Twitch Channel Point Pricing: The Math Most Streamers Get Wrong

You set your reward prices when you first hit affiliate. Maybe you looked at what other streamers charge, maybe you just picked round numbers. Either way, it was a guess.

Since then you’ve been adjusting whenever something goes wrong. A sound effect gets spammed during your horror stream, so you raise the price. Nobody redeems the expensive one, so you lower it. A loyal viewer who’s been with you since day one is sitting on 15,000 points and wants something big to spend them on, but nothing you’ve set up feels worth it.

You’re reacting to problems you can’t see coming because there’s nothing telling you what’s actually happening with your rewards. The prices feel arbitrary because they are. They were set once and they’ve been drifting ever since.

Most of this comes down to a piece of math that Twitch doesn’t make obvious, a framework for thinking about different types of rewards, and a concept called cost scaling that you’re probably already doing by hand without realizing it.

The earn rate math most streamers never learn

Before you can price anything, you need to know what your viewers are working with.

Twitch awards channel points on a fixed schedule that hasn’t changed since Channel Points launched in 2019. Viewers earn 10 points every 5 minutes of watch time, plus a 50-point bonus every 15 minutes that they need to click to claim. That works out to roughly 320 points per hour for a non-sub viewer who’s actively watching and clicking the bonus.

Subscribers earn more. Tier 1 subs get a 1.2x multiplier on the base watch rate, Tier 2 gets 1.4x, and Tier 3 gets up to 2x. The click-to-claim bonus stays the same regardless of sub tier. Watch streaks add a bonus of 300 to 450 points depending on how many consecutive streams someone has caught. Participating in a raid adds 250.

Put that into something useful: a non-sub viewer watching your 3-hour stream earns roughly 960 points. A Tier 1 sub earns about 1,030. That’s the budget your viewers are working with per session.

The per-session budget: ~320 points/hr for non-subs, ~345 points/hr for Tier 1 subs. Over a 3-hour stream, that’s roughly 960 to 1,030 points. Every price you set is a fraction of that budget.

Knowing the earn rate changes how you think about every price you set. A reward that costs 500 points sounds cheap until you realize a new viewer needs to watch for about 90 minutes before they can afford it.

A pricing framework by reward category

Not all rewards should cost the same, and the reason isn’t just “some are better.” Different rewards serve different purposes, and the pricing logic for each one is different.

Quick-hit rewards are things like sound effects, light changes, or a hydrate reminder. These should be cheap enough that any viewer can use them within their first stream. Price them relative to the hourly earn rate. If a viewer earns 320 points per hour and your sound effect costs 100, they can trigger it roughly three times per hour. That feels about right for ambient interaction.

Interactive rewards are things that change how you play or what you do. Remove your glasses, play with an upside-down controller, wear a ridiculous hat for the next 10 minutes. These should cost more because they interrupt your gameplay and they’re funnier when they’re less frequent. Mid-range pricing works here, somewhere that requires 30 to 60 minutes of watch time. Consider limiting them per stream so they don’t stack up.

Aspirational rewards are the expensive ones. VIP status, choose the next game, extend the stream by an hour. These should cost enough that earning them feels like an achievement. They give long-term viewers something to save toward and they create a sense of progression in your channel’s economy. Price them at three to five sessions of watch time. Something in the range of 5,000 to 10,000 points gives viewers a real goal to work toward.

Pricing at a glance: Quick-hit rewards: under 30 min of watch time (~150 points). Interactive rewards: 30 to 60 min (~150 to 320 points). Aspirational rewards: 3 to 5 sessions (~5,000 to 10,000 points).

The framework is simple: think about how often you want each reward to fire, then work backwards from the earn rate to set the price. If your quick-hit reward costs 500 points, a new viewer needs to watch for about 90 minutes before they can use it once. Is that the behavior you want for a sound effect? Probably not.

Why static pricing breaks over time

Even if you nail the initial prices, they won’t stay right forever.

