FAQ

Who is datatactics?

A well-rehearsed engineering team in Germany — senior leadership and junior talent, all business-case-sensitive, no offshoring, no account-management layer. Our systems run productively in aviation, in state air-transport logistics and in global industry. We build for sectors where standard software doesn’t carry — because complexity, compliance and 24/7 operations coincide.

Aren’t you too small for our critical systems?

Our customers include a top European airline group, a state air-transport unit and a joint venture at a major international hub. If size were the criterion, we’d never have won them.

What a well-rehearsed team with a small footprint can deliver comes from four levers:

• Open-source leverage — every person-day gets the leverage of 100+ person-days of community work from the foundations we build on.
• Platform reuse — our own stacks are redeployed in every project; new use cases are configuration, not code.
• Lightweight discipline — one container instead of a VM cluster, MVP in 6 months instead of 18; we actively reject over-dimensioned requirements.
• Long customer relationships — multi-year engagements with our existing customers; discovery overhead disappears.

Isn’t open source + engineering more expensive than standard tools?

You’re comparing the visible price. We’re comparing the total cost over 5 years — including exit.

License budgets in proprietary platforms are typically 30-40% of total cost. Underneath sit: initial integration (30-100% of license cost per year), customizing per requirement, forced upgrade cycles, per-user / per-GB / per-API scaling — and exit costs in the 6-figure range if you want out.

With datatactics, you pay for engineering, infrastructure (of your choice) and no licenses. Realistically the 5-year TCO sits 30-60% below proprietary solutions — with more control.

For very small deployments (< 5 users, < 100 GB), standard SaaS is often cheaper. Then we aren’t the right partner — and we’d rather say so upfront.

Isn’t open source less secure than enterprise software?

The foundations we build on run the critical infrastructure that powers banks, electricity supply and airlines. “Open source is insecure” is a legend from the 2000s.

The real difference:

In a zero-day situation, you can act yourself. With closed source you wait for the vendor release cycle.

We also rely on curated open source: foundation-backed projects with long-term support, updates and patches as part of our service contract. Operationally closer to enterprise software than to a hobby stack.

We already have a lakehouse or data warehouse. Do we need you then?

Yes — but for a different layer. Lakehouses and data warehouses are storage and compute. We are the application layer above: vault with update history, original-document storage, declarative business rules, real-time decision services, AI quality control with audit trail, cryptographically anchored audit chain.

What lakehouse vendors deliberately don’t build, because it isn’t their layer.

We can use your lakehouse as the storage backend if you’ve already invested there — the vault layer and our connectors run on top. We don’t compete with the storage layer, we deliver what moves money operationally and survives audits.

If you really only need classical reporting + BI, your existing lakehouse alone is the right choice. Then we aren’t the right partner.

Can’t we build this ourselves?

You can. The question is whether you want to.

Realistic build times without specialists:

• Productive Kubernetes platform with security, monitoring, backup: 12-18 months
• Integration runtime with hot-deploy and business rules: 9-12 months
• Data catalog with lineage and quality tests: 6-12 months
• Graph-based asset management from 15+ sources: 18-24 months

With us, these building blocks are ready, because we see them running productively at several customers in parallel. Plus: A 2-person team that built the platform internally is a single point of failure — when those people leave, the organization is left with undocumented tribal knowledge.

We have three cooperation models: Fully managed, co-build with later handover, or enablement (we train and audit, your team builds). The dichotomy “build OR outsource” is wrong.

If you want to position platform engineering as a core competence (hyperscaler mindset), build it yourselves. Then we’re at most useful in a review role.

How fast can you start?

• Phase 1 — Tactical Assessment: free, 30 minutes online, within the same week. An experienced engineering lead listens, asks back, gives an honest read — still in the call.
• Phase 2 — Architecture Sprint: fixed price €2,500, 3-4 engineering days (1 day on-site optional, then work, then closing presentation), completed within 2 weeks. Result: architecture document, roadmap, effort estimate for phase 3.
• Phase 3 — Engineering & Operations: three delivery models (greenfield / integration / hybrid). MVP in under 6 months cumulatively.

Do we get SLA, even if we operate the system ourselves?

Yes. What comes out of our hands is operable — and we offer SLA with support, classically when we operate, and also when you operate yourselves:

• We operate — 24/7 monitoring, response times per severity, incident response by our engineering.
• You operate — support SLA with third-level support, defined response times, architecture reviews, curated CVE and version recommendations, escalation path.

This is possible because we live DevOps: Every stack is built as an operable system, not as a demo for later handover. Logging, observability, runbooks, GitOps and backup patterns are part of the delivery. The operating manual is created together with your team and stays in your repo — not in a vendor knowledge base you can’t reach.

What happens if we’re not satisfied with you?

You can exit at any time.

• Our integration runtime is Apache 2.0-licensed and publicly available — freely usable by anyone.
• The datatactics Platform consists 100% of established open-source components — Kubernetes, identity, database, observability, CI/CD; all foundation-backed and without a platform license.
• Your configuration (GitOps, IaC, Helm charts) belongs to you — from day one of every project.
• Your data sits in standard formats — exportable at any time.
• Custom logic (rule sets, data models, custom code) is openly readable and transferable.

If you end the relationship, we provide the handover documentation to a successor or your internal team. That’s not a marketing gesture — it’s a structural property of our platform.

Do you actually not use any hyperscalers?

We do, where it makes sense. One of our global product-search APIs runs on hyperscaler infrastructure, with our architecture patterns on top. What we avoid: service-watering-can consumption — deeply embedding 30 hyperscaler-specific services that later become a migration obstacle.

Cloud-agnostic architecture is the default. Hyperscaler-specific services only where they bring a clear benefit that justifies the later migration effort.

What are you not the right partner for?

The four profiles where another partner is the better fit are on the home page. In short: mass self-service onboarding without engineering involvement, pure hyperscaler buildouts without a sovereignty need, implementation mandates with 100+ consultants, and standard suites as the main business — none of those are our turf. If you recognize yourself there, we’ll say so in the first call — and, where we can, point you to a partner who fits.