The workflow, documented
Including the exceptions people handle without mentioning them, which is usually where the time goes.
Automate operations
Return the capacitymanual work is eating.
Agents and integrations that run repeatable operational work, with your team keeping visibility and approval wherever a decision carries consequence. Eight to twelve weeks, priced to the outcome after a free initial audit.
Every mark here links to the engagement it belongs to. See the work →
Agentic Workflow Automation builds software agents and integrations that run repeatable operational work, over eight to twelve weeks, priced to the outcome after a free initial audit. Approval gates keep a person in the loop wherever a decision carries consequence. The engagement suits businesses where staff re-key, reconcile or route the same work, and where the current time or cost of doing it can be measured.
Choose thiswhen
- People re-key, reconcile or route the same work
- Exceptions and approvals can be described
- Current time or cost can be measured
You get
A process map and baseline, agents and integrations with human approval gates, and a dashboard measuring the result.
Pricing
Priced Outcome
One agreed price for an agreed outcome, quoted after a free initial audit and approved before delivery begins.
What the estimate depends on- The scope of the outcome you want
- The stage the application is at
- The condition of the existing codebase
- The team the work requires
- How the system is used in production
- The operational complexity around it
Work out what it has to return first: the ROI calculator for operational work, how we measure for everything else.
Deliverables
What weactually deliver
The automation itself
Built into your real systems through their APIs, running on your infrastructure, with your rules as testable code.
An exception path that works
What the system will not decide goes to a person with the context attached, and the queue is monitored for depth and age.
Monitoring and a rollback path
Alerts, logs and a way to turn it off and go back to the previous process without losing work in flight.
Where it starts
The four places thisusually shows up
Operations
Work moves between people by hand, and every handoff can stall.
- The same record is typed into two systems
- A spreadsheet sits between two applications
- Volume grows, headcount is the only lever
Finance operations
Invoices and approvals depend on somebody remembering to check a folder.
- Month end is a week of copying
- Approvals live in an inbox rather than a system
- Nobody knows the cost per document
Customer support
The answers exist, in documents and past tickets, and finding them is the job.
- The same answer is rewritten daily
- Response time depends who picks it up
- The demo worked, production did not
Reporting and data
The number the business runs on is assembled by hand, twice, differently.
- A report is rebuilt weekly from exports
- Two dashboards disagree
- The data cannot be queried safely
Proof
Engagements thatlooked like this
CrystalCommerce How CrystalCommerce kept scarce inventory accurate across sales channels
No baseline was captured on this engagement.
COCOLI How COCOLI automated furniture supply across three European markets
No baseline was captured on this engagement.
Tunelark How Tunelark connected lesson sales, scheduling and instructor payouts
No baseline was captured on this engagement.
Is this the right engagement for you?
Describe what is actually going wrong. We come back with the outcome, the engagement that fits and the price basis - after a free initial audit.
Recognise this?
Where manual workusually shows up
The same fact is typed in twiceIt arrives in an email, gets typed into a spreadsheet, then typed again into the system of record.01
Work lives in an inboxThere is no queue, no owner and no age. Whether something is late depends on whether somebody remembered it.02
Approvals stallA request waits because the approver did not see it, and when it moves there is no record of who decided what.03
The handoff is a personTwo teams work in two systems and somebody in the middle keeps them in sync by hand, every day.04
Reporting is assembled, not queriedSomebody spends several hours a week assembling numbers management should already have.05
How we decide
Most of an automation project is not the automation.
The build is usually the shortest part. The parts that decide whether it works are the ones people skip.
- Mapping the process as it actually runs, not as it is described.
- Deciding what happens when the system is unsure, before it goes live.
- Running alongside the old process until the numbers say it is safe to switch.
- Owning it afterwards, when a vendor changes an API and something quietly stops working.
In practice
What fixing manual worklooks like
Email → extraction → validation → ERP → notification
Order processing
Orders arrive however customers send them and end up as clean records, with only the odd ones reaching a person.
Invoice → extraction → matching → accounting → exceptions
Invoice processing
Line items are read and matched against POs and receipts. Finance sees the mismatches, not everything.
Request → policy check → routing → audit trail
Internal approvals
Requests follow your policy automatically, reach the right person, and leave a record you can show an auditor.
System A ↔ System B ↔ System C
Data synchronization
A change in one system is the same change in the others, without a nightly export and someone to babysit it.
Go deeper
The decisions behindthis kind of work
- ArticleWhat is workflow automation? A practical definition for operations teamsA plain-language guide to workflow automation: what it is, what it is not, where AI belongs, and how to choose the first process.
- ArticleWe mapped an order-entry workflow. Here are the 11 automation opportunities we found.One process, one afternoon of watching people work, eleven places where the business was paying for the same information to be handled twice.
- ArticleWe mapped invoice and PO matching. Here are 8 automation opportunities.A mapped accounts-payable workflow showing what to extract, what to compare, which exceptions need people, and what should post automatically.
Questions
About manual work
Whatever survives contact with your environment: workflow platforms where they fit, custom services where they do not, and the APIs of the systems you already run. We do not sell you a platform licence and call it a solution.
No, and it is the normal case. Mapping the workflow is the first thing we do, and clients regularly tell us the map alone was worth the engagement because it was the first time the whole process was visible in one place.
Then we say so. Automating a broken process just makes it produce wrong results faster. Where the process itself is the problem, we tell you what needs to change before anything is worth building.
The outcome, the measure and the acceptance conditions are agreed in writing before development starts, which is what makes that question answerable rather than a matter of opinion. If the route to the outcome turns out to be wrong, changing it is our delivery problem and not a new invoice. If the outcome itself has to change, we document and agree that before continuing.
Eventually, in a controlled way, for the systems the workflow touches. Access is named, scoped to what the workflow needs, documented and revocable. We prefer a dedicated service account in your own identity provider so you can turn us off independently of anything else.
Not this one?
The otherengagements
Our roadmap needs a senior team behind it.AI-Native Engineering Pod3 months minimumPriced OutcomeThe AI demo has to work in production.AI Integration Sprint4-8 weeksPriced OutcomeThe prototype cannot survive real users.Vibe Code Rescue Sprint4-6 weeksPriced OutcomeThe old system is too risky to change.Legacy Modernization2+ quartersPriced OutcomeNobody owns the product after launch.Product CareOngoingPriced Outcome
Before you commit
What you wouldbe signing up for
One bounded outcome
A named result, a fixed window and a scope agreed in writing before delivery starts.
- Commitment
- 4 to 12 weeks
- Price basis
- Priced Outcome
- The problem has an edge you can describe
- You want the cost known first
- Someone inside owns the result
A team against a roadmap
Senior capacity inside your codebase, moving a roadmap week after week.
- Commitment
- 3 months minimum
- Price basis
- Priced Outcome
- The work is continuous, not bounded
- Recruiting would cost a quarter
- You want the same people throughout
Ownership after launch
Monitoring, patching, upgrades and a named engineer for live software.
- Commitment
- Ongoing, cancellable
- Price basis
- Priced Outcome
- It is live and business-critical
- Nobody owns it out of hours
- The risk needs an owner
One agreed price for an agreed outcome, quoted after a free initial audit and approved before delivery begins.
Named quotes go here as clients approve them. See what the engagements produced →
Next step
Pick the task your team complains about most.
That complaint is usually a good estimate of where the hours are going. Describe it and we will tell you how much of it can come off their week, and how much of it should not be automated at all.