Deliverable 01
A prioritised automation register.
Every candidate process scored, ranked, and one-page-costed. Yours to keep — the reference the next round of automation will use, whether we deliver that round or not.
S9 · Build
Automate the tedious parts of your operation — RPA where it fits, workflow orchestration where it doesn't, LLMs and agents where they earn their keep. Tool-neutral by design; measured by outcome.
Workflow automation is where AI and the boring engineering both have to be good. Clark leads the discovery — his methodology and the department demo library sit behind every recommendation. We lead the build. Every process shipped is a real production system, not a proof of concept, and every recommendation stays honest about where a human is still in the loop.
We’re deliberately vendor-neutral. UiPath for RPA where the platform is already there. Power Automate where the org lives in Microsoft. n8n or Temporal where the surface is engineering-owned. LLMs and agents where they earn their keep against a measurable bar — never as a first recommendation for a task a rule-based automation would handle for a tenth of the cost.
If you’re mid-assessment, the automations in your register will land here. If you already know the shape of what you want to ship, we can start with a single process and let the platform grow around it.
When to bring us in
How we approach it
Some processes want an agent. Most want a plain workflow. A few want the process redesigned before you automate anything. We start with an opportunity register, ranked on Value, Feasibility, Speed, and Strategic Alignment — the tool comes after the rank.
RPA (UiPath, Power Automate, n8n), workflow orchestration (Temporal, Airflow, Prefect), LLMs and agents for the decisions humans currently make. We recommend on the merits and integrate whatever your platform team has already committed to.
Tests. Observability. Cost caps. Runbooks. An escalation path for the twenty percent that will always need a human. Anything less is a demo that will surprise you.
Every automation carries a calibrated confidence per output. Above the threshold, it goes straight through. Below, it queues for the human who was doing the whole task before. Volume shifts to the machine at the pace the numbers earn.
Per-process throughput, error rate, cost per handled item, and the exception queue depth. On a dashboard the operations lead looks at anyway — not in a report we send monthly.
What you get
Deliverable 01
Every candidate process scored, ranked, and one-page-costed. Yours to keep — the reference the next round of automation will use, whether we deliver that round or not.
Deliverable 02
Ingestion, orchestration, guardrails, human-in-the-loop routing, observability, cost tracking, exception handling. Deployed, monitored, owned by your team on day one.
Deliverable 03
What to do when a supplier changes an invoice layout. What to do when the LLM returns a valid-looking answer nobody trusts. What to do at 03:00 when a queue is filling faster than it's draining.
Engagement models
SPRINT
Six to ten weeks. One high-value process shipped end-to-end with the platform pieces (observability, cost caps, runbook) around it. Named lead, weekly demo.
PROGRAMME
Two or three processes per quarter, sequenced by the register. Includes the platform investment the second and third automations will benefit from. Ideal after an assessment.
Related services
S0 · Advise
A structured discovery engagement — 8 stages, ranked opportunities, ROI-costed roadmap — delivered by our partner practice so we walk into your build with the shape already agreed.
S2 · Build
Multi-step LLM workflows with clean topology, tool surfaces, memory, evaluation, and safety — built to land in production, not stall in a demo.
S3 · Build
Chat, RAG, and the wider integration stack — measured, regression-tested, and cost-aware from the first prompt.
S8 · Build
Full production software delivery, usually with AI at the core — fixed team, named engagement lead, no staffing pyramid.
A 30-minute discovery call. We'll get to the shape of the work and whether we're the right fit.