Benefits after go-live — the business case is not a wedding album

A business case is a bit like a wedding album. Enormous effort goes into it. Everyone looks lovely. It gets admired once, on the day, and then it goes on a shelf where nobody opens it again until something has gone badly wrong and someone wants to know what was promised.

Every IT and digital transformation programme starts with one, and most business cases are excellent documents. They name the benefits, quantify them, and tie them to the investment. Then the funding is approved, the document is filed, and from that day the programme is managed against scope, schedule and budget. The benefits are mentioned again at the end, once, in a review whose main purpose is to explain the variance.

I have watched programmes deliver on time and on budget and fail, because the thing they delivered no longer produced the value the case promised, and nobody had been looking.

Why benefits get lost

Three reasons, in my experience, and none of them is malice.

The benefit has no owner. "Reduce order-to-cash cycle time by four days" is a fine line in a spreadsheet, but no named person will be measured on it after go-live, so nobody defends it when scope is cut in month seven. The benefit is an orphan, and orphans lose arguments to deadlines.

The benefit has no cadence. Delivery is reported weekly. Benefits are reported never, or once at the end, so there is no moment at which a trajectory going wrong is visible while it can still be corrected. You do not weigh yourself once a year and call it a diet.

The benefit is disconnected from delivery. The plan lines that produce it — the process redesign, the data cleansing, the training wave — are managed as tasks with no visible link to what they were for. When they slip, the benefit slips silently, and the plan still says green because the tasks that survived were done on time.

The Draghi report puts the investment Europe needs in the hundreds of billions a year and asks for investment that demonstrably pays back. Inside a company the same demand lands on the CIO's desk: money moved is not a result. Somebody has to be able to say what came back, and "we went live" is not an answer.

Governance, not a metrics platform

The tempting response is a benefits dashboard: connect it to the data warehouse, chart the KPI, watch the line. That is a metrics platform, and it has two problems. It is expensive to build for every benefit of every project, and it answers "what is the number" without answering "who is accountable and what are we doing about it". A line going the wrong way on a dashboard nobody owns is just a sad line.

Value realisation in TransformRadar is deliberately governance rather than analytics. It holds four things for each benefit the business case promised.

• A value driver with a named owner — a person who will be asked, on

cadence, how it is going, and who is expected to know.

• Measures with a baseline and a target — the number the case promised,

and where it started, so a reading means something.

• Readings on cadence — reported by the owner, including through an

emailed pulse that lets them answer without opening the product, so the trajectory has data points while the programme is still running and can still change course.

• A trajectory the steering group can see — next to delivery, in the

steering pack, on the portfolio view, with the same amber and red logic as everything else.

One deliberate rule sits underneath. A driver's owner may record readings on their own driver even when their project role is otherwise read-only. It is the single place a read-only member writes to the record, and it exists because the person who knows the number is often a business owner who has never opened a project tool and would like to keep it that way.

How AI helps, and where it stops

The AI's job here is to make the setup less of a chore. Given the project's context and business case, it proposes value drivers and measures for the lead to refine, rather than starting from a blank form and a sigh. It draws only on the project's own record, and the lead edits before saving. It does not invent a benefit the case never promised, and it does not report a reading on anyone's behalf. The album stays honest.

What the CIO sees

A portfolio where every programme carries its promised benefits, owned and reported, changes the portfolio conversation. "Which of these should we stop" becomes answerable, because a programme whose value trajectory went amber six months ago while its delivery stayed green is visible as exactly that. A steering group can hold a sponsor to a benefit, and a sponsor can hold a delivery team to the plan lines that produce it.

The Programme Health page shows how value realisation feeds the health assessment, and the product overview shows the value workspace beside the plan it is tied to. Neither of them goes on a shelf.

• value realisation
• benefits
• governance
• CIO