Questions? No problem.

This can be done via Gitlab or by email. We review each report and classify it as a bug/incident, RFC or user question. Depending on the type and urgency, we apply corresponding response times.

No, an SLA is optional. However, an SLA does offer additional benefits such as guaranteed response times, a fixed point of contact and prioritisation of calls. Without an SLA, you can still use support.

Impact sessions are strategic sessions in which we look together at how your digital environment can be further optimised. We link insights to concrete tickets and build an ongoing roadmap together.

We work on a co-creation basis. This means that we work intensively with your team to come up with smart, scalable solutions together. We apply our C-U-B-E model: Create, Invent, Build and Evolve.

Our service package consists of a combination of maintenance, management, hosting, support (SLA) and development (strippenkaart). In this way, we ensure that your digital landscape remains secure, up-to-date and future-proof.

Sprint 0 is our inventory phase. In a few weeks, we completely map out the scope of a project: from goals and user requirements to technical requirements. In this way, we ensure a flying start and a solid foundation for what follows.

With the strip card you reserve a fixed number of hours per month for smaller RFCs (requests for change) and user questions. This gives you control over budget and capacity, and ensures that requests can be dealt with quickly.

With an additional SLA option, we offer out-of-hours reachability for high-prio incidents. This coverage runs from Monday to Sunday between 08:00 and 22:00. All hosting services are monitored 24/7.

Certainly, Cube is growing and our people are growing with us. So you can grow into a position with more responsibilities or a position with a different content. There are many different possibilities, depending on your knowledge and skills, hard- and softskills and your drive.

At Cube, the focus is on customer projects, but developing our own products is no stranger to us and is on the roadmap for the coming years. The scope of assignments varies. We have multi-year intensive collaborations with clients whose entire digital landscape we manage, and we work on projects with a clear head and tail. Every project is different, and every client has different requirements and starting points. Check the  cases on our website to get a good idea of this, At Cube, we focus on customization and innovation. That keeps our work varied. Do you have a preference for one or the other, or does a varied mix suit you well - just let us know! We try to take your personal wishes and preferences into account as much as possible.

You will work in the design team, but you will also collaborate a lot with external stakeholders to fully understand their needs and requirements. You will also work closely with colleagues from other departments, such as development and business, to create the right solutions and ensure the successful execution of projects.

The ideal size of a team is around nine people. Composition: two project managers and seven developers. When starting a new colleague, extensive consideration is given to knowledge, skills and the needs of the teams and the applicant to determine in which team the new colleague will do best. Each team has a Team Lead - this is one of the project managers - and a Tech Lead, this is one of the experienced developers. Each developer at Cube has their own specialism (e.g. front-end, back-end, app, Vue.js, Flutter, React Native or a combination thereof). In addition, several developers have Devops responsibilities.

Developing scalable, user-friendly applications and software solutions. You work together with project managers, UX/UI designers, developers, devops engineers and strategists to create the best solution for the client. Work efficiently, ask questions and learn from colleagues.

Your working day starts between 07.30 and 09.00. We have a joint break between 12.00 and 13.00. Your working day ends between 16.00 and 17.30. At Cube, we work 7.5 hours daily while you get paid for 8. This is in the second half-hour break (12.30 - 13.00). You get this extra time from Cube so you can take a break from work after lunch. Go for a walk with colleagues, play a game of football, darts or table tennis. If you want to go out on your own, no problem. Do what recharges you.

Cube employs both full-time and part-time employees. Depending on the role, we also have part-time options.

Learning on the job, our Hackathons, knowledge development hours, personal training and development programmes; all these contribute to the development of teams and individuals. Do you have a wish or an idea regarding development, training or workshop? Let us know. Then we will see what we can do. Basically, there is a training budget of EUR 750 per year for each team member.

Certification is possible! If you wish to obtain a professional certificate, we will be happy to help you. In recent times, for instance, colleagues have obtained certifications for Laravel, Vue.js and the implementation of the ISO 27001 standard.

Our office is in the heart of Oldenzaal, at the foot of the Plechelmus basilica and opposite the town hall. In front of the door is a large car park where you can park your car. If you come to meet us, we will switch on the parking meter for you. If you start working at Cube, we will arrange a parking permit for you. The current office has 45 fully equipped workstations. These are all equipped with a modern standing desk, ergonomic office chair and one or two 28" monitors.

Scalability is essential, because you want your API to grow with your business/product. We use techniques like load balancing, where traffic is distributed across multiple servers, and caching with tools like Redis, to make frequently requested data available quickly.

Security comes first. We implement the latest security measures, such as OAuth 2.0 for authentication, API keys for access management and SSL/TLS encryption for secure communication.

We provide customised integrations with all your external systems, such as your ERP, CRM, payment systems and marketing tools. This way, your systems exchange data with each other efficiently and securely.

