How to verify a streamer's audience and spot fake viewers

To verify a streamer's audience before you buy, work through five steps: see the creator's own dashboard, check it against public history, treat every gap as a question, pay per verified view rather than promised reach, and test small if doubts remain. No single step proves that a channel does or doesn't use bots.

Definitions and warning signs are in Twitch viewbots: how to spot fake viewers. This one walks through how to check an offer like “5,000 viewers per stream”.

The audience check at a glance

Step 1: Get the analytics first-hand

Ask the streamer to open their dashboard live on a screen share; a screenshot can be cropped. Here is what each platform shows the creator (checked 12.11.2026):

Twitch's Creator Camp advises creators to analyze longer periods; the three-month minimum is our recommendation, not a Twitch requirement.

Step 2: Check public data without channel access

Without access you see the channel page, saved past streams, the schedule and history on third-party trackers, which work from public numbers. Use trackers for two questions: does the order of magnitude match the offer, and do the time slots you'd buy perform like the monthly average? An odd curve is a reason to ask, not proof.

Step 3: How to spot fake viewers on Twitch without jumping to conclusions

None of the platform help pages cited here gives a threshold that proves bots, so compare within the same channel.

• Average vs peak. Kick shows them separately; YouTube Help defines “concurrent viewers” as a maximum. Pin down which number the offer quotes. More on the terms in CCV, average and unique viewers.
• Viewers vs chat. Set Unique Chatters or chat messages against the average, stream by stream.
• Jumps. Ask what caused each spike: a raid, a collaboration, a tournament, an embed.

Can you buy Twitch viewers, and what happens?

If you pay per stream based on viewers, you pay for bought viewers too. For the channel, platform rules apply (checked 12.11.2026):

These are platform rules, not a legal assessment. A channel capped or suspended mid-campaign takes the promised reach with it.

Step 4: What to write into the contract and media plan

• Billing unit: per verified view, not per promised viewer count or a flat fee per stream.
• Proof: a screenshot captured at the moment of display, with channel, time and audience size, not sent by the streamer afterwards.
• Anti-fraud consequences: what happens to suspicious traffic, and is it removed from the report and the invoice?
• Right to exclude: approve every channel before launch and be able to exclude one during the campaign.
• Your own tracking: a separate link per channel, UTM parameters or your MMP.

Step 5: Still unsure? Run a pilot

Book a small pilot under the same rules as the full campaign. It gives you your own numbers instead of promises, but doesn't guarantee the full campaign's result.

Checking the audience and checking the ad are two jobs

The check before you buy asks whether this audience exists; the check during the campaign asks whether your creative was fully visible, on the right channel, at the agreed time. For the second, see How to check that your ad actually ran and Brand safety in live streams.

How APIQ limits this risk

APIQ bills verified views at one CPV rate with a guaranteed volume, and checks each one:

• A screenshot of the stream at the moment of display; the viewer count is recorded at that moment.
• Computer vision checks that the creative is present, intact and in the right position.
• Anti-fraud across 14 parameters: chat activity, audience dynamics and stability, sudden viewer spikes, user behavior, clicks and conversions. The full list is available on request.
• Suspicious traffic is excluded from the report and the invoice; a streamer with bots is removed from APIQ.

This doesn't replace the platforms' detection or promise zero bots. The media plan arrives within 24 hours, and the launch follows 1–2 days after the streamer and creative list is approved. More on the brands page.