A post takes off. The traffic graph spikes. You reload it five times, grinning — "see, social works, let's post way more from tomorrow." That decision just got made by the height of a spike.
The night you go viral is exactly when one screen matters: RPS by channel (revenue per session).
Volume spikes and money-carrying traffic are different things#
From the sample store's real 30-day channel data, four contrasting rows:
| Channel | Sessions | Revenue | RPS |
|---|---|---|---|
| Google Search | 278 | ¥108,558 | ¥390 |
| Direct | 172 | ¥100,437 | ¥583 |
| Gemini | 24 | ¥33,188 | ¥1,382 |
| X | 27 | ¥0 | ¥0 |
Twenty-seven sessions from X earned ¥0, with 33 seconds of average dwell — visitors came, but they weren't shoppers. Meanwhile Gemini brought just 24 sessions worth ¥33,188 — ¥1,382 per visit. The same "one session" differs in value by orders of magnitude depending on the door it came through.
That's the viral night in miniature. A pageview spike is a volume event; revenue is a quality-times-volume event. Let the spike's height drive your investment call, and the high-quality doors (here, AI referrals and search) get deprioritized. The three-layer way to read traffic quality is detailed in this article.
Three steps for the viral night#
1. Celebrate first. Going viral is genuinely great. 2. Open RPS by channel. Did today's spike move the revenue row? Check the bot-excluded real numbers (crawlers ride along with virality). 3. Decide in the morning. If the spike carried revenue, that door deserves more. If not, log it as an awareness event and leave your investment plan unchanged — that alone prevents most high-emotion misjudgments.
RevenueScope shows you quality on the night of the spike
Sessions, revenue, RPS, dwell and bounce per channel on one screen, bots excluded. Learning that X earned ¥0 stings — but it's far cheaper tuition than shifting your budget toward social without knowing.
One boundary: social and AI referrals often convert days later, so under default last-touch attribution the entry channel may not get credited. You can switch attribution models and compare.
FAQ#
Q. Does this mean social doesn't sell?
No. It means X earned ¥0 in this sample's 30 days — that's all. For some sites social is the strongest door there is. The point is deciding from your own measured RPS, not from anyone's generalization.
Q. Can I tell whether a spike was mostly bots?
Each channel row carries its bot-excluded count. "It spiked, but most of it was crawlers" is visible on the same screen.
See which ads actually drive revenue, at a glance
Free up to 5,000 sessions/month, AI analyst included. No credit card required. Up and running in 5 minutes.
