Memecoin trending mechanics
What trending means on each surface, which parts of it are published, which parts are inference, and why one token can top a screener while nobody is talking about it.
Read the overviewHow a token becomes visible
Nobody finds a memecoin by scanning the chain. They find it on a screen that already decided what to show them: a trending list, a filtered feed, a chat somebody trusted enough to stay in, a chart shared without a caption.
This desk describes those screens and the sorting rules behind them. Where a platform documents how a list is built, we quote the documentation. Where it does not, we say what is observable and label the rest as inference, because guessing at an algorithm and presenting the guess as fact is how most launch advice goes wrong.
What this desk separates
Four words that get used interchangeably and mean four different things.
- Surface
- A place a token can appear: a screener list, a feed, a chat, a launchpad board.
- Signal
- The quantity a surface sorts or filters by, whether or not the surface publishes it.
- Activity
- Transactions that happened. Countable, on chain, and indifferent to who caused them.
- Attention
- People choosing to look. Not directly countable, and never produced by an instruction.
The seven discovery surfaces
Almost every first impression of a memecoin arrives through one of these. Each was built for a different job, which is why the same token can dominate one and be entirely absent from another.
01
Launchpad boards
The live board on the platform where a token was created, showing new mints and the ones approaching a migration threshold. Audience is already there and already deciding.
02
Screeners and aggregators
Third-party sites that index pairs across venues and let people sort and filter them. Most of what anyone calls a trending list lives here rather than on the chain.
03
Trading terminals
Fast-execution front ends with their own token lists and their own filters, aimed at people who intend to trade within seconds rather than read anything.
04
Telegram and Discord rooms
Private and semi-private space. Nothing is indexed and nothing is searchable, which is exactly why one forwarded message can outperform an entire public feed.
05
The public timeline
Crypto Twitter and its equivalents: an open feed with a ranking layer nobody outside the platform can inspect. High ceiling, high variance, shaped by who reposts.
06
Alert bots and scanners
Automated posts fired by an on-chain condition into a channel that subscribed to it. They reach people who never open a screener and never read a thread.
07
Wallets and swap front ends
The token list inside an app somebody already trusts. The slowest surface to reach, the hardest to influence, and the one that carries the most weight when reached.
Featured field notes
Four pieces that cover the ground most people arrive looking for: what trending means, what each list rewards, why turnover and interest are different numbers, and what happens on day two.
What ranking signals reward
Sorting a list is a design decision with consequences. What a turnover sort, a percentage-change sort and a recency sort each select for, and the failure built into each.
Read the analysisAttention is not volume
Turnover counts value that moved. Attention would have to count people who looked. The four cases where the two numbers pull apart, and what each case looks like from outside.
Read the comparisonSustaining attention after day one
Launch attention decays because the event that produced it has ended. What a second week can honestly be built on, and the four things that only look like a plan.
Read the methodThree sections, one question
Surfaces are where a token can be seen. Signals are what those places sort by. Communities are the layer that decides whether anyone is still looking a week later.
Surfaces
The screens, feeds and lists a token can appear on, what each one is built to do, and which of them a launch team can observe directly rather than guess at.
Open the sectionSignals
Sorting rules, published filters and the difference between a measurement a surface states openly and a mechanism people infer from watching it behave.
Open the sectionCommunities
Chats, holder bases and the people who keep talking after the first day. How groups form, what makes them durable, and where honest growth stops being possible.
Open the sectionThree claims this desk does not accept
Each of these is repeated often enough to sound settled. In each case the claim and the mechanism it rests on say different things.
The claim
Trending is one thing
People say a token is trending as though there were a single list. There is not. Every surface builds its own, from its own data, over its own window.
The mechanism: a screener ranking pairs by turnover and a feed ranking posts by engagement are measuring unrelated quantities. Topping one implies nothing about the other.
The claim
Enough activity forces a listing
The idea that a number crossed is a switch flipped. Some surfaces do publish numeric filters; most also apply review, deduplication and exclusions they describe only loosely.
The mechanism: a published minimum is a floor to be eligible, not a guarantee of placement. Where a surface says nothing about the rest of its process, the honest position is that we do not know it.
The claim
Attention can be manufactured
Activity can be produced, and openly is. Attention is a decision made by a person who could have kept scrolling, and no amount of purchased motion makes that decision for them.
The mechanism: surfaces mediate, they do not create. Producing activity can change what a screen displays. Whether anyone stops on it is settled somewhere no vendor controls.
Activity is one input among several
A screener that sorts by turnover is measuring something real: value moved through a pair inside a window. It is not measuring interest, and it does not claim to. That gap is why some teams treat produced activity as a distribution expense, in the same category as an ad placement, and run it openly.
The desk takes no position on whether that is worth doing. What it insists on is describing the practice accurately. Produced activity changes what a turnover-sorted screen displays. It does not change whether a person who sees the row decides the token is worth a second look, and any vendor claiming otherwise is selling a result nobody can deliver.
The reason to understand the tooling is that its output is one of the things you will be reading when you try to interpret a chart. Knowing what produced flow looks like is a prerequisite for telling it apart from the rest.
What the transaction record shows
- Which program executed a swap, and therefore which venue carried it.
- Whether one signer sent many transactions or many signers sent one each.
- How evenly spaced the trades were, which no crowd manages naturally.
- Whether buy value and sell value across a window came out near symmetric.
- Whether pooled depth grew alongside turnover or stayed exactly where it was.
- What share of a session passed through a single pool rather than several.
How this desk observes attention
Three commitments that decide what gets published here and, far more often, what does not.
Documented, observed, inferred
Every claim about a platform belongs in one of three buckets. Documented means the platform published it. Observed means anyone can watch it happen. Inferred means the desk worked it out and could be wrong. The bucket is stated on the line, not in a disclaimer at the bottom.
No tactic is promised to work
You will not find a claim that posting at a particular hour, seeding a particular chat or crossing a particular threshold produces attention. Those claims require evidence nobody has. What can be described is the mechanism a surface runs on and what it therefore tends to select.
Nothing that deceives a reader
No purchased followers, no fabricated chat activity, no coordinated posting dressed as spontaneous enthusiasm, no undisclosed paid promotion. Paid promotion appears here only in its disclosed form, because the disclosed version is the only one that survives being found out.