I've been making tier lists for developer communities, side project roundups, and the occasional "tools I actually use" thread since around 2022. I thought I was doing it right. I was not doing it right.
Here are 12 things I've noticed — some from my own mistakes, some from watching other people make the same ones.
1. The default tier labels are a trap.
S / A / B / C / D feels intuitive until you realize nobody agrees on what "S-tier" means. Does S mean "the best"? "Situationally perfect"? "I would recommend this to a stranger"? I've had comment threads derail entirely because two people had different mental models of the same letter. If your tiers don't have explicit criteria, you're not ranking anything — you're just sorting by vibes.
2. The number of tiers you use changes the argument you're making.
Five tiers implies a spectrum. Three tiers implies a judgment. Two tiers is basically a binary. I used five tiers for years because it felt more nuanced. Then I realized I was using B and C interchangeably and the whole middle of my lists was meaningless. Sometimes three tiers is more honest than five.
3. Long tier lists are decision fatigue machines.
I ran an informal experiment: I made a 30-item list and a 15-item list on the same topic, one week apart. The 15-item list generated twice the comment engagement per item. People actually argued about specific placements. The 30-item list got a few "interesting" replies and nothing else. Shorter lists force you to be more selective, and that selectivity is visible to readers.
4. The items you rank last are almost always ranked wrong.
This is the one I'm most confident about. By the time you're placing item 25 of 28, you've already made 24 decisions. Your brain is tired. You stop asking "where does this actually belong" and start asking "where can I put this quickly." My B-tier is a graveyard of items I ran out of energy to think about. (I suspect yours is too.)
5. Tool friction shapes your thinking more than you realize.
I switched from a spreadsheet-based system to a dedicated tier list maker about six months ago — specifically Tier List Maker — and the change in my process was immediate. Not because the tool was smarter than a spreadsheet. Because the interface was designed for this specific task, which meant I spent less cognitive energy fighting the tool and more actually thinking about the rankings. The output quality improved. I don't think that's a coincidence.
6. "I'd use this in production" is a better tier label than "A."
I started labeling my tiers with actual criteria instead of letters or grades. "Ship it," "worth evaluating," "specific use cases only," "avoid." The comment quality changed immediately. People stopped arguing about whether something was A or B and started arguing about whether it actually belonged in "ship it." That's a more useful argument.
7. The items you're most uncertain about are the most interesting ones.
I used to hide my uncertainty by defaulting to the middle tier. Now I flag uncertain items explicitly — sometimes with a note, sometimes with a separate "genuinely unsure" row. Those flagged items consistently generate the most discussion. Readers have context I don't. Showing your uncertainty invites them to contribute it.
8. A tier list is an argument, not a poll.
This took me embarrassingly long to understand. A poll asks "what do you think?" A tier list says "here's what I think, and here's the implicit reasoning behind it." The difference matters because it changes what you're asking readers to engage with. If you're not making an argument, you're just making noise.
9. The visual layout affects how people read the ranking.
Items at the top of a tier get more attention than items at the bottom of the same tier. I started treating within-tier ordering as meaningful — left-to-right as a secondary ranking. Some readers pick up on this. Others don't. But it forces me to be more precise about my own thinking, which is worth it regardless.
10. Revisiting old tier lists is more useful than making new ones.
I went back through twelve months of lists last winter. About 30% of my placements had changed — not because the tools or topics had changed, but because my understanding had. Publishing a "revised" tier list with explicit notes on what moved and why turned out to be more engaging content than the original list. The delta is the story.
11. The best tier lists have a point of view that's slightly
uncomfortable.
The lists that generate real discussion aren't the ones where everyone agrees. They're the ones where at least one placement makes someone go "wait, really?" That placement is the hook. It signals that you actually thought about it rather than just reproducing consensus. If your tier list looks exactly like every other tier list on the same topic, you haven't added anything.
12. Most tier list drama is actually about criteria, not placements.
When someone says "X doesn't belong in B-tier," they almost always mean "I'm using a different definition of B-tier than you are." The argument isn't about the item — it's about the framework. Making your criteria explicit doesn't eliminate disagreement. It makes the disagreement more productive. Which is, honestly, the whole point of publishing a tier list in the first place.
I'm still making tier lists wrong in various ways. But I'm making them wrong more deliberately now, which feels like progress.
Somewhere around observation 7, I started keeping a running doc of "items I'm genuinely unsure about." It's longer than I'd like. I add to it more than I remove from it.
I'm not sure what that means yet.
Top comments (0)