antidosis
$ cd ../blog
philosophyratingstrustguide

Why Antidosis Ratings Default to 10/10 — The Default Excellence Philosophy

9 May 20266 min readantidosis

Why Antidosis Ratings Default to 10/10 — The Default Excellence Philosophy

Every rating system sends a message about what it expects from humans.

Uber defaults to 5 stars and silently trains riders to downgrade drivers for trivial inconveniences. Airbnb's review system created an entire culture of hosts anxiously messaging guests to beg for 5 stars. eBay's feedback system became so toxic that the company had to overhaul it entirely.

At antidosis, we chose a different starting point: the slider defaults to 10.

Not because every exchange is perfect. Because every exchange starts from a place of trust.

The Psychology of Default Ratings

Research on defaults is clear: whatever you pre-select becomes the anchor.

  • When a form defaults to "satisfied," most people leave it there
  • When a rating defaults to 5 stars, people look for reasons to deviate downward
  • When a rating defaults to 10/10, people look for reasons to maintain it
  • This isn't about manipulating users. It's about framing the mental model.

    A 5-star default says: "Prove you deserve a good score."

    A 10/10 default says: "Everything went well unless something genuinely went wrong."

    Why 10/10 Makes Sense for Barter

    Barter exchanges are different from commercial transactions in three ways that make a 10-point default rational:

    1. Both Parties Are Vulnerable

    In a cash transaction, one person has money and the other has a service. The power dynamic is clear.

    In a barter, both people are giving something personal. The plumber is trusting the musician to show up for guitar lessons. The musician is trusting the plumber to fix the toilet properly. Neither holds all the cards.

    When both parties are vulnerable, starting from suspicion hurts both. Starting from trust helps both.

    2. Reciprocal Exchange Creates Shared Investment

    If I pay you $200 for a service and you're rude, I leave a bad review and move on. I lost money. You lost a customer. End of story.

    If I teach your kid maths for 10 hours and you're rude, I also invested 10 hours of my life. I have skin in the game. You have skin in the game. We both want this to work.

    That shared investment means most exchanges genuinely do go well. The default should reflect that reality.

    3. Small Communities Self-Correct

    On a platform with 10 million anonymous users, you need brutal ratings to surface bad actors. The sheer volume makes individual reputation less meaningful.

    On the Central Coast, with a few hundred active traders, everyone knows everyone. A bad reputation travels fast without any rating system at all. The formal rating just confirms what the community already knows.

    In small networks, the rating system doesn't need to be a weapon. It just needs to be a signal.

    What the Scale Actually Means

    RatingWhat It SignalsWhen to Use It
    10Everything went as expected. No issues.Default — use this unless something was genuinely wrong.
    9Minor inconvenience — late by 15 minutes, small miscommunication.When the exchange worked but friction existed.
    8Noticeable issue — quality below expectation, but still usable.When you're dissatisfied but the trade wasn't a disaster.
    7 or belowSignificant problem — no-show, misleading description, conflict.Only when the exchange genuinely failed.

    The "Uber Problem" We Refused to Copy

    Uber's rating system created a bizarre dynamic where:

  • Drivers with 4.7 stars get deactivated
  • Riders feel pressured to tip to avoid retaliation
  • Everyone is paranoid about the hidden number
  • The result? A marketplace where ratings don't measure quality — they measure anxiety.

    We didn't want that for antidosis. We wanted a system where:

  • A 10 means "this person shows up and does what they say"
  • A 9 means "small hiccup, still would trade again"
  • Anything below 8 means "talk to me before you trade with this person"
  • What Happens to Bad Actors

    Some people worry: "If everyone defaults to 10, how do you catch scammers?"

    Three mechanisms:

    1. Volume matters more than individual scores. One 10 means nothing. Fifteen 10s means something. Three scores below 7 means everything.
    1. Written reviews carry weight. The rating is just a number. The review text tells the story. "Showed up late but did great work" is very different from "Never showed up, stopped responding to messages."
    1. Verification layers exist for a reason. SMS, email, and credential verification create friction for fake accounts. A scammer can make one fake profile, but making ten with different phone numbers is hard.

    The Real Test: Do People Actually Use Lower Ratings?

    Early data from the Central Coast trial shows something interesting:

  • ~85% of ratings are 10/10
  • ~10% are 9/10
  • ~4% are 8/10
  • ~1% are 7 or below
  • Some platforms would see this as "rating inflation" and panic. We see it as reality matching the default. Most people who voluntarily join a barter marketplace are decent humans who want to help each other. The rating system should reflect that, not fight it.

    What We Tell New Users

    When someone asks "how should I rate this person?" we say:

    "Start from 10. Only move down if something genuinely went wrong. And if something did go wrong, be specific in the review so the next person knows what to watch for."

    That's it. No algorithms. No hidden scores. No gamification.

    Just humans rating humans the way humans actually behave when they're not being trained to look for flaws.


    Your first exchange is waiting. Post a need on antidosis. Trade something. Rate honestly. See what happens when trust is the default.

    Found this helpful? Post a need and put it into practice.

    Post a Need