The cost of creating a web application depends on the complexity, features and integrations required. At Cube, we start by identifying your needs and goals so that we can provide a solution that fits within your budget. Whether it's a simple tool or a complex platform, we ensure price transparency and an efficient development process.

Application development is the process of designing, building and maintaining software applications for mobile, web or desktop. This includes everything from UX design and architecture to programming, testing, deployment and support.

At Cube, we specialise in customised applications using modern technologies such as Flutter,  React Native,Vue.js, Livewire, Laravel. Whether it's a mobile app, web platform or fully integrated system, we deliver secure, scalable solutions that align with your business goals.

With MCP, the integration layer runs on your own infrastructure. You control which data is visible per tool and who has access. MCP supports authentication via OAuth or API keys, rate limiting, and logging. Data does not leave your environment unless you explicitly allow it.

A DMP ensures data is centralised and structured. MCP is the layer that gives AI models access to that data. They complement each other.

Yes, when implemented correctly. MCP runs on your own infrastructure. Cube always implements with permission management, authentication, and audit logging in line with ISO 27001.

The investment depends on the number of systems, the complexity of your data landscape, and whether a central data layer is needed first. A first MCP server for one system is typically operational within two to four weeks. Contact us for a tailored estimate.

An API is static and pre-programmed for a specific data flow. MCP is dynamic: the AI model discovers which tools are available and decides when to use them. In most organisations, both work alongside each other.

MCP focuses on server-side integrations with internal systems. WebMCP is a proposed browser standard from Google and Microsoft that allows websites to declare their capabilities as tools for AI agents.

MCP is an open standard that enables AI models to communicate in a controlled way with external systems such as CRM, ERP, or internal databases. Originally developed by Anthropic, it is now governed by the Linux Foundation.

Claude (Anthropic), models from OpenAI, Google, and Microsoft all support the protocol. A well-built MCP server works with any model that implements the standard.

Any system with an API can be connected, and some systems already have their own MCP server. Think CRM systems like HubSpot or Salesforce, ERP packages, planning tools, customer portals, and ticketing tools. What the MCP server can do depends on what that system's API allows.

Anthropic (the company behind Claude) introduced MCP in November 2024. In December 2025, governance was transferred to the Agentic AI Foundation under the Linux Foundation.

Most modern ERP integrations operate via an API. In more complex environments, a middleware layer is used to manage, translate, and monitor data traffic between multiple systems simultaneously.

At Cube, a functional initial version is typically launched within four to nine months. This is faster than the market is accustomed to, because we reuse components from previous ERP and portal projects (authorization, audit logging, integrations, reporting) and utilize AI-assisted development. After going live, further development continues as usual.

That depends heavily on the scope and complexity. You pay more upfront than with a standard package, but you have no license fees and no obligation to keep up with a vendor’s updates. Over the entire lifecycle, you often end up paying a comparable amount or less than with a standard package that requires a lot of customization.

In the Netherlands, these include AFAS, Exact, Microsoft Dynamics 365, SAP S/4HANA, and Unit4. The best choice depends on your industry, the size of your organization, and the complexity of your processes.

There are three options. Option 1: Keep a standard package and connect it to your online store, CRM, or WMS using custom integrations. Option 2: Have a fully custom ERP system developed around your processes. Option 3: Replace the ERP as the central system with an accounting package plus custom software that handles the business logic. Which option is best depends on process fit, organization size, desired lifespan, and data architecture.

ERP stands for Enterprise Resource Planning: software that centralizes core processes within an organization, such as finance, procurement, inventory management, and HR. Well-known systems in the Netherlands include AFAS, Exact, Microsoft Dynamics 365, SAP S/4HANA, and Unit4.

The cost of an ERP integration varies depending on the situation. It depends on factors such as the number of systems that need to be connected, the quality of existing APIs, and the complexity of the business logic. Because no two integrations are the same, both the cost and the timeline can vary significantly. During an introductory meeting, an initial estimate can be made based on your specific situation and requirements. This quickly provides clarity on what is needed and what to expect.

A custom ERP system is an Enterprise Resource Planning system that is built entirely around your organization’s processes, rather than a standard off-the-shelf package that you have to adapt to your processes. It’s worth it when your core processes are so specific that no off-the-shelf package can cover them without extensive customization.

An ERP integration is a technical connection between your ERP system, such as AFAS, Exact, SAP, or Microsoft Dynamics, and another application like an e-commerce platform, WMS, or CRM. Data is exchanged automatically, eliminating the need for manual input and making processes more efficient. The cost and implementation time of an ERP integration vary per situation. This depends on factors such as the complexity of your systems, the amount of data, and the desired functionalities. An initial estimate can be made during an introductory meeting, providing clarity on what is needed for a successful implementation.

