For operations and customer service leaders.Redesign the work first. Then decide what AI does.
Operations is where AI's promise is most concrete and where it most often disappoints. The difference is usually whether the work was redesigned before it was automated, and whether anyone was honest about the saving.
If you run operations or service, you may be hearing this.
- “We were told the chatbot would cut contacts. It moved them.”
- “Finance has already banked a saving my team hasn't made.”
- “My people don't know whether this is here to help them or replace them.”
- “We automated the process. It's still the wrong process.”
Capacity is not the same as cash
The most common flaw in an operational AI business case is one line: hours saved, multiplied by a salary, presented as a saving. Nothing has been saved until spending falls, or until the hours are used for something worth more.
Released capacity can still justify the investment through shorter queues, better service or room to grow without hiring. But it has to be named for what it is, or the operations leader ends up owning a target that was never real. We keep the four kinds of value in separate columns from the first estimate, and records how confident anyone should be in each.
The decisions on your desk.
- 01
What the work is for
Before any automation, ask what value each step creates and what should simply stop. Automating a poor process preserves the wrong work.
- 02
What stays with a person
For every workflow, decide what stays with a person, what AI assists and what AI takes on within agreed limits. Write it down so your team knows where they stand.
- 03
Who is accountable when AI acts
Each redesigned workflow names an accountable owner and sets out how a decision can be reviewed and reversed.
- 04
What the saving really is
Time saved creates capacity. It is a cash saving only when spending is reduced or avoided. Decide in advance which you are aiming for, and tell finance.
What tends to go wrong.
Savings counted before they exist
Released hours are booked as cost reduction. The headcount and the budget stay the same, and the business case quietly fails.
Deflection instead of resolution
Contacts are pushed to a channel that cannot resolve them, and come back as complaints.
A team left guessing
When nobody has said what stays with people, the people assume the worst and adoption stalls.
How the four stages apply to an operation.
Stage 01
Decide
Choose the workflows where a change would move cost, capacity or customer outcomes most, with current performance measured where evidence exists.
Stage 02
Design
Redesign each workflow, its roles and its decision rights, including the boundary between people, AI assistance and automation.
Stage 03
Deliver
Pilot in a small, time-boxed way with measures agreed in advance. Proceed, adapt or stop.
Stage 04
Embed
Train people on the new work using real cases, name owners and track results your team can run.
Start with one or two workflows and an honest measure of each.
The Diagnostic takes a week and looks at one or two priority workflows. You get prioritised opportunities, an initial value hypothesis with its assumptions stated, a readiness, risk and dependency view, and a recommended next step. Where the main gap is confidence and skill, AI Training for Leaders and Teams is built on your own work.
Engagement 02
AI Advantage Diagnostic
£6,0001 week
- Prioritised opportunities for the areas in scope
- An initial value hypothesis, with its assumptions stated
- A readiness, risk and dependency view
- A recommended next step
Also worth a look: AI Training for Leaders & Teams (quoted after scoping).
Questions we hear.From this seat.
Will this lead to job cuts?
That is your decision, and it should be made openly. The work makes the choice visible by separating released capacity from cost savings, and by setting out what stays with people in each workflow. Many businesses use the capacity to grow or to improve service.
Do you implement contact-centre or automation platforms?
We do not sell or install software, but we stay with the work until it is live. We redesign the workflows, specify pilots and acceptance criteria, help you choose and brief the partners who implement, and test the result against the agreed measures.
How do you measure whether it worked?
Against measures agreed before the pilot starts. Revenue, cost savings, released capacity and customer outcomes are tracked separately, with a named owner for each result.
Can you train my team?
Yes. Team sessions use real work from your operation, cover what stays with people, and set working norms that fit your governance. There is also a session for leaders on judging opportunities, value and risk.
We are in a regulated sector. Does that change things?
It adds a step. For financial services and insurance, redesigned workflows are tested against Consumer Duty outcomes and conduct risk before they reach a board.
Related reading
Use caseAI operating model
Workflows, roles and decision rights redesigned around AI, with the people boundary made explicit.
Use caseAI governance
Accountability, Consumer Duty, UK GDPR and the EU AI Act, built into the design.
IndustryFinancial services & insurance
Consumer Duty, conduct risk and board scrutiny, designed in before anything is built.
Start with a conversation.Thirty minutes, free.
Tell us what you are trying to decide. If we can help, we will say how and what it would cost. If we can't, we will say that too.