Ready to Go Live Isn’t the Same as Ready

There’s a question that gets asked in almost every go/no-go meeting I’ve sat in.

“Is the system ready?”

It’s a fair question. It’s also only a fraction of the right one.

By the time an HR system reaches go-live, the technology is usually the best-understood part of the programme. It has test scripts, defect logs, exit criteria and a vendor contractually obliged to fix it. What’s far less understood is whether payroll can pay everyone correctly on the first live run, whether Finance’s numbers will still reconcile at month-end, whether managers know what they’re now expected to do themselves – and whether anyone has actually asked them.

That gap is why I’ve rebuilt my HR system implementation health check.

Where it started

The bones of it go back to a readiness assessment from my first implementation, then developed further on a group-wide e-recruitment programme in 2004 – in fact, the one where we recommended halting the platform (I’ve written it up in my case study Enterprise Programme Governance and the Judgement to Call Time on a Failing Platform). Reading it again recently, what struck me wasn’t how dated it was, but how little the problems had changed. Stakeholders who said yes in meetings and no with their actions. Local processes nobody had told the programme about. Paper forms still in circulation long after go-live. Savings in the business case that depended on switching off things nobody owned.

The technology has moved on enormously in twenty-odd years. The reasons programmes struggle mostly haven’t.

So I’ve brought it up to date – drawing on current assurance practice, people-readiness research and a fair amount of hard-won experience – and widened it well beyond HR.

Two questions, not one

One idea from the original I’ve kept, because it’s still the most useful thing in it: every measure is rated twice.

• Progress – are we doing what we need to, on time?
• Risk – how exposed are we if we’re not?

It sounds like a small thing. It isn’t. A single RAG status lets a late item with a solid recovery plan look as alarming as an on-time item carrying a serious unmitigated risk – or worse, lets the second one sit there quietly as green. Separating them changes the conversation at programme board almost immediately.

And everything is rated on evidence. No reconciliation, no test exit report, no training data, no survey result – no green. Reassurance isn’t evidence.

What it looks at

The health check covers fourteen readiness areas and around a hundred measures, which fall broadly into seven themes:

• Direction and control – is the business case still credible, are the benefits owned, and does governance actually make decisions?
• Design, data and technology – have processes genuinely been redesigned, is the data clean enough to trust, and does every integration work end to end?
• Payroll and finance – will people be paid correctly, and will the numbers reconcile?
• The connected functions – IT, data protection, pensions, procurement, audit, communications, employee relations and more.
• People readiness – are people informed, willing and able, and are their managers equipped to lead them through it?
• AI and automated decision-making – which AI features will be switched on at go-live, who decided, and can people still review and challenge what they do?
• Go-live and beyond – cutover, hypercare, decommissioning, and whether the benefits ever turn up.

There’s a UK edition, with the payroll detail that real-time reporting, pensions and statutory payments demand, and an international edition for organisations running payroll across several countries, entities and currencies.

HR systems don’t live in HR

The part I’ve expanded most is the one programmes most often underestimate: the functions that depend on HR data but aren’t in the room.

Payroll is the obvious one. But Finance relies on the HR system for its largest cost line. IT relies on it to give people access on their first day – and to take it away on their last. Data protection, pension providers, recruitment, learning, health and safety, employee representatives: each has a stake, and each tends to hear about go-live last and feel it first.

So the health check maps those dependencies and asks each function to rate, and sign off, its own readiness. It shouldn’t be completed by HR or the programme team alone. If Finance hasn’t said it’s ready, it isn’t.

People readiness is a gate, not a workstream

Prosci puts it neatly: technical readiness gets your system live; people readiness gets you the return.

Yet in too many programmes, people readiness is measured by activity – emails sent, sessions delivered, e-learning completed. Those are inputs. The health check looks for outcomes. Do the groups most affected understand what’s changing for them, and why? Were managers equipped before their teams? Has anyone checked that people can actually do the new task, rather than just attended the training? Will anyone be measuring whether the new way of working is still being used three months on?

People readiness has sixteen measures of its own, and it’s assessed with exactly the same rigour as testing and data.

If people aren’t ready, the organisation isn’t ready – however good the system is.

How deep does it go?

Deep enough to be uncomfortable – in a useful way. A few of the questions it asks:

• Can you show me the reconciliation from your last payroll parallel run – with every variance explained?
• If someone can be added to payroll without going through the HR process, how would you know?
• Who agreed the go/no-go criteria – and was it before the go/no-go meeting?
• What proportion of managers could approve a change in the system today, without phoning HR?
• Which savings in the business case depend on switching something off – and who owns switching it off?

None of these are trick questions. But it’s surprising how often they don’t have a ready answer. I learnt that from one of the best project managers I’ve worked with. She was brutal: if you told her something was done, she’d ask whether it was “done, done and done” – carried out, documented and proven. It was hard and uncomfortable at times, but it worked.

When to use it

Findings are cheapest early. I’d use it at four points – design sign-off, before user acceptance testing, at go/no-go, and around ninety days after go-live – with a lighter monthly check in between and an optional benefits review at around twelve months. The same finding that costs a conversation at design sign-off can cost a recovery programme after go-live.

What you get

• An evidence-based view of readiness for the sponsor and board, with an overall delivery confidence rating they’ll recognise.
• A shared language across HR, payroll, finance, IT and the business.
• Risks surfaced while they’re still cheap to fix – each with an owner, an action and a date.
• Signed-off readiness from the functions that have to live with the result.
• A go/no-go decision made on evidence rather than optimism.

I’m not going to publish the full tool here – partly because every programme needs it tailored, and partly because the value is less in the checklist than in knowing which answers to push back on. But if you’re three to six months from go-live, or you’ve already gone live and the benefits aren’t showing up, I’d be glad to talk.

Because the real question isn’t whether your system is ready to go live.

It’s whether your organisation is ready to live with it.

Related insight

What a Transformation Health Check Should Really Tell You

Go-Live Is Not the Finish Line

The Governance Gap: Why HR and Payroll Need More Than a One-Way Integration

Sources

• Prosci, “Technical Readiness Gets Your ERP Live; People Readiness Gets You ROI”.
• Prosci, “The Prosci ADKAR Model”: awareness, desire, knowledge, ability and reinforcement.
• UK Government, “Gate 4 Review: Readiness for Service” and the delivery confidence assessment used in project assurance.