Exchange status is operational fact, not a vibe. Before you send coins, you check whether deposits, withdrawals, trading, and the specific asset-network pair are working on that venue. A green marketing site is not a status page. A status page is also not a solvency statement. It answers a narrower question: is this component degraded right now.
Most transfer pain is avoidable. People broadcast a transaction while withdrawals are paused, then spend an hour blaming the chain. Or they deposit an asset on a network the venue has not opened yet. The status page will not make you whole. It will tell you not to start.
This is a calm checklist. The range is the story on price days. On transfer days, the incident banner is the story. Read it before you sign.
Who this is for
Anyone who moves coins onto or off a centralized exchange. Beginners who treat the app as a bank. Intermediate users who already “know this” and still send during a maintenance window because the candle looked urgent.
It is also for people helping a friend. If you walk someone through a first withdrawal, you own part of the outcome when they ignore a yellow banner. Slow them down. Status literacy is cheaper than a support ticket.
If you only self-custody, still read the deposit half. You will use a venue someday for fiat or a book you cannot get on-chain. Venue risk does not disappear because your keys are elsewhere.
Prerequisites
- Bookmarks to official status pages, not search-result clones. Start with Coinbase Status and Kraken Status if those are your venues. Add every other venue you actually use.
- The exact asset, network, and destination address you plan to use, written down before you click send.
- Willingness to wait. Maintenance windows are boring. Stuck deposits are not.
- A basic read on custody: the venue holds the keys while coins sit on the exchange.
- A market snapshot habit from how to read a crypto market snapshot so a price spike does not become a reason to skip the banner.
You do not need API access. You need the official URL and the discipline to read component names.
Steps
Run this sequence before every non-trivial deposit or withdrawal. Tiny test sends still follow it. Panic sends especially follow it.
1. Open the official status page from a bookmark
Type nothing into a stressed search bar if you can avoid it. Phishing sites copy status layouts when a venue is actually down. Your bookmark should hit the domain you verified on a calm day.
If you use more than one venue, open both. A Coinbase incident does not describe Kraken. A Kraken withdrawal pause does not describe your hardware wallet. Label the page with the venue name in your head before you read a single sentence.
RSS history pages exist for a reason. If you want the last few incidents, use the venue’s own history feed, not a screenshot from a group chat. Chat lag is real. Status pages update. Threads do not.
2. Read the banner, then the component list
A global “All Systems Operational” line is the start, not the finish. Scroll the components. Trading can be up while withdrawals are degraded. The website can be up while an individual asset is paused. API can be down while the app still loads candles.
Incident language matters. “Investigating” means they do not yet have a clean story. “Identified” means they think they know the cause. “Monitoring” means a fix is in and they are watching. “Resolved” is the only state that should make you resume size. Maintenance is scheduled on purpose. Treat it as a closed window, not a dare.
Timestamps are part of the fact. Note the last update time. A two-hour-old “investigating” banner is stale information dressed as a live page. Refresh. If the timestamp does not move and withdrawals stay paused, you wait. You do not “just try.”
3. Match the incident to the action you were about to take
Deposits, withdrawals, trading, staking, and fiat rails are different machines. A trading outage can strand you in a position. A withdrawal outage can strand you in custody you wanted to leave. A deposit outage can leave coins in limbo if you already broadcast. The status line has to match the verb you are about to perform.
Then match the asset and the network. USDT on Tron is not USDT on Ethereum. Bitcoin on Lightning is not Bitcoin on the base chain. If the venue paused one network, the other may still work. Do not guess. If the status page is silent on your pair, check the in-app deposit screen for a pause badge. Two sources agreeing is better than one shrug.
If you were about to trade, read whether the book is actually open. Status can say trading is up while a specific pair is halted. Halted pairs still show prices in some UIs. A price without an open book is a museum label.
Fiat rails deserve their own glance. A crypto withdrawal can be green while ACH or wire is a mess, or the reverse. If your plan was “sell then wire out,” you need both legs. People check the coin rail and skip the dollar rail, then discover they converted into a trapped cash balance. The status page will usually say so if you scroll past the coins.
