180 people in round one, only 14 left in round two — can the crowd carry itself?

What this is

One of our published methods, written in plain language: what it can do, when to use it, what goes in and what comes out.

Why it matters

Most methods are published as papers or code. This one is written so you can decide whether it fits your problem before you invest time in it.

How to use it

Read the page top to bottom. If it matches your question, use the online tool - or subscribe and run it on your own data.

What you get

A clear answer on what the method does and does not do, plus a live curve from our own demand series. Free to read.

Free to readOnline tools from $5 per runSee it →

Demand curve read by this method — A live reading from our own demand series - the shape of the signal this method works on. 570 industries are tracked daily; this is one of them.

Data base: InfoGap own records - 391 Chinese cities x 8 dimensions, 213,208 city-industry rows, 40,499 POIs (central / lifechain, 2026-10-08).

Algorithm Handbook · one algorithm per page · plain language throughout · formal name: Decay ("decay" means "fewer every round")

1In one line: what it does

Enter how many came in round one, how many each person brings, and how much each round drops, and see how many are left round by round.

2When to use it

Three situations call for it:

You spent money on a forum push, 180 people came, and after that nothing happened.

What most people do: "That wave is over, buy another one."

Using it: enter 180 and "each person brings 0.23" and see how many are left in round two — one look tells you whether this wave was one-off.

You want to know "how many people can I reach in total".

What most people do: just add up the headcount of each campaign.

Using it: it adds the rounds together and gives you a total, instead of leaving you to do the sum.

Someone tells you "build a good product and users will bring users".

What most people do: believe it and go back to reworking the product.

Using it: put your two current numbers in and see whether the curve is "down nine tenths per round" or "down half per round" — the two cases call for different fixes.

3What goes in · what comes out

Input (this is the exact set used in the real run)

· How many came in round one: 180
· How many each person brings on average: 0.23
· How much each round drops: 0.35 (a drop of about 35%)

What comes out

These numbers are what it produced on the spot, not a single digit changed:

Round 1: 180 people (cumulative 180)
Round 2: 14.35 people (cumulative 194.35)
Round 3: 1.13 people (cumulative 195.48)
Round 4: 0.09 people (cumulative 195.57)
Total people reached: 195
Its verdict: decay is significant — only 0.1% left in the final round

—— with "each person brings 2.5, each round drops only 0.05" ——
Round 2: 22.28 people, round 3: 2.73 people, round 4: 0.33 people
Total people reached: 205

How to read it: look at how much drops between two rounds. Here round one to round two drops 92%, which means almost none of those 180 people brought anyone in for it. The total is only 195, just 15 more than round one's 180 — so this wave was basically one-off. The other case (the second set of numbers) is declining too, but the total reaches 205, so people-to-people spread is happening, just slowly.

📈One chart: round by round, how many are left

180 people in round one, only 14 left in round two — where exactly does the cliff fall.

4How to use it

There is no public calculator for this yet — the method above is the whole procedure, and on this site the inputs have to be keyed in by hand. The inputs are three numbers: round-one headcount, average invites per person, and how much each round drops.

Measure three numbers first: how many came in round one, how many each person brings on average, and roughly how much each round drops
Work the three numbers through the steps above and look at the round-by-round headcount
Read the per-round headcounts and the "total people reached"

5Who it's for

People who ran an acquisition push and want to review whether it was worth it; people who want to know whether people-to-people spread actually happened; people who want to run the numbers before spending on the next wave.

6Common questions

Q: How is this different from the "how many people does one person bring" one?

A: That one gives a single number (how many people one person brings on average); this one lays out the round-by-round headcounts and the total for you. Using both double-counts, so pick one.

Q: Why does round two drop so much?

A: Because "each person brings 0.23" is itself below 1, and a share of each round is lost on top of that, so round two naturally leaves only a dozen or so.

Q: Does a total of 195 mean this wave was money down the drain?

A: Not entirely. It means the spend bought 180 real people who actually clicked in — it is just that those people will not bring others. Paying to acquire people is not wrong; just don't expect it to grow on its own.

7Related algorithms

k-factor · one number to see whether it can spread by itself water-flow · turn campaign data into a curve risk-grade · how dangerous is it right now ← Back to the Algorithm Handbook Handbook index

✓Why this scenario exists (where the content comes from)

This scenario comes from the records of the "algorithm meets demand" round: the overseas Chinese community cold-start demand was matched against several algorithms, and this one was judged "useful", on the grounds that it gives "with the same inputs, results consistent with the first few rounds of the water-flow algorithm, and it outputs 'decay is significant, only 0.1% left in the final round'".

The input and output above were produced by re-running the numbers recorded there: 180 people / 0.23 invites each / 0.35 drop per round → per round 180 / 14.35 / 1.13 / 0.09, total 195.

Source: section 5, product 4 of the overseas scouting report; the report's measured wording is "$700 → 77,000 impressions, 180 clicks, 0 signups". The D7 group in the pairing ledger.