SHYPIARD
OBSERVATORY · TRACKING CLUSTER

A monthly report always arrives too late.

In the Shypiard observatory, 38% of struggling launches are a post-launch velocity problem — a drop that, caught in time, gets fixed in a few days. Caught a month later, it's already spread through the algorithm.

01 — WHY WEEKLY, NOT MONTHLY

The algorithm measures in days. Your tracking should too.

Steam evaluates velocity over rolling windows of a few days to two weeks. A monthly report catches a problem after it's already affected three or four algorithmic measurement cycles — the time to react was already lost before the report even lands in the inbox.

Weekly tracking isn't about excessive rigor: it's simply matching the pace at which the system you're trying to influence makes its own decisions.

02 — THE FOUR NUMBERS, IN DETAIL

Each one answers a different question.

Wishlist velocity(7-day delta, not the cumulative total) answers "is it still climbing." 30-day rolling positive review ratio(not the displayed overall rating) answers "is recent reception declining." 7-day rolling CCU answers "are the players who buy sticking around." Mentions outside your community(Reddit, third-party Discords) answer "is a conversation starting without you."

None of these four numbers alone tells the whole story — it's reading them together that distinguishes a normal slowdown from a real warning sign.

03 — THE CONCRETE ALERT THRESHOLD

Two flat weeks, not one.

One week of flat wishlist velocity can be noise — a quiet weekend, a competing piece of news. Two consecutive weeks with no movement is a signal: it's time to publish the next visibility event rather than wait for a third flat week that would only confirm what you already know.

04 — WHEN TO REACT, WHEN TO WAIT

Not every drop is treated the same way.

A velocity drop with a stable positive-review ratio is a reach problem — publish content or re-engage creators. A velocity drop paired with a falling review ratio is a reception problem — a patch or a page clarification matters more than a new visibility push that would only amplify an unresolved issue.

RESEARCH · 17 AUGUST 2026

Post-launch shows up on the store page

DLC: 75/100 versus 16/100. Updates: 99/100 versus 81/100 (5 documented silences on the low-review side). News without traffic is still zero velocity — Diets & Deities has 19 announcements and 15 reviews.

Read the 200-game study →

05 — WHAT TRACKING DOESN'T REPLACE

Tracking these four numbers reveals that a problem exists and what kind — it doesn't by itself provide the action plan or the creators to re-contact. The free diagnostic sets up this tracking along with the plan that goes with it.

YOUR CASE

We monitor these four numbers
every week, not every month.

Test your page first in 90 seconds — post-launch tracking then includes weekly monitoring and continuous outreach, $100/month.