI where it earns its place, and apps for the people doing the work.

Fixed-scope pilots. On site in Czechia and across the EU.

The gap

Where the gap shows up

The pattern repeats whatever the company makes or sells. The people doing the work know what is happening. The people deciding find out later from a number somebody assembled by hand for a week that is already over.

Reported / Actual

  • Reported
  • Actual
This is not customer data. The shape is the point: the version that reaches management is smooth, the work never is, and nobody can act on the distance between them. Some companies can already name that distance. Others only know that something takes too long, and finding it is the first job.
  • Visibility

    Management sees last month. The floor sees today.

    Reporting is assembled by hand after the fact, so by the time a number reaches the person who could have acted on it, the week it describes is over. We put the live number in front of them instead.

  • Spreadsheets

    The process lives in one person's spreadsheet.

    It works until they are on holiday. Approvals, handovers, checklists and planning end up in files nobody else can safely touch, with no record of who changed what or why.

  • Systems

    Two systems, one person retyping.

    The ERP, the CRM, the shop, the supplier portal and the accounting export each hold a piece of the truth. Somebody keeps them in sync by hand, and that somebody is the integration.

  • Production

    On a line, the same gap is measured in minutes.

    Manufacturing is one example of many: output counted on paper, downtime remembered rather than recorded, traceability reconstructed after the complaint arrives. The same shape of problem, closer to the machine.

  • Starting point

    There is nothing to compare against yet.

    Plenty of companies cannot point at the gap because there was never anything written down to measure it against. They only know that something takes too long, costs more than it should, or depends entirely on one person. Working out what is worth changing is the first piece of work, and it does not start by buying software.

Where we work

Different industries, the same gap

Production is the example we are asked about most, and it is not the only one. The work is the same wherever the people doing something and the people deciding about it are looking at different information.

  • Manufacturing and automotive
  • Logistics and warehousing
  • Field service and maintenance
  • Wholesale and distribution
  • Professional services
  • Back office and finance

Scope

What we build

Custom software, sized to one problem at a time. The first item is not software at all.

Process analysis and an improvement plan
The deliverable is the plan, not software. We map what actually happens, find where the time and the money go, and write down what is worth changing and in what order. Some of it will not need software, and we say so.
Deliverable: written scope
Reporting and live dashboards
One live number in front of the person who can act on it, with enough history behind it to put it in context. Usually the fastest way to close the information gap.
PostgreSQL / BI
Internal tools and workflows
Approvals, checklists, handovers, planning boards. We put the process your company already follows into something built for it, rather than an overgrown spreadsheet.
Web / role-based
System integration
ERP, CRM, e-shop, accounting, supplier portals. We build the integration as an interface with a contract, not as a nightly CSV nobody owns.
REST / SQL / EDI
AI where it removes the typing
Reading invoices, orders and supplier emails into structured data. Searching your documents and procedures and answering questions in plain language. Classifying, routing and flagging the odd case for a human. Hosted in the EU or on your own hardware.
LLM / on-premises option
Mobile and field apps
For people working away from a desk: technicians, drivers, warehouse and shop-floor staff. Large controls, offline-first operation and clear behaviour when the signal drops.
iOS / Android / offline
Machine data and OEE
Where the process is a production line, we read availability, performance and quality straight from the equipment, with serial-level traceability behind it.
OPC UA / Modbus / MQTT
Custom web applications
When nothing on the market fits the process, we build the thing that does, and hand over the repository with it.
Web / self-hosted

Routing

How we work

The steps are numbered in tens, the way a routing sheet numbers operations, so another step can be inserted later without renumbering the rest.

  1. OP 10

    Walk the process

    Half a day on site. We follow the process end to end, talk to the people running it, and find the point where the information stops moving. You do not need to know what to fix beforehand, because that is what this step is for. You get a written scope out of it whether or not you continue with us.

  2. OP 20

    Fixed-scope pilot

    One process, one problem, a fixed price and a date. It runs on your real data before anyone signs up for a company-wide rollout.

  3. OP 30

    Rollout

    The pilot becomes the standard: more teams, more sites, the integrations and the user management that make it survive without us in the room.

  4. OP 40

    Run

    Monitoring, backups, a number to call when a stoppage turns out to be software-shaped, and changes as the process changes.

Terms

How we run a project

You own the code

The source code, repository and deployment scripts are handed over. No licence trap, no vendor who has to be paid to read your own data.

Your data stays on your network

On-premises by default, including the AI parts. We use the cloud only when you ask for it, and then in the EU, under a DPA.

Fixed scope before fixed price

We quote a written scope, not an hourly guess. If the scope changes, that is a decision you make with a number in front of you.

Engineers, not account managers

The person who walks your process is the person who writes the code and answers the phone at 22:00.

AI only where it pays

We do not put a language model on a problem a database query solves. Where AI does fit, such as reading documents, searching procedures or flagging the odd case, we tell you up front what it will get wrong and what a human still checks.

Sometimes the answer is not software

If the fix is a change to the process itself, that is the recommendation you get, even though it is the smaller invoice.

Working with

  • PostgreSQL
  • Node.js
  • .NET
  • React Native
  • Docker
  • REST and EDI
  • Claude API
  • Local LLM
  • OPC UA
  • Modbus
  • TimescaleDB
  • On-premises

Contact

Tell us where the gap is.

Tell us about the process, the systems around it, and the thing you cannot see today. If you are not sure yet what needs fixing, that is a normal place to start. You get an answer from an engineer, usually within a working day.

Write to us
Email
hello@appforgap.com
Based in
Czechia, working across the EU
Languages
Czech, English