Most software vendors stop paying attention once the system is live
The conventional model in healthcare IT is heavy at the start, light afterwards. A long implementation, then a support contract. The customer gets a system; the vendor moves on.
We don’t think that’s enough. The interesting questions only start to surface once a system has been in real use for a while - and by that point, the people who deployed it are usually long gone.
This is a story about what happens when they aren’t.
Who's behind this approach
The question that started it
Hvidovre Hospital, in the Capital Region of Denmark, has been using Columna Flow Task Management for some time. Earlier this year, a question came up - one that required understanding the hospital’s workflow in more detail than a meeting could cover.
“It wasn’t something I could answer with a demo or a slide deck,” says Laura Aagaard Nielsen, the Customer Project Manager who took the work on. “I needed to actually understand how their porter team works. What their patterns look like. What’s getting in the way.”
Hvidovre already shares this kind of operational data with us through their existing setup. Laura got the go-ahead to take a closer look - and that’s when the real work started.
Sitting with the data
Laura - who, in her own words, “accidentally became a Power BI wizard” along the way - built the analysis herself. A colleague in Support helped with the data extract. The interpretation, she did from scratch.
The engagement at a glance:
Customer: Hvidovre Hospital, Capital Region of Denmark.
Product in use: Columna Flow Task Management.
Data analysed: One year of patient transport records.
Method: Data analysis paired with several on-site working sessions.
Status: Ongoing.
The first finding was familiar to anyone who has worked in a hospital. Tasks peak around 10am as morning rounds get underway, and waiting times build through the day. Nothing unusual.
The second finding was more interesting. Imaging departments were a disproportionate contributor to waiting times - no great surprise - but the pattern within imaging was unexpected.
What the data showed about imaging
“In the morning, transport runs one direction. Patients to imaging only,” Laura explains. “In the afternoon, it runs both ways. The same team is doing two jobs at once, so output goes up without any change in effort. It’s the shape of the work that changes, not the team.”
That single observation reframed the question. The afternoon wasn’t faster because of the team’s behaviour. The morning was constrained by the structure of the workflow itself.
A third finding sealed the picture. Looking at how the porter team handled tasks during peak hours, Laura noticed something quietly impressive: porters and supervisors were manually assigning tasks ahead of the morning peak, getting work distributed before waiting times built up. They weren’t waiting for the system to flag the problem. They were managing the peak themselves.
“They’ve already worked out how to handle the morning surge. The interesting question now is how we can support what they’re already doing well — rather than trying to tell them how to work.”
Pairing data with the floor
Laura visited Hvidovre several times over the course of six weeks. Each visit was both an analysis review and a working session with the portering management team.
The data told her what was happening. The conversations told her why.
“You can stare at a Power BI dashboard for a week and miss things that take five minutes to understand if you sit with the people doing the work. They know exactly when their pressure points are. They know which corridors slow them down, which wards send last-minute requests, which transports tend to spiral. None of that is in the data — but once you’ve heard it, you read the data differently.”
This pairing - data plus on-site observation - is the part she keeps coming back to. Not the analysis itself. Not the findings. The combination.
“It’s not a methodology I learned anywhere. It’s just what the work asked for. Hvidovre had questions that needed answering, and answering them properly meant doing both.”
What this kind of work produces
Five things have come out of the engagement so far, with the work ongoing.
A clearer picture of how Hvidovre’s porter team actually works - what their constraints look like, where their pressure points sit, how the system fits into their day.
A baseline. Detailed pre-engagement waiting time data is now documented at the level of individual departments, days of the week, and hours of the day. Anything that changes from here can be measured against it.
Direct input to product development. Several observations from the porter team are now being considered for future task management functionality. The kind of insight that doesn’t come from a roadmap workshop.
A stronger relationship with the customer. Hvidovre’s porter team has been part of the analysis from the start. They aren’t waiting to see what we deliver - they’re shaping it.
And a way of working that we want to do more of.
This is not a service line. It’s not a product offering. It’s how we think about being a software vendor in healthcare.
What customer success actually looks like
Software working as designed is the floor, not the ceiling. The value of a system like Columna Flow is realised over years of operational use — and that requires sustained attention from the people who deployed it, not just the people who sold it.
How we approach customer success:
We treat go-live as the start of value realisation, not the end.
We pair system data with on-site observation.
We build measurement baselines before we change anything.
We bring real-world insights back into product development.
We partner with operational teams, not just IT departments.
For Hvidovre, that attention has meant a Customer Project Manager spending real time in their data and on their floor. It has meant pairing what the system shows with what the porter team knows. It has meant turning a year of records into a picture detailed enough to actually act on.
We don’t think this is unusual. We think it’s what customer success in healthcare IT should look like.
We’ll have more to share as the work continues.
Do you want this kind of partnership?
The Hvidovre engagement is one example of how we work with customers post go-live. If you're evaluating how a vendor will support you over the long term - or you're an existing customer wondering what we could find in your data - we're happy to talk.
Contact Jacob Gade, Columna Flow Sales Director.
Email: jacob.gade@systematic.com