Advisory, design and commissioned development — and our own products.
For teams bringing data and AI into their own services. We clarify the need, shape the solution and build the software — with clear responsibility from the first conversation to the delivered product.
Services
A fit for: Teams that need product leadership, design or technical expertise over time.
Typical shape Ongoing · 1 senior · agreed capacity
Support with product choices, technical direction or capacity in your team. We work within an agreed mandate and contribute where you need us most.
You get: A prioritised recommendation, visible risks and a plan for the next step. Ongoing support has agreed capacity and responsibilities.
A fit for: An idea, a difficult product decision or a workflow you want to improve.
Typical shape 4–6 weeks · 1–2 people
Make an idea or workflow tangible. We explore user needs, create prototypes and assess where AI could be useful.
You get: A prototype tried with relevant users, prioritised needs and a focused proposal for what to build first.
A fit for: Taking a defined need through to software people can use.
Typical shape 3–9 months · 2–4 people
We build software, integrations and AI features on commission. An AI product engagement starts with a concrete workflow. We agree on scope, quality and handover together.
You get: Testable releases, documentation and an agreed code handover. Operation and further development are scoped as part of the engagement.
How we work
AI supports analysis, prototyping, development and testing. People review the work and make the decisions. When AI is part of the product, we agree on data use, quality requirements and human control.
Understand the problem, the users and the intended outcome.
Try assumptions and prototypes before building further.
Build in focused increments with visible results.
Use experience and feedback to guide the next priority.
An AI engagement might involve an internal knowledge assistant, document processing or support within a specialist workflow. Where relevant, we apply lessons from Datakilde: source references, explicit access levels and answers that can be checked.
Agree sources, access and which information may be sent to external services. Use anonymised examples in the initial discussion.
Agree representative test cases, acceptance criteria and what happens when an answer is wrong or uncertain.
Establish who approves results, how errors are followed up and who owns operation and further development.
Our products
Our own products give us firsthand experience with complex data, demanding workflows and ongoing development.
Closed beta · access by invitation
An independent building-project product developed by Approach. Connects property information, 2D and 3D work and document drafts in one workflow.
Property information. Drawing drafts. Professional review.
Explore ProsjektKlar
Open demo · extended access by agreement
Norwegian address, company and area data through documented APIs and MCP. Inspect sources, freshness and limitations in each response. Restricted fields require a separate agreement.
Use the data for address enrichment, company insight and area analysis. The address demo here shows one example.
Explore DatakildeAPI and MCP documentationTRY DATAKILDE / PUBLIC DEMO
Search a Norwegian address and preview public data with sources. Explore the full catalogue on Datakilde.
About Approach
Approach leads direction and delivery. Any additional expertise or capacity is scoped for each engagement. Who contributes, and their responsibilities, are agreed clearly before work starts.
More about ApproachWe welcome conversations with specialists and small consultancies about collaboration on specific engagements.
Send a few lines about the problem, who it affects and what you want to achieve. You do not need a finished requirements specification.
post@approach.no