db3 Labs · Applied AI for established companies

Pay for outcomes, not hours.

db3 Labs is the applied side of db3.ai, led by me. We build working AI systems inside established companies. We agree the outcomes before we start, and you pay for each one when it is delivered.

See the engagements Start a conversation ↗

How we work

Priced per agreed outcome
Led by Steve O’Brien throughout
Strategy and hands-on engineering

Why outcomes

You take on enough risk already.

Most companies do not need another AI workshop. They need to decide where AI creates real value, build the right thing and make it dependable enough to become part of the business.

Paying by the day rewards activity. Paying per outcome means we both care about the same thing: a system that works in your business and does what we said it would.

How the pricing works

Agreed up front. Paid on delivery.

Agree the outcomes

Before any work starts we agree what each outcome is, how we will measure it, how we will check it together and what it costs.

Build it into real work

We build the system, test it with the people who will use it and put it into your existing operation.

Check it, then pay

When an outcome meets the measure we agreed, you pay for it. If it does not, you don’t.

An outcome is specific and checkable: a system in live use that meets a measure we both signed off before work began. We agree outcomes the work can control. “Revenue up 20%” depends on too much else, while “qualified leads routed and answered within five minutes” does not.

Choose the right depth

Two ways to start building.

Engagement 01 · Six weeks

3 outcomes

Implement 3: Revenue

Select and implement three tightly scoped AI plays designed to create, convert or retain revenue.

What you get

Commercial opportunity and customer-journey mapping
Three working AI implementations in your existing operation
Evaluation, measurement and sensible human safeguards
Technical handover, team walkthrough and 30 days of support

Pricing

Per outcome

Three outcomes, each priced before work starts and paid when delivered.

Discuss Implement 3

Engagement 02 · 90 days

4 outcomes

The AI-Native Company

Install the first working version of an AI-native operating model—across customers, operations, knowledge and product.

What you get

AI opportunity map, priorities and 90-day transformation plan
A portfolio of working systems around the highest-value priorities
Shared knowledge, evaluation and governance foundations
Team enablement and a board-ready roadmap for the next 12 months

Pricing

Per outcome

One outcome per stage, each priced before work starts and paid when delivered.

Discuss the 90 days

Both engagements assume one company, an available decision-maker and reasonable access to the people, systems and data involved. Third-party software, model usage and substantial new platform infrastructure are priced separately where required.

Inside Implement 3

A deliberate path from opportunity to shipped value.

Find the commercial leverage

Map your customer journey, evidence and constraints. Select the three opportunities most likely to create or protect revenue, and agree each one as an outcome.

Build the working systems

Prototype quickly, test with the people who will use them, then integrate the strongest version into your existing operation.

Prove and transfer

Measure each outcome against what we agreed, document the decisions and leave your team able to operate and improve what has been built.

Inside the 90 days

Change the operating system, not just the tools.

Ninety days will not finish the AI transformation. It will create the priorities, working systems, internal capability and decision-making rhythm needed to make that transformation real. Each stage ends in an outcome you can check.

Days 1—15

Diagnose

Outcome: agreed commercial goals, a workflow baseline, the data reality and a prioritised map of where AI makes a material difference.

Days 16—45

Prove

Outcome: the first useful systems in real work, with human review, evaluation and clear measures of value.

Days 46—75

Expand

Outcome: what works extended into connected workflows, shared knowledge and repeatable ways of building with AI.

Days 76—90

Embed

Outcome: ownership transferred, practical governance in place and the next 12 months of change agreed.

Principles

Practical by design.

Working software, not theatre

The output is something your team and customers can use—not a strategy deck full of hypothetical use cases.

Commercial value first

We start with the customer and the economics, then decide where models, agents or automation genuinely help.

Built with your team

Decisions, safeguards and technical knowledge stay inside the business, so progress continues after the engagement.

Good fit

Built for companies ready to act.

This is likely right if…

You have customers, operating data and a meaningful commercial goal.
Leadership can make decisions and give the work visible backing.
You want senior thinking paired with hands-on implementation.

It is probably not right if…

You only need a generic AI training day or a list of ideas.
No one can own the change once the engagement ends.
The brief is to remove people without redesigning the work responsibly.

Rather build it yourselves?

Learn to implement AI with your own team.

I implement one AI process for you, then one with you, and your team gains the skills to keep going.

The course →

Start with the business

Bring the goal. We’ll agree the outcomes.

The first conversation is a short fit check: what you want to change, why now and whether one of these engagements is the right way to do it.

Talk to Steve