Three Groups, No Authority
An NHS clinical-tool rollout at Oracle. Getting people who did not report to me to agree.
- Facilitates and convenes
- Senior stakeholders
- Shapes strategy
- Healthcare
- Where
- Oracle, for an NHS client
- When
- 2025
- My role
- Senior Service Designer and Programme Manager
- Themes
- Influence without authorityStakeholdersClinical safetyPhased rollout
90 min
to an agreement 3 groups had blocked for 2 months
-
01Situation
Clinical leads, information governance and operational managers all had a real say in a new clinical tool, and they wanted 3 different things. None of them reported to me.
-
02The move
I met each group alone and asked what they thought the other 2 wanted. That surfaced the caricatures, so I replaced them with what each group had actually said, then brought everyone to 1 page with 3 possible designs.
-
03Result
They agreed in about 90 minutes: optional for 8 weeks in 2 pilot wards with a safety review built in, then mandatory, with the data restriction designed in from the start.
Three circles that meet
3 groups, what they said about each other, what they actually meant, and where they met.
The situation
At Oracle I was running a programme for an NHS client to introduce a new digital tool for clinical teams. There were 3 groups of stakeholders who all had a real say in it. The clinical leads, who cared about patient safety and about not adding to their teams’ already heavy workload. The IT and information governance team, who cared about security and data protection. And the operational managers, who cared about cost and about hitting the targets they personally were being measured against.
None of them reported to me. I did not even work for their organisation. I had exactly as much formal authority over them as a stranger would.
The clash
About 2 months in, it became clear that they wanted fundamentally different things from the same product. The clinicians wanted it simple and optional to begin with. The operations managers wanted it mandatory from day one, because the entire benefits case depended on adoption. And information governance wanted to restrict what data the tool could show, which would have made it far less useful to both of the others. Each group had escalated to their own senior leader, so I was receiving 3 different sets of instructions at once.
What I did
I did not try to solve it in a big meeting, because I had learned that when people are dug in, a big meeting just gives them an audience to perform their positions for. Instead I met each group separately, and I did something slightly unusual. I asked each of them to tell me what they thought the other two groups wanted, and why.
What came out was that each group was carrying a caricature of the others. The clinicians thought operations only cared about numbers. Operations thought the clinicians were simply resisting change for its own sake. And both of them thought information governance was just being obstructive for the sake of it. None of these was true, but each one was blocking any possibility of agreement.
So in each of those conversations, I shared, carefully and fairly, what the other groups had actually told me. When the clinicians heard that operations were worried because their funding genuinely depended on showing adoption, they softened. When operations heard that the clinicians were worried about a specific, real patient-safety scenario, they became far more open to a phased approach. And when both of them heard from information governance about the specific data rule that was worrying them, which turned out to be a real legal constraint rather than caution, they stopped seeing it as obstruction.
Then I brought them together, but not to argue. I put up a single page showing what each group actually needed, as opposed to what they had each been demanding, alongside 3 possible designs. We went through them together.
The moment that tested me
The hardest moment came earlier in that process. The senior operations director told me, quite firmly, to just make the tool mandatory and let the clinicians adjust. I disagreed, and I told him so, privately. I explained that I thought a mandatory launch without genuine clinical support would lead to low-quality use and workarounds, which would quietly destroy the very benefits case he cared about most. And I told him clearly what I would need from him if he still wanted to go ahead anyway: a visible clinical sponsor. He did not enjoy hearing it, but he agreed to let me try the other approach for 2 weeks.
Outcome
In about 90 minutes together, they agreed a phased approach: optional for the first 8 weeks in 2 pilot wards, with a specific safety review built in, then mandatory, with the data restriction designed in from the start. Everyone gave something, and everyone got the thing they actually cared about most.
The lesson
Influence without authority is mostly about understanding. People dig in when they feel misunderstood.
If you can show each group that you genuinely understand their position, and then help them understand each other, the solution very often becomes obvious on its own. And when you do have to disagree with someone senior, you do it privately, with evidence, and you give them something to hold on to.
How I found out .
Each method, and why I chose it.
WhyDug-in people perform their positions in big rooms. Alone, they explain them.
The people it had to work for.
Composite and anonymised. Needs and frictions kept, names removed.
Persona 1 of 4
The clinical lead
Accountable for patient safety on wards that are already stretched.
- Needs
- They need a tool that is simple, optional to begin with, and demonstrably safe.
- Friction
- Mandates that create workarounds, and workarounds that hurt patients.
Written on the
inside cover
- Each group was carrying a caricature of the other 2. None of them were true.
- The IG objection was a legal constraint, not caution. Once named, it was easy to design in.
- Operations' funding genuinely depended on adoption. Once the clinicians heard that, they softened.
- The patient-safety scenario was specific and real. Once operations heard it, phasing made sense.
-
“Just make the tool mandatory and let the clinicians adjust.”
The senior operations director, before he gave me 2 weeks to try it the other way
What we made to make the decision.
Artefacts first, arguments second.
Most of this is under NDA, so each piece is described rather than shown.
-
Perspective-taking interview guide
4 questions per group. The third was always "what do you think they think of you?"
-
Needs versus demands page
3 columns, 2 rows. The page the room agreed from.
-
3-design trade-off sheet
Mandatory day 1, optional forever, phased with safety review. The third won.
-
Phased rollout plan
8 weeks optional in 2 wards, safety review, mandatory with the restriction built in.
What changed, and what I keep.
- 90 min
- to an agreement 3 groups had blocked for 2 months
- 8 weeks
- optional phase in 2 pilot wards, with a safety review
- 2 weeks
- the operations director gave me to try it his way or mine
- 0
- reporting lines I had over any of them
- The data restriction information governance needed was designed in from day 1, not retrofitted after a breach.
- The benefits case the operations director cared about survived, because adoption came with clinical support instead of workarounds.
- The clinicians got their safety review in writing, which set the pattern for how the tool was changed afterwards.
- 01 Influence without authority is mostly about understanding. People dig in when they feel misunderstood.
- 02 Translation beats debate. Carry each group's real reason to the others, carefully and fairly.
- 03 When you disagree with someone senior, do it privately, with evidence, and give them something to hold on to.
I would have run the perspective-taking conversations in week 1 rather than month 2. By the time I asked each group what the others wanted, the caricatures had already hardened into escalations.