With rating systems, companies like Uber and TaskRabbit turn customers into ruthless middle managers
The rating game — Soon, you'll be able to go to the Olive Garden and order your fettuccine alfredo from a tablet mounted to the table. After paying, you'll rate the server.
Context & Ripple Effects
When this piece ran in 2015, the rating system was framed as an experiment spreading from rideshare into sit-down dining — the Olive Garden's tableside tablet was the thought experiment, with customers rating the server after paying. The subsequent coverage shows the prediction landing from both directions: UberEATS extended five-star ratings to restaurants and even individual menu items, while [[a:930909|BuzzFeed found servers at chains deploying tabletop tablets contending with inaccurate scores and inappropriate comments]].
What changed since is that the rating loop closed on the customer too. Uber first made rider ratings more visible to encourage good behavior, then added mid-trip rating with categorized written feedback, and finally moved to enforcement — banning riders whose ratings stay significantly below average. Two-way rating has become the platform's de facto HR department.
First-order effects
- Servers and drivers are now directly managed by anonymous customers in real time — the BuzzFeed reporting shows workers absorbing unfair scores and comments with no manager filtering them.
- Riders lose their one-sided position in the rating system: Uber's visibility update, mid-ride feedback, and below-average ban policy make passengers themselves subject to performance review.
Second-order effects
- Restaurants gain item-level demand signals — thumbs up/down on specific menu items feeds personalized recommendations — so the rating infrastructure doubles as a menu-optimization tool, not just worker discipline.
- TaskRabbit-style labor marketplaces compete on trust metrics: once ratings gate access to income on one platform, rivals must match the enforcement apparatus or cede the 'quality' positioning.
Third-order effects
- Customer-facing work across the service economy is restructuring around continuous, two-way scoring — the middle manager's evaluative role is being replaced by an aggregate rating that no one individual administers and no worker can appeal face-to-face.
- If the pattern holds, expect pressure for due-process standards around algorithmic termination — Uber's multi-step warning path before banning low-rated riders is an early template, but whether regulators or platforms set the rules is genuinely unresolved.
The trend: Service platforms are replacing human supervision with bidirectional rating systems that evaluate workers and customers alike, turning reputation scores into hiring, firing, and banning decisions at scale.