Questions readers send
Twelve questions that arrive often enough to be worth answering in one place. None of them repeats an article; these are the practical questions that sit around the edges of the reading method rather than inside it.
Do I need to write code to watch launches properly?
No. Everything described on this site can be done with a public feed, a block explorer and a text file. Code helps with one thing only: recording consistently, so that a hundred observations exist rather than a hundred impressions. If you never write a line, keeping a plain log of what you saw and what would have changed your reading will still put you ahead of watching without one.
Is a paid launch feed better than a free one?
Sometimes, and for narrower reasons than the marketing suggests. What you are buying is usually lower latency and access to fields a free screen aggregates away, such as per-transaction detail. Neither of those makes a reading correct. A paid feed with an undocumented ranking method is worse than a free one that publishes how it sorts, because an unstated instrument cannot be corrected for.
What is the difference between a launch feed and a token screener?
A launch feed shows tokens in the order they were created, so its subject is newness. A screener shows tokens that meet thresholds, so its subject is whatever the thresholds select for. The two answer different questions and are routinely used interchangeably. If you are reading a screener while thinking about newness, the selection has already been made for you.
Does the venue check anything about a token before it is listed?
Creating a token on a permissionless venue is an on-chain action, not an application. Name, image and links are supplied at creation and nothing in the process establishes that any of them are connected to anything. Treat every piece of descriptive metadata as a claim made by whoever paid the creation fee, because that is exactly what it is.
Do the social links on a listing mean anything?
They mean a link was entered. They do not mean the destination belongs to the token creator, that it existed before today, or that anybody at the other end knows the token exists. Links are the single cheapest field on a listing to populate. If you are going to use them, use them as something to check rather than as something that has been checked.
Can I reconstruct a launch after it happened, or only watch live?
Reconstruction is usually better. The ledger keeps every transaction, so the opening sequence of a launch from last week can be read at leisure, with the outcome already known. Live watching adds urgency and removes nothing. Anyone trying to build a reading method should spend most of their early time on finished launches, where checking a reading costs nothing.
Is there a way to see a launch before anyone else?
Not meaningfully. A token becomes visible when the creating transaction is confirmed, and automated watchers see that within the same slot or the next one. The realistic difference between a fast setup and a slow one is a few seconds, and no amount of speed converts an unreadable opening sequence into a readable one. Speed is a smaller edge than it is sold as.
How much does it cost to watch this way?
Watching costs nothing beyond attention. Reading a block explorer is free, and no transaction is required to observe anything described here. The only real cost is time, and the method on this site is deliberately built to be run without ever sending a transaction, because participating changes what a reader is willing to conclude.
Does Launch Feed publish alerts, signals or a live feed?
No. This desk publishes written pages about how to read a feed and nothing else. There is no alert list, no channel, no scanner, no scoring product and no paid tier. If you encounter something using this name and offering any of those, it is not us, and we would rather say that in advance than after somebody has paid for it.
Can the reading method be automated?
Parts of it. Counting addresses, measuring spacing and flagging first appearances are mechanical and worth scripting. Funding-graph judgement and the decision about what an unusual shape means are not, because both depend on a baseline that changes and on context the ledger does not carry. Automate the recording, keep the judgement manual, and be suspicious of anything that automates both.
Does any of this apply to launches on other Solana venues?
The reading discipline transfers; the specific mechanics do not. Counting addresses rather than summing amounts, checking funding one hop back and separating protocol fields from computed ones work anywhere on Solana. Curve parameters, completion behaviour and migration rules are venue-specific and change over time, so read those from the venue rather than assuming they match what you learned elsewhere.
What should I do if a page here turns out to be wrong?
Write in and say which page and what is wrong with it. Corrections are added to the page as corrections rather than edited in silently, because a desk that quietly rewrites its own record has no standing to ask anyone to check it. Protocol details in particular go stale, and a reader who notices before we do is doing the most useful thing anyone can do with this site.