“How to redesign a company so AI actually does the work?”
See what we built
- of processes: what can be automated and where AI makes sense
- Map
- CRM, accounting, projects and Google or Microsoft workspace, all connected
- Toolset
- built around the services the company sells
- Custom AI
The starting point
The agency wanted to be an AI-first company. Not with an extra chat window, but with AI doing real work under human supervision. It had plenty of tools, but they weren’t connected to each other.
Goal: map the processes, set up the architecture, choose and connect the right tools, and build a knowledge base and infrastructure on top of them.
Why redesign the company
Three ways a company starts with AI.
| Buy tools | A consultant’s presentation | Redesign and build | |
|---|---|---|---|
| Result | licenses | a document | running systems |
| Connection | none | a proposal | connected tools and knowledge base |
| Team | learns on its own | hears theory | workshops on its own processes |
| Decisions | based on ads | based on slides | based on tests with its own data |
What we built
Company map
How the company works today, who does what, what can be automated and where AI makes sense.
AI-native architecture
What the company will look like when AI does real work in it, including the economics.
The right toolset
CRM, accounting, invoicing, projects and Google or Microsoft workspace. We keep what people already know, so they don’t have to learn a million new things.
Connections and custom AI
We connect the tools so they share data. We set up AI around the services the company sells.
Knowledge base and permissions
Company knowledge that both people and AI draw on, with user permissions set up.
Timeline
May 2026today
- Company map
- Architecture
- Choosing tools
- Connections
- Knowledge base
“First we drew the whole company. Only then did we decide where AI would work.”
Management and support
- Leadershiprules, limits
- People and partnerscapacity
- Legal and compliancecontracts, data, rights
- AI and technologyplatform, agentsAI
- Internal operationsday-to-day running
- Brand and PRreputation
provide rules, capacity and technology to the whole flow
Main flow
- Marketingbrings in leads
- Leads and CRMcapture, qualificationAI
- Opportunity and proposalneed, solution, price
- Signed contracttrigger event
- Project kickoffmaterials, access
- Deliverycarrying out the serviceAI
- Handover and resultsapproval, closing
- Relationship and next demandfeedback
Finance and control
- Contract signalprice, milestones
- Financial planbilling plan
- Work and changescosts, scope changes
- Control and closingmargin, reporting
Knowledge and data
- Company knowledgemethodology, rulesAI
- Client knowledgeclient context
- Sales and proposalscope, price
- Project knowledgeoutputs, results
- Approved insightsno client data
approval into the knowledge base
where we brought in AI and agents
Under the hood
For those who want to know how it’s built.
- Architecture
- the company map and its AI-native form
- Tools
- CRM (e.g. Pipedrive, Raynet), accounting (Pohoda, Money, Abra), invoicing (SuperFaktúra, iDoklad), projects (ClickUp, Trello, Notion), chosen to fit the company
- Connections
- data sharing between tools, user permissions
- Knowledge
- knowledge base infrastructure
- Training
- workshops on the team’s own processes
“We didn’t buy another tool. First we understood our own company, and only then did we start building.”
“Want something similar?”
Let’s start with your problem.
Next story · Instant quotes
How to send a customer an exact quote within 10 minutes, while they still care?”
“Have a question? Just ask.”
Want to know more about our services? Interested in concrete use cases and references, whether we’ve already solved a problem like yours, or what it might all cost? Talk it through with our AI agent. It knows a lot more than you’d think.