Game designers who build virtual economies talk about faucets and sinks. A faucet is anything that puts currency into the system. A sink is anything that takes it out. In your channel, the faucet is watch time, pouring roughly 320 points per hour into every viewer’s balance at a rate you can’t control. Your rewards are the sinks. The only thing keeping the economy balanced is whether your sinks are compelling enough to drain points at roughly the same rate the faucet fills them.

When that balance tips, you get one of three problems. Most streamers run into at least one.

The first is spam. Cheap rewards get hammered. If you stream horror games and have a jump-scare sound effect at 50 points, your regulars will trigger it constantly. One streamer described having to increase the price on their horror sound effects to stop the spam. In a thread with 35 comments, others described deleting rewards entirely, tweaking others, raising prices, and adding per-stream limits. By the time you notice the problem, your chat has already turned your carefully set-up reward into a meme.

The second is hoarding. Loyal viewers who watch every stream build up points faster than they can spend them, especially if your rewards are all priced for casual viewers. One streamer described a regular who’d been sitting on a massive point balance and kept asking for something worth spending on. Another said their viewers end up hoarding huge amounts of points because there aren’t enough high-value redeems that still feel exciting. The economy stalls when your most engaged viewers have nothing to reach for.

The third is staleness. Rewards that were exciting at launch become furniture. Nobody redeems them, but they still take up a slot. You built that reward because you thought it would be fun. Now it’s just taking up space, and you’ve forgotten it’s there. You might not notice for weeks because Twitch doesn’t surface this information. The reward just quietly sits there, eating one of your 50 slots, doing nothing.

The problem isn’t that you set bad prices. It’s that prices need to respond to how people actually use your rewards, and a number you typed in once can’t do that.

Cost scaling: prices that adjust themselves

If you’ve ever raised a reward’s price because it was getting spammed, you’ve already done cost scaling by hand. The idea is simple: instead of setting a fixed price and adjusting it when something breaks, you set a price that moves with usage.

A reward starts at a base price. Every time someone redeems it, the price goes up. This slows down spam because the first few redemptions are cheap but it gets more expensive each time. Then the price resets at some interval, like the end of the stream or the start of a new day, so it’s fresh again next session.

There are two common approaches.

Linear scaling increases the price by a fixed amount each time. A reward that starts at 100 points goes to 200 after the first redemption, then 300, then 400. It’s predictable and gentle. Viewers can do the math and decide if the next redemption is worth it.

Geometric scaling multiplies the price each time. That same 100-point reward goes to 200, then 400, then 800. It ramps up fast, which makes it a natural rate limiter. The first couple of redemptions are accessible. After that, only viewers with deep point balances are buying.

Linear vs geometric: Linear adds a fixed amount per redemption (100 > 200 > 300 > 400). Gentle, predictable. Geometric multiplies (100 > 200 > 400 > 800). Aggressive, self-limiting. Pick based on how rare you want the reward to be.

To make this concrete: say you have a jump-scare sound effect on your horror stream priced at 100 points. With linear scaling, the first viewer pays 100, the next pays 200, then 300. By the fourth trigger, your chat has to decide if that jump scare is worth 400 points. The spam slows down naturally, but it’s still accessible. With geometric scaling, that same reward goes 100, 200, 400, 800. Three redemptions in and it costs more than an hour of watch time. The spam dies fast.

The choice depends on the reward. Linear works well for interactive rewards where you want steady usage that gradually slows. Geometric works for things that should be rare by nature, like choosing the next game or spawning a zombie horde in 7 Days to Die.

You can do this manually. After each stream, look at which rewards got hammered and raise the price. Look at which ones nobody touched and lower it. But that only works if you remember to do it and if you have some way to see the usage numbers.

