From a business task to a defined project, with a budget that covers advice, implementation and adoption.
Funding can make it easier to start an AI project. The investment still has to fit your business. We would not expand a project simply because more expenditure could qualify for a grant. What matters is which work will improve and what you want to keep running once the project ends.
Start with a specific task. You might want to prepare quotations faster, make knowledge in project files accessible or extract information from incoming documents. Once you have defined the first use, you can describe the work and its cost. That plan gives you a basis for choosing the right financing.
Understand where work gets stuck
“We want to use AI” gives you little to base an investment decision on. Describe where the process slows down: who is waiting for which information? What gets entered more than once? Which questions keep coming back? And how would the team recognise an improvement?
Consider a hypothetical trades business. To prepare a quotation, the office gathers information from an email, site-visit photographs and the supervisor’s notes. Missing information creates another round of questions. Here, the AI idea is to bring the available information into one case, highlight gaps and prepare a quotation draft using the business’s approved templates. The employee responsible still decides on the work and prices.
You now have something useful to discuss. You can show which part of the task should change. Conversations with your team, an implementation partner and the funding body become much more specific.
Choose a first use you can put into operation
For this business, the first assignment could cover one recurring type of enquiry. It would include an approved quotation template, the necessary data sources and a clear review step for the office. Other quotation types can follow once that initial workflow works reliably.
This boundary helps you estimate the cost. Using information already available in one system is different from connecting email, document storage and business software. Preparing documents, permissions, testing and introducing the workflow to the team also take time. They belong in the project scope.
Describe the first use on one page: today’s task, the intended result, the systems and data involved, who will own it in operation and the cases you will use for acceptance. That gives a supplier something concrete to quote for and your team something it can assess.
Budget for advice, implementation and training together
A useful budget explains how the solution will enter everyday use. Advice covers understanding the process and choosing a suitable approach. Implementation includes configuration or development, connections to existing systems and testing. During adoption, the team learns to use the workflow and recognise difficult cases.
Plan these activities together, even if you commission them separately. An application offers little value if the required documents are missing or nobody can judge its results. Equally, not every project needs custom software. First check what your existing tools can already do.
Show ongoing costs separately. Hosting, model usage, maintenance and support affect the investment beyond the initial project. Eligible expenditure and the full cost of ownership can differ.
A grant, a loan or your own funds?
A grant can reduce your contribution to an eligible project. Its suitability depends on the programme, your business and the specific work. The German state programmes discussed here are geographically limited. WTSH’s DKU programme in Schleswig-Holstein covers advice, implementation and related staff development. This makes a clear breakdown of the work particularly useful.
A subsidised loan is still financing you must repay. KfW’s ERP Digitalisation Loan also covers AI projects and is applied for through a financing partner. Certain projects can receive a supplementary grant. Assess the loan and the grant separately.
Your own funds may suit a small, well-defined starting point. Compare the application and coordination effort, the timing and the possible funding advantage with the project itself. Funding does not have to be the first step for every AI idea.
Keep cash flow in view. Under Brandenburg’s BIG-Digital programme, payment follows review of the evidence of expenditure. An approved grant therefore does not remove the need to plan how you will pay for implementation initially.
Establish the funding conditions before placing an order
Discuss the project plan early with the relevant funding body or financing partner. A specific proposal lets you establish which work may qualify, which documents are missing and when you may commission the work.
Timing matters. For DKU, WTSH counts a concluded supply or service contract as a project start; projects already started are excluded. For BIG-Digital, ILB permits a start at your own risk after the digital application receipt. That receipt is not a funding commitment. The rules differ between programmes.
Our German-language guides to AI funding in Schleswig-Holstein and AI funding in Brandenburg explain those state programmes in more detail. Your project plan gives you a basis for asking the right questions. These examples concern German funding programmes, with their respective location and eligibility requirements.
Bring a clear project to the first call
You do not have to choose an AI model or write a complete specification before a first call. Bring the task you want to improve, the systems involved and some anonymised examples from daily work. Tell us which budget and timing would make sense for the business.
As a 360° partner, we work with you to understand the workflow, choose a suitable technical approach and define a scope we can implement. Adoption and operation belong in that work alongside the application. Our AI process automation explains how we connect those activities.
A free first call can establish the next step: a focused assessment, a clearly costed first project or a decision to set an idea aside for now. You can use that basis to discuss funding with the relevant body. Discuss your AI project.
