Black Friday breaks every alerting setup that was fine the rest of the year. The logic is simple: an alert config tuned for 40 orders a day becomes unusable at 400. The channel that gently pinged you about new orders in October turns into a scrolling wall of noise by noon on the 27th, everyone mutes it by early afternoon, and the one thing you actually needed to see - a $2,000 order flagged as high risk - flies past unread in the flood.
I have been through enough BFCM weekends to have strong opinions about this. The instinct under pressure is to turn on more alerts so you do not miss anything. That is exactly backwards. During a surge you want fewer, sharper alerts, because the volume is already doing the job of telling you sales are happening. What you need Slack for is the handful of events that still demand a human, and the discipline to keep everything else out of the way.
Turn these on
These are the alerts that earn their place during a surge, because each one points at a decision someone has to make in the next few minutes.
- High-value orders. Not every order - your channel would melt. Set a threshold well above your average order value, higher than your usual all-year setting because BFCM carts run bigger, and only surface the orders that deserve white-glove fulfillment or a VIP flag. These are also the orders where a fulfillment mistake hurts most.
- High-risk and fraud flags. Fraud attempts climb right alongside legitimate volume on Black Friday, and the noise of a busy weekend is exactly what fraudsters count on to slip through. A high-risk order flag going straight to a review channel lets you hold the shipment before it goes out the door, which is your only real window.
- Low stock on hero products. During a surge your bestsellers can go from healthy to zero in hours. An early low-stock ping on your top SKUs is the difference between pulling the ad or swapping the hero product in time, versus paying to send traffic to a sold-out page.
- Failed and declined payments. A spike in failed payments is not just lost sales, it is often a signal: a payment gateway hiccup, a fraud filter set too aggressively, or a checkout bug under load. During normal weeks a few declines are background noise. During BFCM a sudden cluster of them is a five-alarm fire worth investigating immediately.
- Chargebacks and disputes. These run on deadlines that do not pause for your busy weekend, and the volume makes them easy to lose. Keep them in their own channel so they never get buried.
That is the whole list. Notice what is not on it: every single order, every customer signup, every cart. During a surge those are pure noise.
Keep everything else from drowning you
Turning on the right alerts is only half the work. The other half is making sure the surge itself does not render them useless. This is where most setups fall apart, and where the controls actually matter.
- Route ruthlessly. One giant #alerts channel is where signals go to die, and on Black Friday it dies by 10 a.m. Split it: high-value orders to one channel, fraud and risk to another, inventory to ops, payment failures to a channel the person who can call the gateway is watching. Each channel should have a narrow purpose and a named owner.
- Batch the bursts. This is the single most important control during a surge. Instead of a message per order, collapse the high-volume events into a rolled-up digest every few minutes. You still see that 180 orders came in during the last window, but as one readable summary rather than 180 separate pings. Save the individual messages for the genuinely exceptional events, like a high-value or high-risk order.
- Use conditions, not just triggers. A raw trigger like new order is useless at scale. A condition like order over $400 or order flagged high risk is a signal. The tighter your conditions, the more trustworthy your channels stay when volume triples.
- Be deliberate about quiet hours. Most alerts can respect quiet hours even on Black Friday - nobody needs a low-stock ping at 3 a.m. if the reorder happens in the morning anyway. But carve out the exceptions. Fraud, disputes, and payment-failure spikes are worth waking up for, because the cost of waiting is real.
The goal is a set of channels your team still trusts on Sunday night, because they never got trained to ignore them on Friday morning.
Set it up before the weekend, not during
The one piece of timing advice I will repeat until people listen: do not touch your alert configuration on Black Friday itself. The weekend is for execution, not for discovering that your thresholds were wrong or your routing was backwards. Get it dialed in now, in early November, while you still have calm days to test it.
A sane pre-BFCM checklist:
- Run a dry run. Place a few test orders, trip a low-stock threshold, and confirm each alert lands in the right channel with the right people in it.
- Raise your thresholds for the season. Your all-year high-value threshold is probably too low for BFCM carts. Bump it so the channel does not flood with orders that are large by normal standards but ordinary this weekend.
- Confirm batching is on for your high-volume events, and off for your exceptional ones. Getting this backwards is the classic mistake - batched fraud alerts and un-batched new-order alerts is the worst of both worlds.
- Assign owners. Every channel needs a person who is responsible for acting on it over the weekend. An alert with no owner is theater.
This is exactly the kind of setup a dedicated app makes painless. Our own Slack Store Alerts covers all of these events across its 35+ triggers, with the conditions, routing, quiet hours, and batching that keep a surge readable instead of overwhelming. However you build it, the principle holds: Black Friday is not the time for more noise. It is the time for the few alerts that matter, delivered where your team is already looking, at a volume they will still be reading when it counts most.