4. Check the other side of the transfer
A move has two venues or a venue and a wallet. If you withdraw from Kraken to Coinbase, both status pages matter. If you withdraw to self-custody, the chain still has to be usable, and your wallet still has to be on the right network. Dual-status checks feel slow. Dual failures feel slower.
For exchange-to-exchange, confirm the receiving deposit address and memo or tag while both venues show deposits healthy. Memos on some networks are not optional. Status being green does not fill in a missing memo. The chain will still credit a valid address, and the venue may still not credit you.
If either side shows degraded withdrawals, delay. Moving size into a venue that cannot send it back is how people become involuntary holders during an incident. Liquidity you cannot access is not liquidity.
5. Decide whether to send a test, wait, or stand down
If everything you need is operational, send a small test. Confirm credit. Then send the rest. If anything you need is degraded, wait. If the incident is “investigating” with no component detail, stand down. Curiosity is not a reason to be first in a broken queue.
Price moves will try to veto this step. A spike is not a status all-clear. Pair the wait with a snapshot so you can admit what you are actually chasing. If the only reason to hurry is a candle, you are not managing operational risk. You are volunteering to be the person who deposits into a paused rail.
Write the decision in one line: venue, component, asset, network, action, go or no-go. That line is what you reread when a friend says “just send it.”
6. After it credits, confirm you can still leave
A completed deposit is not the end of the check. Look at withdrawals again. Look at the specific asset. If you moved coins to trade, you also need a path out. If withdrawals flipped to paused while your deposit confirmed, you now have venue risk you did not plan for. Size the next action with that fact, not with hope.
Keep a short personal log of incidents on venues you use. Repeat pauses on the same rail are information. One blip is weather. A pattern is climate. Climate belongs in how much you store on that exchange.
When the incident is marked resolved, still test with a small amount if you are about to move life-changing size. Resolved is a status state. Your credit is a ledger state. Believe the ledger.
Common mistakes
- Using Google’s first result for “Coinbase down” instead of the bookmark.
- Reading “website operational” as “withdrawals operational.”
- Broadcasting a deposit after the pause banner appeared because “it might still credit.”
- Ignoring network and memo fields because the asset ticker matched.
- Checking only the sending venue on an exchange-to-exchange move.
- Treating a status page as proof the venue is solvent.
- Letting a Bitcoin spike override a yellow withdrawal component. The tape can wait. Stuck coins cannot.
Status pages do not replace on-chain explorers. If you already sent, the explorer tells you whether the transaction exists. The venue tells you whether they will credit it. You need both after a send. You needed the status page before the send.
Do not treat support chat as a faster status page. Support repeats the incident. It does not clear a paused rail because you are polite.
Related reading
Venue incidents travel next to price chatter. Keep the snapshot habit so you do not confuse a halt with a thesis. For Bitcoin-specific tape while a large venue is degraded, use Bitcoin news as context, not as a reason to skip the banner.
If the move is off-exchange, pair this guide with wallet hygiene. Status is the venue half. Keys are the other half. Green components do not read your address for you.
I keep the two status bookmarks in the same row as the market snapshot. Flows and incidents belong in the same morning pass. If withdrawals are paused on a major venue, that fact sits next to the range. It is part of how the market can actually move coins today, which is more useful than a slogan about being your own bank while you still use an app.
Build a tiny incident glossary for yourself. Write what “degraded deposits” meant the last time it happened on your venue: did credits lag, did they fail, did they later appear with a surprise network? Your personal glossary beats a generic thread. Venues rhyme. They do not copy-paste the last outage. The log is how you stop treating every yellow banner as a unique mystery.
When a friend sends a “is Coinbase down” screenshot from an unofficial account, reply with the bookmark. That is the whole social protocol. You are not winning an argument. You are pointing at the component list. If the unofficial account was right, the official page will agree. If it was wrong, you just avoided a phishing day. Either way you practiced the habit, which is the only part that compounds.