Star ratings are the wrong model for care work

When people talk about making care work visible, the conversation often jumps straight to ratings. Give every worker a score out of five, the thinking goes, and quality becomes legible. We think that is the wrong model, and a risky one in a safeguarding context.

What ratings get wrong

A star rating is reductive by design. It compresses a person's judgement, reliability, warmth and skill into a single number, then strips the context that made it meaningful. In consumer marketplaces that trade-off is tolerable. In care, where the stakes are a vulnerable person's wellbeing, it is not. A number out of five tells a future employer almost nothing about whether a worker is good with people living with dementia, or calm in a crisis, or trusted by the families they support.

Ratings also tend to be extractive and fear-driven. They reward the avoidance of complaints rather than the presence of skill, and they put the people doing the hardest work permanently on the defensive.

What other industries actually did

The industries that made invisible work visible did not do it with stars. They did it with portable professional identity:

• Developers built reputations through a visible body of work that follows them between jobs.
• Professionals carry recommendations and a record of what they have done from one employer to the next.

The common thread is not a score. It is a record of real work and real regard that belongs to the person, not the platform, and travels with them.

Recognition, not ratings

Care needs the same idea, shaped for its context. Recognition that is specific and meaningful ("trusted with complex care", "reliable across years of shifts") tells a future provider something a number never could. And it can run both ways: workers should be able to learn which providers are good to work for, just as providers learn about workers.

How feedback works in practice

In the product we call this feedback, and we have built it to lean towards recognition rather than rating. It only carries weight when it comes from someone real, so every piece of it is anchored to either a verified identity or a working relationship we can stand behind. It runs in both directions.

A provider's feedback on a worker. A provider can only leave feedback for a worker whose Career Passport they have access to, or previously had access to. That single rule means it always comes from a genuine working relationship: someone who actually engaged that worker, never a stranger or a competitor. The feedback is specific, and it becomes part of the worker's profile, so good work is visible to others rather than locked inside one employer's files. The point is recognition: a record of regard that travels with the worker.

A worker's feedback on a provider. It runs the other way too, and this is the part the sector rarely offers workers. Someone who has worked for a provider can share how likely they would be to recommend it as a place to work, and the context behind that. This counts only once the worker has verified their identity, so it comes from a real, known person rather than an anonymous account, and we pair it with specifics so it stays meaningful rather than collapsing into a number.

These are two different kinds of feedback, and it helps to keep them apart. A worker's feedback on a provider answers "is this a good place to work", and it gives workers something they have rarely had: a way to choose good employers. A provider's feedback on a worker answers "what is this person like to work with", and it builds a professional identity the worker carries with them. Different questions, different audiences, both grounded in something real.

We are very conscious that this is sensitive territory, so the design leads with fairness. Most feedback is everyday signal: reliability, teamwork, how someone handled a tricky shift. Anything that rises to a serious concern goes through a deliberate review process and is examined thoroughly before it is ever made visible. If we have reason to believe feedback is false or misleading, we can act on the profile that left it, but we do so carefully and proportionately, never on a hair trigger.

Some issues are graver than a platform should settle on its own. For those, the responsible thing is making sure the people who should know are informed, including the agencies that maintain the banning registers. We have not finished designing how that escalation should work, and we would rather say so plainly than overstate it. Getting it right matters more than getting it quickly.

The whole point is to make good care and good workplaces visible, and to help people feel recognised for work that usually goes unseen. Because every contribution is anchored to a verified identity or a real working relationship, the same design quietly makes the network hard for bad actors to game. That is a welcome by-product, not the goal.

How this connects to the Career Passport

A Career Passport is built to carry both: a worker's reviewed credentials, and, over time, the recognition of their work, owned by the worker and travelling with them. The point is not to score people. It is to make sure that years of good care do not vanish the moment someone changes employer.

That is what visibility should mean in this sector: not a number, but a professional identity that finally follows the person who earned it.