Quick Ratio

While growth was described by a single growth rate, two very different companies got written down in the same sentence. Users up 10 percent month over month, and the description was finished.

The limit of that shows up in the question Grace got back after telling an investor the company was growing 10 percent a month: write the arrivals and the departures separately. One company goes from 30,000 users to 33,000 by adding 4,000 and losing 1,000. Another goes from 30,000 to 33,000 by adding 12,000 and losing 9,000. Both lines in the deck read 10 percent. To hold that same 10 percent next month, the first company needs to find 4,400 people and the second needs to find 13,000.

The quick ratio exists to pull that difference into one number. Divide what came in by what went out and the first company scores a 4 while the second scores 1.3, so a single cell shows that the same growth rate is sitting on top of two different structures.

Divide what was gained in a period by what was lost in the same period. The unit is either users or revenue, and a team picks one and stays with it.

The user version starts by splitting a month's change in user count into four buckets. Writing those buckets out is growth accounting, and the quick ratio compresses that table into one division.

1. New — people who arrived for the first time this month.
2. Resurrected — a resurrected user who went quiet for a while and came back.
3. Churned — people who used the product last month and not once this month.
4. Retained — people who used it in both months. This bucket stays out of the formula.

The formula adds new and resurrected, then divides by churned. In a month with 30,000 monthly active users (MAU), 4,000 new, 900 resurrected and 4,900 churned, the ratio is exactly 1.0 — net change of zero. Drop churn to 1,400 and the same month scores 3.5.

The revenue version runs the same formula in dollars. The numerator holds revenue from new contracts plus revenue added by existing customers spending more; the denominator holds revenue lost to cancellations plus revenue lost to customers moving down a plan. The unit is monthly recurring revenue (MRR), the portion that arrives again every month. With $9,000 new, $2,300 in expansion, $3,000 cancelled and $800 in downgrades, the ratio comes to 3.0.

The part of the calculation that trips teams up is the definition of churn. A team that counts someone as churned after 30 quiet days and a team that waits 60 days get different numbers from identical data, because the shorter window puts more people in the denominator. Once the window is set it stays set, and changing it means recomputing several past months on the new rule before the series can be read as one line.

One is the dividing line. Gains equal losses, so user count sits still, and anything under one shrinks every month.

For early-stage software products, a number around 4 is the figure people describe as healthy: four gained for every one lost. Below 1.5 is treated as a warning even when the growth rate reads well, because most of what comes in is being spent replacing what left, and a small rise in acquisition cost takes net growth to zero.

The trend matters before the absolute value. A ratio that fell from 3.1 to 1.4 in two months is worse news than one sitting at 2.0 for four months straight, because whatever pushed it down is still in place. When it moves, teams reach for something like north star decomposition, which breaks a headline metric into the factors that multiply into it, to find which factor moved.

A small denominator makes the number unreliable. A product with 300 users, 30 new and 3 churned scores a 10, and churn rising to 6 next month halves it to 5. Below a few thousand users, a three-month average reads better than any single month.

A product that holds near 4 for a long stretch is usually one that has found product-market fit — the state where what was built matches real demand in a specific market. Products in that state keep the people who arrive, which shrinks the denominator on its own. A ratio parked near 1 for a year points instead at weak retention, the rate at which existing users keep coming back, and that is the place to look before buying more traffic.

"Growth rate looks fine, but quick ratio is 1.2. Arrivals are covering departures."

Net growth is real, and the structure underneath it is leaking.

"Quick ratio is 3.4 this month against 3.1 last month, so the trend holds."

The direction is the report, not the number itself.

"We moved the churn window to 30 days, so recompute the last six months and send both."

A changed definition can't be placed next to older values.

At Pagedeck, Ravi puts this number on the first slide of the monthly metrics review. Last month came to 4,200 new, 800 resurrected and 3,100 churned, a ratio of 1.6, down from 2.9 three months earlier.

Grace asked whether the denominator had grown or the numerator had shrunk before anyone mentioned the ad budget. If new signups look about the same as three months ago while churn climbed from 1,700 to 3,100, the next question is which path the leaving users arrived through. When teams from ads stay half as long as teams from search, the thing to fix is the gap between what the ads promise and what the product does, not the size of the budget.

The most common misreading is that a high number means fast growth. Big numbers feel good, so the assumption comes easily. But this ratio measures leakage, not speed. A product with 100 new users and 10 churned scores a 10 and grows by 90 people a month; a product with 10,000 new and 4,000 churned scores 2.5 and grows by 6,000. The second one is twenty times faster. Cutting a channel on the strength of this number alone shrinks the leak and stops the growth along with it.

The other one is reading a falling ratio as rising churn. A growing denominator is the intuitive cause. A shrinking numerator does the same thing, though. Hold churn flat at 3,000 while new signups fall from 9,000 to 4,500 and the ratio halves from 3 to 1.5. Assign people to churn work without checking, and two months later the number has not budged, because the blockage was on the acquisition side.