The DAP method
Understanding before technology
Before automating anything, we understand how your organization works. Then we define the outcome, put AI and data to work, and verify it's delivered.
Four steps, and what you get from each
- 1
Understand
How your organization actually works: processes, decisions, data and people.
Deliverable: Operating map and diagnostic - 2
Define
The expected outcome, its success metrics and its limits on risk, privacy and budget.
Deliverable: Outcome sheet with indicators - 3
Orchestrate
The AI agents, geospatial data and integrations that deliver that outcome, with people in the loop.
Deliverable: Solution running in a real workflow - 4
Verify
Quality, traceability and performance, continuously, with reports ready for donors and regulators.
Deliverable: Verification report
Verification feeds the next cycle: every measured outcome sharpens the next definition.
Ways to work together
Start small. Scale on evidence.
Durations are indicative and adapted to each organization. Every proposal is scoped and priced individually.
| Model | Indicative length | What you get | Pricing |
|---|---|---|---|
| Diagnostic (AI Readiness) | 2 to 4 weeks | Process map, workflows ready for agents, economic baseline and roadmap | Fixed fee |
| Sprint or pilot | 6 to 10 weeks | A solution running in a real workflow, with agreed metrics | Fixed fee by milestones |
| Managed operations | Ongoing | Operation, improvement and support of the solution | Subscription or pay-for-outcome |
| Audit and verification | Per cycle | Evaluation of agents or systems (yours or third-party), MRV and traceability | Fee per cycle |
Principles
How we always work
- People in the loopThere are always people overseeing the AI.
- Your data, your controlClient data stays under the client's control.
- TransparencyWe're open about the models we use and what they cost.
- Metrics from day oneSuccess indicators are agreed before we start.
FAQ
What people usually ask
Do you work with tenders and development funding?
Yes. We have experience with projects funded by development agencies and banks, and we adapt deliverables and reports to their requirements.
Who owns the data and the code?
Client data belongs to the client. Code ownership is set in the contract, and we use open components when the funder requires it.
Which AI models do you use?
We choose the model case by case (cost, language, privacy and data control) and document that choice. We don't depend on a single provider.
What happens when the project ends?
We hand over documentation and training, and offer two paths: managed operations by DAP or full transfer to your team.
Do you work in Spanish and English?
Yes. We work in Spanish and English across Central America, the Caribbean and with international partners.
Next step
What outcome do you need to prove in the next 12 months?
Tell us about your challenge. In a first conversation we'll see whether a diagnostic makes sense for your organization.
Book a diagnostic →