So what would actually work?

Three posts in and you're probably wondering where am I going with this? Am I disillusioned and decided to air out all my pent up frustrations? Well, maybe... (checks when my next therapy session is scheduled). The truth is I shared my stories to set the stage for this post, for what would have been a better alternative. Applying a human centered thoughtfulness to dealing with the humans in organizations: their well-being, development and feedback.

And here's the thing: we all know this. Know your audience, understand power dynamics, solve the actual problem not the assumed one, etc are not radical design principles. They are not even exclusive to design. The question is why do we get to only apply them rigorously to customers and abandon the entire approach when the users are employees?

Let's use the frame we already have.

HR is a system and its products may be the listening circle, the survey, the leadership training. They are designed for a specific human being in a specific context. If you pitched any of them to a design review the way they're currently built, any design practitioner (an outspoken one at least) would flag the problems in the first five minutes.

The listening circle: place users in a room with the people who control their career outcomes and ask them to share honest feedback. Power dynamics in play that compromise the data before anyone opens their mouth. Format doesn't support the stated goal. Back to the drawing board.

The leadership training: deploy a clinical therapeutic framework without licensed facilitation, confidentiality protections, or consent mechanisms. Military imagery for visual interest. No ethical design review passes that.

The survey: remove names but retain enough demographic data to re-identify individuals in small teams. The anonymity is performed not structural. The tool is collecting data it claims not to collect. Fails the basic privacy review.

The employee — back office, front line, head office, support, call center agent, advisor, analyst, whatever the level — is. the. user. in. the. system.

Let's design for them accordingly, starting with the actual user. Not the user leadership imagines, the one who actually exists: whether the over achievers or the ones struggling. What's going on there?

Design for the constraint, not around it. If employees won't speak honestly what conditions should we design for the speaking to follow?

Test the prototype before you deploy it. Would this format survive a basic psychological safety audit? Would the anonymity mechanism survive a re-identification test? Would the training content survive a cultural sensitivity review? If the answer to any of those is no. The prototype isn't ready.

Separate the people who need to hear from the people being heard. Basic research methodology. You don't put the subject and the evaluator in the same room and call it research.

Protect the data like it matters. Because it does.

Now. What's in it for the organization?

Brace yourselves I'm going to go all business speak. The average cost to replace one employee in Canada is $30,680. Sounds manageable until you multiply it by your voluntary turnover rate. An organization of 10,000 people losing staff at the national average of 10.2% is replacing roughly 1,000 people a year. That's $30 million, annually, in replacement costs alone. Before productivity loss and institutional knowledge walking out the door. Before the remaining team absorbs the gap and starts doing the math on whether they're next. All adds up to cost and cost is bad.

That's balance sheet drama dressed up as a people problem.

It produces a dashboard someone presents to the board as evidence of a healthy culture while the actual state of things lives in quiet conversations between people who trust each other. Off the record, untracked and unsubmitted.

Here's an exercise. In a moment when mass layoffs are making headlines, count how many people you know who left voluntarily in the last six months. Before the cuts, of their own choosing. Interesting number, no?

Designing badly for people is expensive. Producing inaccurate information and calling it insight is worse.

To org leaders out there: if it helps to think of your employees as batteries — invest in recharging them. I'm not asking for decency or the right thing to do, just pointing out that drained batteries don't perform.

There is space (albeit shrinking) for making the case to advocate for people inside the system that is an organization. Whether that looks like HR and design teams working together, or people committees doing side of desk cultural work.

The shrinkage is happening in real time since organizations are already deploying AI to replace the human touch points of the employee experience — the very touch points that were already failing because they weren't designed for humans properly in the first place. Yes, apparently the solution to broken human centered interventions is less human intervention.

Here's a thought before we automate HR — design it properly first just so we don't scale the broken version. But then again we might all be replaced anyway so who cares?

That went dark very quickly lol here's a candle: in a world where we're all going to be replaced by AI anyway — treat the ones you've got left thoughtfully.