Custom solutions are the way to go when three conditions are met: your core processes are highly specialized, your organization is large enough (50 or more employees) to support ongoing annual development, and the system will last seven years or longer. If fewer than two of these three apply to you, or if 80 percent or more of your processes overlap with a standard package, then that package is almost always the smarter choice.

It starts with mapping out the current workflow and identifying manual steps. Next, a suitable digital solution is built or configured, tested with real users, and rolled out in phases.

That depends on the complexity. A simple digital workflow is different from bespoke systems with multiple integrations, so the costs depend on the scope and the number of integrations.

Digital processes are workflows in which manual steps have been replaced by software. Examples include automated order processing, digital approval workflows, and automated reporting.

Core processes deliver direct value to clients, support processes keep the organization running, and management processes provide direction and ensure quality.

Digitization is the process of converting manual or analog processes into digital workflows, so that they can be carried out more quickly, with fewer errors, and more effectively using software.

Yes. With customization, you use the same catalog but with separate pricing rules and checkout flows for each segment. B2B customers see their own environment; consumers see a standard shopping experience.

Yes. We start by conducting a quality assessment of your current platform to evaluate its technical condition and identify migration risks, after which we set up a replatforming process that minimizes disruption to your business.

Yes. Cube integrates with AFAS, Exact, SAP, and other systems via API or middleware. The integration synchronizes product, customer, and order data bidirectionally and in real time.

Yes. Any online store that processes EU personal data is subject to the GDPR. At Cube, security and privacy are standard components of the development process.

Via API integration or middleware. Cube integrates with AFAS, Exact, SAP, and Microsoft Dynamics through a middleware layer that keeps the online store and ERP systems independent of one another and ensures their future-proofing.

Through a targeted onboarding campaign, a first-order experience that’s as simple as possible, and active support from your inside sales team during the transition to the new platform.

Off-the-shelf platforms work well for standard e-commerce. However, if you need complex processes, deep integrations, or custom workflows, a custom solution is better. Cube always provides advice based on your specific situation—even if an off-the-shelf platform is the better choice.

Custom pricing, volume discounts, user roles per account, a quote module, order history, purchase on credit, and ERP integration are the core features. Additionally, configurators, EDI integrations, and multi-warehouse support are available.

A B2B e-commerce platform supports business buyers with customer-specific pricing, volume discounts, multiple user roles, quote workflows, and ERP integration—features that a consumer platform does not offer.

OCI allows you to place orders directly from your online store through an external procurement system such as SAP. Products are selected in the online store, and the order line is automatically sent back to the procurement system. This is often a strict requirement for corporations and government agencies.

A standard platform is quick to set up but offers limited customization. A custom-built online store is designed around your processes, provides deep system integrations, and avoids vendor lock-in. In the long run, the total costs are often lower

The tipping point usually occurs between the second and fourth year. License fees, transaction costs, and plugin maintenance all factor in—for growing organizations, the TCO comparison often favors custom solutions.

For standard scenarios, Magento B2B or Shopify Plus are good options. For complex pricing logic, deep integrations, and custom order flows, a custom solution is the better choice. Cube provides recommendations based on your specific processes.

Yes. We handle redesign and replatforming projects for existing corporate websites. We always start by assessing whether a visual redesign is sufficient or if a new technical foundation is needed.

We integrate with popular ATS systems so that job postings are automatically synchronized and applications are routed directly to the right place.

Yes. We regularly conduct UX analyses and redesign projects for existing websites. Sometimes, a targeted adjustment to the conversion strategy is all it takes to generate significantly more leads.

An employer branding platform requires ongoing maintenance. Cube provides maintenance and management services to ensure the site remains technically sound and allows you to easily add content yourself.

An average corporate website project takes between twelve and twenty weeks, depending on the scope. We work in iterations so that you can see interim results throughout the process.

The cost depends on the complexity and duration of the project. A simple campaign website is faster and less expensive than a comprehensive microsite with interactive elements. We always provide a customized proposal following an initial consultation.

The investment depends on the scope of the project. A basic employer branding platform starts at a different price than a fully customized recruitment platform with an ATS integration. We assess your needs and provide a fair, customized proposal.

A corporate website is a custom solution. The cost depends on the complexity, integrations, and the number of target audiences the site needs to serve. After a thorough discussion, we’ll provide you with a fair, customized proposal.

A focused campaign site can sometimes go live within two to four weeks, depending on the scope and the delivery of content. We work in short sprints and keep you informed throughout the process.

For specific campaigns or target audiences, a dedicated landing page almost always performs better than the homepage. We’ll help you determine when a separate page makes sense and when it’s better to optimize the homepage.

A strong career site features a clear presentation of the company culture, a well-organized list of job openings, and an accessible application process. Authenticity is key: real employee stories are more effective than marketing jargon.