Some streaming tools can automate this. Streamer.bot has a Set Cost sub-action with math operators that lets you wire up a reward redemption trigger to automatically increase the price by a fixed amount or multiply it. Firebot can do something similar with its Update Channel Reward effect and custom scripts for more complex formulas. Mix It Up has an Update Channel Point Reward action that can modify cost on redemption. None of them offer this as a one-click feature. You’re building the logic yourself through triggers and actions.

If you’re comfortable with that kind of setup, any of those tools will work. If you want cost scaling as a built-in feature where you just pick linear or geometric and set your parameters, BetterStreams ships it natively.

Either way, the concept is what matters. Prices that respond to demand solve the spam problem, create natural scarcity, and keep your reward economy from going stale. Do it by hand, wire it up in an automation tool, or use something built for it. The result is the same.

How to know if your pricing is working

Setting prices is only half the job. The other half is knowing whether they’re actually working. After your next few streams, there are four signals worth paying attention to.

Redemption frequency per reward. If one reward accounts for 50 or 60 percent of all your redemptions, it’s probably too cheap relative to the others. You want usage spread across your reward list, not stacked in one or two cheap options.

Unique redeemers. Fifty redemptions sounds healthy, but it matters who’s redeeming. Fifty redemptions from two viewers means two people are camping a cheap reward. Fifty from thirty different viewers means the reward is appealing and fairly priced. The same number tells two very different stories.

Redemptions over time during a stream. If there’s a spike in the first 30 minutes and then nothing, viewers are burning their points too fast on cheap rewards and have nothing left for the rest of the stream. If redemptions stay relatively flat throughout, your pricing is giving people enough to engage with across the whole session.

Dead rewards. Zero redemptions across multiple streams. Something is wrong. Either the price is too high relative to what the reward offers, the reward itself isn’t interesting, or viewers don’t understand what it does. Reprice it, rework it, or remove it and free up the slot.

The hard part is actually getting this data. Twitch doesn’t give you a dashboard that breaks down redemptions by reward, by viewer, or over time. The Twitch dashboard shows you which rewards exist and lets you manage the queue, but it doesn’t tell you which rewards are popular, which ones are dead, or who’s redeeming what.

None of the major streaming tools solve this either. Streamer.bot, Firebot, CastMate, and Mix It Up are all automation tools at their core. They trigger actions when a reward gets redeemed, but they don’t track or visualize redemption data. Mix It Up has a statistics page that tracks stream-level metrics, but it doesn’t break things down per reward.

You can track this manually. Keep a spreadsheet, note which rewards fire during your streams, and look for patterns over a few sessions. It’s tedious but it works if you’re disciplined about it.

BetterStreams tracks redemption analytics natively, including per-reward popularity, unique redeemers, and usage over time. If you want the data without the spreadsheet, it’s currently the only tool that provides it.

Regardless of how you track it, the signals are what matter. Learn to read them and your pricing decisions stop being guesses.

What to do next

You don’t need to overhaul your entire reward setup in one sitting. Start with the piece that makes the biggest difference and build from there.

1. Learn the earn rate math. Calculate what a typical viewer earns during one of your streams. Non-subs earn roughly 320 points per hour. Subs earn more. Knowing this number turns every pricing decision from a guess into a calculation.

2. Categorize your rewards. Look at each reward and decide if it’s quick-hit, interactive, or aspirational. Check whether the price makes sense for that category. A quick-hit reward that costs two hours of watch time is priced like an interactive one.

3. Identify candidates for cost scaling. Anything you’ve had to manually reprice because of spam or underuse is a candidate. You can do this by hand after each stream, wire it up in Streamer.bot or Firebot, or use a tool like BetterStreams that handles it natively.

4. After your next few streams, check which rewards are working. Look for the four signals: redemption frequency, unique redeemers, usage patterns over time, and dead rewards. Even rough tracking in a spreadsheet is better than guessing.

The goal isn’t a perfect economy on day one. It’s building the habit of checking what’s happening and adjusting based on what you find instead of waiting until something breaks.