Juan Camilo Auriti audited reports from his monitoring product and found that 78% showed a score delta of exactly zero: nothing had changed since the previous run. The reports were correct and on time, but empty of new information. His first response was to tune change-detection thresholds. He later concluded that the deeper problem was the content and cadence of the reports, not the detector’s sensitivity.
What the 78% and 22% figures mean
In his account, Auriti found that 78% of reports had no score change. He also defined a measure he called the “content rate”: the share of generated artifacts containing at least one thing the recipient did not already know. He reported a content rate of 22% for his product.
As an Amazon Associate I earn from qualifying purchases.
These are figures from Auriti’s self-reported product audit, not industry benchmarks. The available account does not state the number of reports audited or provide independent validation. A zero score delta is also only a proxy for an empty report: two equal scores do not prove that every part of an artifact lacked new information. Teams should define value in terms of what their own recipients learn or can act on.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWhy changing thresholds was the wrong first move
Auriti initially suspected that his change-detection thresholds were too coarse and spent time tuning them. But detector sensitivity and the frequency of real-world change are different questions. A detector can be perfectly sensitive to a signal that rarely changes.
#1 Best Overall
As Auriti later put it, “A report whose content is "no change" is not a report with a sensitivity issue. It’s a report with nothing in it.” If the monitored properties remain static until a site owner edits them, making the detector more sensitive cannot create new developments to report. It may instead surface smaller differences without improving the report’s usefulness.
Check report value before tuning detection
A useful first diagnostic is to measure how often an artifact contains information that is new to its recipient. Auriti’s score-delta comparison was a quick way to find repeated reports in his system, but equal scores are not a universal definition of “nothing new.” A product team might instead track changed fields, newly actionable findings, or recipient-confirmed value, depending on what the report promises.
For a score-based system in PostgreSQL, the core comparison can be written as:
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →WITH ranked AS (
SELECT
domain,
created_at,
score,
LAG(score) OVER (
PARTITION BY domain
ORDER BY created_at
) AS previous_score
FROM reports
)
SELECT COUNT(*) AS unchanged_pairs
FROM ranked
WHERE previous_score IS NOT NULL
AND score IS NOT DISTINCT FROM previous_score;LAG(score) retrieves the score from the preceding row within the same domain, with rows ordered by created_at. The partition matters: without it, one domain’s report could be compared with another domain’s. The ordering matters too, because it determines which report counts as the previous one.
Rank #3
- Business Analytics: Data Analysis and Decision Making with MindTap, 7th Edition
- Product Type: ABIS_BOOK
PostgreSQL documents that ordinary comparisons involving NULL yield NULL, while IS NOT DISTINCT FROM treats two NULL values as equal. That makes the operator useful when NULL scores should count as unchanged relative to other NULL scores. The example filters out rows without a prior score; adapt that condition if your definition of an unchanged pair treats missing values differently. See the PostgreSQL 18 comparison operators documentation.
This query counts equal score pairs, not necessarily empty reports. It is a diagnostic for a specific signal, and teams should choose a content-value measure that matches the artifact and what its recipient already knows. Auriti described the check as “one query, before the tuning, not after.”
Rank #4
- LOOSE LEAF VERSION Still enclosed in shrink wrap. Excellent Saving opportunity. NO CDS supplements of codes are included.
Match reporting cadence to how quickly signals move
A default weekly schedule can produce a steady stream of reports even when the underlying signal changes only occasionally. Auriti’s proposed response was to match cadence to the rate of change rather than assume that frequent delivery is inherently useful.
For some products, an event-triggered report may make more sense than a periodic one: notify the recipient when a meaningful change occurs, rather than send a report on a fixed schedule regardless of whether anything happened. Periodic reporting may still suit a product whose promise includes regular confirmation that conditions remain stable. The right choice depends on the recipient’s need and the behavior of the monitored signal; Auriti presents these as product choices, not as independently tested rules.
Best Value
- Wiley
- Language: english
- Book - storytelling with data: a data visualization guide for business professionals
Expand the signal mix when monitoring should find independent change
If a product promises ongoing monitoring, consider whether it watches only properties controlled by the user. Auriti suggested adding signals that can change without the site owner making an edit, including competitor position, citations appearing or disappearing, and new crawler user agents in logs.
These examples broaden what a report can detect, but they do not guarantee a higher content rate. Each signal needs to be relevant to the recipient, measured consistently, and presented in a way that distinguishes meaningful change from noise.
Decide what a no-change report is meant to promise
Stability can itself be useful information. A report that confirms a monitored condition remains healthy may fulfill a product promise even when its score is unchanged. But if recipients expect discoveries or action items, repeated no-change reports can feel empty.
The product decision is whether to position monitoring as finding problems, confirming stability, or doing both. That choice should shape what counts as a valuable report, how often it arrives, and which signals it includes. A detector threshold is only one part of that design.
Quick Recap
Sources
- Juan Camilo Auriti, “78% of my reports had nothing to say, and I spent a month tuning the wrong thing”, DEV Community.
- PostgreSQL Global Development Group, PostgreSQL 18: Comparison Functions and Operators.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