A lead generation website is designed to convert visitors into potential clients through forms, downloads, or direct contact options. The difference from an informational website lies in the strategy behind the design: every decision is made with conversion in mind.

The terms are often used interchangeably. Both are standalone websites with their own URL and purpose. The difference sometimes lies in their size: a microsite typically has more pages.

A "Careers" page is part of the main website. An employer branding website is a standalone platform dedicated entirely to talent acquisition, with its own structure and optimized for candidate searches.

A corporate website serves multiple stakeholders at once: clients, partners, the press, and job candidates. The difference lies in the breadth of the messaging and the higher standards for consistency and structure.

A standard website provides information. A lead-generation website persuades and drives action. The difference lies in the objective behind every element of the page, which is based on behavioral analysis and data rather than aesthetics alone.

We always discuss this in advance. A campaign site can be taken offline, converted into an evergreen page, or archived. We work with you to consider the long-term value of the investment.

That depends on your organization and editorial workflow. We work with several modern CMS solutions and always recommend the one that best suits your situation.

Yes. We deliver campaign websites equipped with a CMS that allows you to update text and images yourself, without having to rely on a developer.

Yes. Cube develops the software or provides guidance to your own team as they learn to work with AI tools. The GitLab Duo Agent Platform brings together tools, agents, and development in a single environment, with the oversight you need. This allows you to maintain control in-house while gaining experience with this way of working.

Yes. We integrate agents and copilots into your existing systems via MCP, so your business logic remains intact. There’s no need to overhaul your architecture: the agent is granted targeted, controlled access to exactly the data and actions required for its task.

Yes. We set up role-based access, record an audit trail for each AI action, and build in human checkpoints for irreversible actions. You determine in advance what the agent is allowed to do on its own and where human approval is required.

No. Cloud models are often the practical choice, but self-hosted solutions may be required if privacy requirements or governance mandates them. It’s a deliberate decision made on a project-by-project basis, not an automatic choice. Cube builds both.

No. The GitLab Duo Agent Platform is at the heart of our development environment, but we aren't tied to a single coding agent. We choose what works best for each situation and adapt when something better comes along, without vendor lock-in.

Only as a conscious choice, not as the default. The cloud tools run in enterprise versions with IP protection, and it is determined in advance what data may be sent to external AI. If requirements are strict, self-hosted is an option.

With role-based access: an AI instance sees only the context associated with its role. Cube uses separate contexts to prevent data from leaking between roles, clients, or projects. MCP serves as the governance foundation for this.

That depends on the process and the systems. An initial, well-defined automation can often be up and running within a few weeks; more complex workflows with multiple integrations take longer. Cube takes a step-by-step approach, so you see results quickly and can scale up.

The costs depend on the number of processes, the integrations, and the complexity of the decision-making. Cube starts with a single, well-defined process that can be automated, so that the return on investment is measurable. During a no-obligation consultation, we’ll work together to identify where the benefits lie.

Costs that skyrocket without cost caps, prompt injection in public interfaces, hallucination of tool calls, and data breaches via prompts. Mitigation: rate limiting, cost caps, strict validation of AI actions, and logging from day one.

Model independence means that you consciously choose the appropriate model for each use case, rather than using a single model as the default. Different models have different strengths. This prevents vendor lock-in and keeps your system flexible when models are updated.

RAG (retrieval-augmented generation) allows a language model to generate responses based on your own documents rather than just its training data. You need it when responses must align with internal knowledge—for example, a knowledge assistant based on your own documentation. It limits made-up answers and keeps information up to date.

AI safety refers to how you safely build and manage AI in practice: guardrails, evaluation, and monitoring. The EU AI Act sets out the legal requirements. The two support each other: good safety practices make compliance achievable, but the legal assessment remains the responsibility of a lawyer.

A copilot supports a person while they work and responds to direct input. An agent independently carries out a task, in multiple steps and across systems, within established parameters. The distinction lies in autonomy and human oversight, not in the technology.

RPA mimics human actions within existing software; it clicks like a human. Cube deliberately does not do that. We connect systems directly via API and build AI-driven decision-making into critical points. That’s more robust than a bot that relies on an unchanging screen.

Processes that are repetitive and rule-driven, involve digital input and output, have sufficient volume to justify the investment, and contain variations that AI can assess. Examples include request processing, booking flows, invoice recognition, and data transfer between systems.

No. AI is taking over an increasing share of the hands-on work, but decisions regarding architecture, quality, and security remain the responsibility of experienced engineers. As a result, the developer’s role is shifting from writing code to providing direction. As AI becomes more capable, that guiding role becomes more important, not less.

Yes. Cube connects your systems directly via its API, at the level where the data converges, and builds the automation into its own middleware. This way, the automation runs within your existing environment and remains robust, even if a system changes.