01
Context
What it knows
The documents, pages and records the system is allowed to read, and the boundary where that reading stops.
AI consulting
AI that can act.And knows when not to.
We design AI systems that research, remember, decide, create and operate inside real business workflows, with explicit boundaries around what they can touch and when a person stays in control.
New enquiry
“We are getting traffic but almost nobody books.”
System interpretation
Qualified service enquiry · brand strategy
Proposed action
Send the relevant case studyCreate the CRM recordSchedule a follow-up
Human approval required
The system proposes. A person releases the action.
The real problem
AI is easy to demo. Reliability is harder.
A clever answer is not the same thing as a dependable system.
The difficult questions begin after the demo.
- What does the system know?
- What does it remember?
- What can it touch?
- What actions require approval?
- How does it verify success?
- When does it stop?
System anatomy
A trustworthy agent is designed as a chain of constraints.
02
Memory
What persists
What is kept between conversations, for how long, and who can see or remove it.
03
Tools
What it can access
Each connected system is a decision: what it may read, what it may write and what stays out of reach entirely.
04
Permission
What it may do
Actions are separated into what runs without asking, what waits for a person, and what is never allowed.
05
Action
What it changes
Every write is named, logged and reversible where the underlying system allows it.
06
Verification
How success is checked
The system has to be able to tell the difference between having answered and having answered correctly.
07
Escalation
When a person takes over
The point where the system stops guessing and hands the decision to someone accountable.
- The exception, designed in
- Escalation is not a failure state. It is the part of the design that decides what the system is allowed to be wrong about.
Not a screenshot
Use it.
The interesting part of an agent is not the reply. It is the state around the reply: what it used, what it was allowed to do, and where a person had to agree.
User
We are getting traffic but almost nobody books.
System
Before recommending more acquisition, I would inspect the point where interest becomes commitment. Do people understand the offer, see enough proof and know what happens after they click?
User
Then tell me what is broken.
System
I can audit the public pages and show you where confidence breaks. I cannot contact anyone or change your CRM until a person releases that action.
Context used
Website and service model
Tool access
Site audit · read only
Allowed action
Analysis and drafts
Human approval
Required before any outreach
User
We need a month of content without sounding like everyone else.
System
I will work from the positioning, the customer objections and what you have already published, so the first draft argues a point rather than filling a calendar.
User
Who signs it off?
System
Nothing publishes itself. Drafts and schedules are prepared; a person approves every send.
Context used
Positioning, objections and published work
Tool access
Publishing queue · drafts only
Allowed action
Drafts and schedules
Human approval
Required before every send
User
Our support inbox is the same twelve questions.
System
Then the job is retrieval and routing, not conversation. Answer what the knowledge base genuinely covers, and escalate the rest with context attached.
User
What if it gets something wrong?
System
That is why it cites its source, and why anything financial, contractual or medical goes to a person immediately.
Context used
Knowledge base and service model
Tool access
Support inbox · read only
Allowed action
Answer and route
Human approval
Anything financial, contractual or medical
Each panel shows the state a person would see before releasing anything: what was read, what was reachable, what was allowed and who decides.
What we build
Start with a workflow. Not with “we need AI.”
Revenue operations
Lead arrivesEnrichClassifyDraftApprovalCRM
Knowledge systems
DocumentsRetrieveReasonCite the sourceAnswer
Customer-facing agents
Customer questionClassifySolveEscalate with context
Operational automation
RequestPlanTool actionVerifyReport
Filled and teal where the system acts.
Outlined and dashed where a person decides.
One chain per workflow, in the order it runs.
Our own systems
We use the same principles inside Pulse.
We sell operating systems we are willing to operate ourselves.
Audit
Reads the public site and reports what it finds, without being told.
Assistant
- Contextpublic site and service model
- Tool accessread only
- Memorynone beyond the session
- Escalationa person, on request
Growth engine
Turns the decisions into ongoing work someone actually owns.
AI visibility
Can machines understand what your company means?
A person reads
“Build for attention. Design for trust. Optimize for choice.”
A human weighs the sentence, the work around it and the reputation behind it.
A machine extracts
- OfferWhat is being sold, in one line
- EvidenceWhat the business can prove
- EntityWhich company, where, since when
- AudienceWho it is for, and who it is not
- Next actionWhat a reader is expected to do
Being indexed is not enough. AI assistants increasingly interpret a business through its published language, evidence and structure. We examine whether a site is legible to people, search systems and AI without flattening the distinctions that make the business worth choosing.
Start here
Tell us what should happen. We will work out whether AI belongs there.
We map the task, the information it requires, the actions involved and where human judgment should remain. If AI is the wrong instrument, we say so.
Book an alignment call
Two lines and the studio comes back with times.