Custom Software Development with AI
In custom software development, datadice builds the business application, portal or ERP extension that your standard software does not cover and connects it to your systems. AI writes part of the code, experienced developers review every change and automated tests safeguard it.
For SAP customers time is pressing: mainstream maintenance for SAP ECC ends on 31 December 2027, and every custom development inside the ERP makes the migration bigger.
Goal:Develop custom applications
Last reviewed: · reviewed by Thomas Knorr, Managing Director datadice
Example
Funding:Small companies in Bavaria receive a grant of up to 50% from the Digitalbonus for the introduction. We prepare the application with you free of charge.
Go to the DigitalbonusTypical starting point
You want to introduce a business application for service, a customer portal or a tool for your field staff. But building it ties up capacity that your day-to-day business needs. At the same time, the new application and your existing systems have to complement each other reliably.
Our answer
We combine AI support with the experience of human developers. Your team stays in its day-to-day business. The new application works reliably with your systems.
We are convinced: companies must turn their business ideas into productive applications sooner. The time saved only counts if the software performs its tasks. It must fit into existing processes and work reliably day to day.
For your project management
Then datadice works with you to define the first version that is to go live. After that, we take over development. Your team provides requirements and acceptance.
The size of the first step matters most. That is why we ask which task the first version has to perform, which systems it connects to and how you will accept it.
For a field service application, for example, that means: record service reports and spare parts on site, read orders and items from the ERP, show customer data only within the order context. Everything else goes into version 2. Your project management receives a plan with deliverables, acceptance cases and the contributions of your team.
Architecture
We define architecture, access rights and data protection at the outset, so security and data protection are part of your application from the start.
Step 1 of 4
We clarify requirements, priorities and acceptance criteria.
Requirements and acceptance
Task: Record service reports on site
Priority: First version: report and spare parts
Acceptance: Report fully completed on site
Quality
AI-generated code sometimes contains errors, just like code written by people. At datadice, experienced developers review every change before we accept it. Automated tests run with every change.
We define architecture, access rights and data protection at the outset. Proven building blocks handle sign-in, rights and audit logging, so development can concentrate on what only your company needs.
AI speeds up the writing; responsibility for architecture and quality stays with named people.
Outcome
You receive a custom business application, a portal or an interface on top of your existing systems, with agreed interfaces, reviewed code, automated tests and documented architecture. We explain the way of working to your team and support rollout, enablement and operation to the agreed extent.
Building blocks
datadice develops web applications in which user journeys and system integration belong together, such as an ordering portal that reads prices and availability directly from the ERP.
We clarify requirements, priorities and acceptance criteria with you. Building on this, we design user journeys and interfaces and plan architecture, security, data protection and system integrations.
Company OS (a company's operating system: one sign-in, one rights model, one data foundation) describes our target vision for how your most important systems, such as ERP, CRM, project management and invoicing, work together on a shared data foundation. Your system landscape therefore grows step by step according to your priorities.
Company OS describes the direction. It is implemented step by step with the core modules, in the order of your priorities.
Customer information, quotes and orders are connected for everyday sales work, aligned with your quoting rules.
You receive a module for customer records, contacts, tasks, follow-ups, quotes and orders, with agreed adjustments and system integrations.
Connected operational information makes deviations visible and supports the steering of day-to-day business.
You receive overviews and analyses for agreed processes, with documented metrics and detail views down to the individual transaction. The functions are integrated into your application and given the required access rights.
SAP ECC end of maintenance 2027
Mainstream maintenance for SAP ECC ends on 31 December 2027. The effort of the migration is driven mainly by your Z programs and custom developments.
datadice builds business-critical custom developments as a separate application next to the ERP and connects them via an SAP interface. Your ERP remains the leading system and stays close to the standard (clean core), and the migration gets smaller.
Book an initial consultationArticle: SAP ECC end of maintenance 2027We record your custom developments and check their usage. Afterwards you know what can go, what the standard covers and what belongs next to the ERP.
Your ERP stays close to the standard. We build business-critical logic as a separate application and connect it via interfaces.
Less code in the ERP reduces the scope of your migration. This applies to S/4HANA and to any other ERP.
Comparison
The table helps you decide between standard software, custom development and a combination of both.
| Criterion | Standard software | Custom development | datadice |
|---|---|---|---|
| Starting point | The product's feature set | The project's specification | Requirements and acceptance criteria for the first productive use |
| Integration of existing systems | via the vendor's existing interfaces | depending on the contract | Connection to your existing systems, planned into the architecture |
| Your own business rules | within the configuration options | freely implementable | Functions for your business on proven building blocks |
| Quality assurance | by the vendor | as agreed in the project | Code review by experienced developers and automated tests |
| Operation and further development | with the vendor's releases | by separate agreement | Operation to the agreed extent, documented architecture |
| Enabling your team | Product training | as needed | Rollout and enablement to the agreed extent |
Starting point
Integration of existing systems
Your own business rules
Quality assurance
Operation and further development
Enabling your team
Application
Yes, if it passes acceptance. datadice derives the acceptance cases from your real transactions, including the special cases. A version only counts as finished once it passes these cases.
Your business department knows the exceptions that appear in no requirements list: the customer with special terms, the order with partial delivery, the item without master data. We turn them into automated tests that run with every change.
Your team tests every version with its own examples before it goes live. Where available, we show a demo system with sample data in the initial consultation.
AI App Development
A business application for service, field staff or customers
An application with agreed interfaces, reviewed code, automated tests and documented architecture.
See the exampleCustomers, Quotes and Orders
Quotes and orders according to your own rules
A module for customer records, follow-ups, quotes and orders. Your team sees every case in full: contact, quote and order.
See the exampleOperational Steering
Metrics directly in the application
Your teams see overviews directly in their applications. The metrics are documented. From the overview, your teams jump straight to the individual case.
See the exampleKnowledge
An objection we often hear
„Our own team can do this with an AI tool.“
Every custom development has to be checked. We move business-critical Z programs out of the ERP and build them as a separate application next to it. Your ERP remains the leading system. The migration gets smaller.
AI supports the implementation, experienced developers steer architecture and quality. Automated tests safeguard every change.
An application with agreed interfaces, reviewed code, automated tests and documented architecture, connected to your systems.
Company OS is our target vision for how your existing systems work together on a shared data foundation. We implement it step by step in custom software development projects.
Getting started
Which application moves your company forward next?
In the free initial consultation, we clarify the functions for the first productive use and prepare the requirements analysis. Scope and fixed price are therefore set on a clarified basis.
Starting point:Requirements analysis