THE SHELFSPRING FIELD NOTES · 07 NOV 2026
Your watched stores,
timed.
A watch should keep looking after you close the tab. We inspected the recorded checks for ten stores in ShelfSpring’s eligible monitoring scope. The result: a 5.993-second median gap, measured against our five-second target.
22:00–22:30 UTC
November 7, 2026
A WATCH-SCOPE RECEIPT
Five seconds is the aim.
These are the observed times.
The typical recorded gap was slightly under six seconds. We are publishing that difference instead of rounding the measurement down to five.
- Recorded store checks
- 2,959
- Anonymous store series
- 10
- Median check interval
- 5.993 s
- 95th-percentile interval
- 6.793 s
- Longest observed interval
- 9.180 s
- Unsuccessful checks in sample
- 0
FOLLOW ONE SERIES
What did repeated checking actually look like?
The timestamps below are consecutive, not selected for being fast. They show the first eight entries of the first anonymous series in the download. UTC times refer to completed store checks on November 7, 2026. This is a dated observation, not a live demonstration.
- 01
First completion inside the sample window · successful
- 02
6.144 seconds after the preceding check · successful
- 03
6.042 seconds after the preceding check · successful
- 04
5.842 seconds after the preceding check · successful
- 05
6.055 seconds after the preceding check · successful
- 06
5.671 seconds after the preceding check · successful
- 07
6.085 seconds after the preceding check · successful
- 08
5.745 seconds after the preceding check · successful
Across all ten series, 100% of the 2,949 measured gaps were within ten seconds. Only 0.1% were within five seconds. This is evidence of repeated seconds-apart checking, not evidence of a strict five-second maximum.
| Gap length | Count | Share |
|---|---|---|
| ≤ 5 seconds | 3 | 0.1% |
| > 5–6 seconds | 1,518 | 51.48% |
| > 6–7 seconds | 1,348 | 45.71% |
| > 7–10 seconds | 80 | 2.71% |
| > 10 seconds | 0 | 0% |
FROM A NUMBER TO A DECISION
What can a shopper take from this?
A saved watch concerns a store and a phone configuration. Frequent store checks are useful context, but these records do not certify that every possible color or capacity was included in each check.
The sample shows repeated checks over the full half-hour. It does not measure the performance of a particular person’s account, browser or phone, and it cannot promise that a future restock will last long enough to buy.
Monitoring cannot reserve an iPhone. Treat an alert as a reason to verify Apple’s current checkout and pickup options, not as confirmation that an order is ready.
Where does the five-second claim stop?
Five seconds describes the intended monitoring interval for eligible active member targets. The measured completion spacing in this report is 5.993 seconds at the median. Neither number is a promise that a notification appears within five seconds of new stock becoming available.
A useful alert has to reach a device as well as detect a change. This study does not time that journey. It also does not compare ShelfSpring with another tracker or calculate a buyer’s chance of securing stock.
OPEN EVIDENCE, PRIVATE OPERATIONS
Read the measurement without seeing anyone’s watch list.
The ten stores were in ShelfSpring’s eligible-target scope when the report was extracted. We included their logged priority checks during 22:00–22:30 UTC. We did not reconstruct membership or selection changes throughout that half-hour. This is a store-level subset of shared inventory observations, not an independent collection experiment.
Each row represents one recorded store-check completion. The CSV uses anonymous labels and contains no account identifiers, actual store selections, device details or implementation information. A series label is meaningful only inside this export.
To reproduce the interval calculation, sort each series by completion time and subtract consecutive timestamps. The initial row in each of ten series has no previous in-window check: 2,959 completions therefore yield 2,949 gaps. All logged results remain included. The 95th percentile is the nearest-rank value, not an average of the slowest checks.
We observed no unsuccessful checks in this subset, but that is not a promise of future reliability. Missing or unlogged activity cannot be recovered from this export. This is our own dated observation, not an independent audit or an all-day service-level guarantee.
Identify this exact file
SHA-256: aba9871e673067b6f937472bc752149f6f1f777b721981b1c0a2fed7b381f1a8
The hash identifies the downloaded bytes; it does not independently verify the measurement.
Three things worth asking
Does a free saved selection receive the same monitoring?
A saved selection and an eligible active paid watch are different states. This report concerns the eligible-target store scope at extraction time; it does not grant or verify monitoring for a particular account. Check your plan and active rules in your account.
Can I use 5.993 seconds as a future guarantee?
No. It describes one fixed half-hour. Keep its date and scope attached when quoting it. A new measurement is needed before making a claim about a later period.
Why are the stores anonymous?
The measurement should demonstrate timing without publishing customer watch selections. Anonymous series preserve consecutive timing while leaving those selections private.
Review your watches · Set up notifications · Send a measurement question
Cite: ShelfSpring, “Your watched stores, timed,” November 7, 2026, 22:00–22:30 UTC. Measured store-check median: 5.993 s; target: 5 s; notification latency not measured.