iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Juan Camilo Auriti says 78% of the reports in an audit of his monitoring product had a score delta of exactly zero: each report arrived on time and was technically correct, but showed no change since the previous run. He initially spent a month adjusting detection thresholds. His eventual diagnosis was different: the monitored properties often did not change unless a site owner changed them, so finer detection could not make an empty report useful.
That 78% is Auriti’s account of his own product audit—not an industry-wide rate. His article does not give the number of reports or a full audit method. The useful lesson is to check whether reports deliver anything new before trying to make a detector more sensitive.
Measure what recipients learn, not just whether a score changed
Auriti calls the share of generated artifacts containing at least one thing the recipient did not already know the “content rate.” He reports that his product’s content rate was 22%. That figure, like the 78% zero-delta figure, describes his own product and is not independently validated or representative of monitoring tools generally.
A score delta is a convenient proxy, not a universal definition of value. A report can have an unchanged score yet include a new explanation, context, or useful confirmation. Conversely, a changed score may not tell its recipient anything actionable. Define what counts as new or useful for your product, then measure that against the reports recipients actually receive.
#1 Best Overall
Separate detector sensitivity from how often the signal changes
A threshold controls how much movement a system must detect or report. It cannot create movement in a property that is static. Auriti first treated zero-delta reports as evidence that his thresholds were too coarse; he later concluded that the monitored properties often changed only when a site owner changed them.
Before tuning thresholds, compare successive outputs and ask whether the underlying signal changed at all. If it did, investigate detection and threshold settings. If it did not, making detection more sensitive is unlikely to improve the report’s content. As Auriti puts it, “A report whose content is ‘no change’ is not a report with a sensitivity issue. It’s a report with nothing in it.”
Audit repeated results with a per-domain SQL comparison
Auriti’s example uses PostgreSQL window functions to retrieve the previous score for each domain and count reports whose score matches it. The chronological ordering matters: without it, “previous” would not mean the preceding report. Auriti’s article presents the query as a check to run before tuning.
WITH reports_with_previous AS (
SELECT
domain,
created_at,
score,
LAG(score) OVER (
PARTITION BY domain
ORDER BY created_at
) AS previous_score
FROM reports
)
SELECT
COUNT(*) FILTER (
WHERE score IS NOT DISTINCT FROM previous_score
) AS unchanged_reports,
COUNT(*) AS reports_with_a_previous_report
FROM reports_with_previous
WHERE previous_score IS NOT NULL;This example excludes rows where previous_score is NULL. That means it omits both the first report for each domain and a report whose preceding score was NULL. If NULL is a meaningful score in your data, use a separate way to identify whether a prior row exists rather than filtering on the prior score’s value.
Rank #3
- Business Analytics: Data Analysis and Decision Making with MindTap, 7th Edition
- Product Type: ABIS_BOOK
The equality test also needs care. PostgreSQL documents that ordinary comparisons involving NULL yield NULL, while IS NOT DISTINCT FROM treats two NULL values as equal. PostgreSQL 18’s comparison-operator documentation explains this behavior. The query measures equal scores; it does not establish that a report contained nothing new. Define a content-rate test that fits the fields and information your recipients care about.
Match reporting cadence to the signal’s pace
A weekly schedule is not automatically useful for every monitored signal. If a property changes rarely, a frequent report may repeatedly say the same thing. Consider whether reporting should be periodic, triggered by a meaningful change, or a combination—for example, event-triggered alerts with less frequent status summaries.
Rank #4
- LOOSE LEAF VERSION Still enclosed in shrink wrap. Excellent Saving opportunity. NO CDS supplements of codes are included.
Choose a cadence based on how quickly the monitored information can change and how soon the recipient needs to know. This is a product-design choice, not a fix for a detector that is failing to notice real changes.
Free tools Windows power users keep installed
One-click scans. No signup required.
Add signals that can change without user action
If a product promises ongoing monitoring, signals that move independently of a site owner’s edits may make reports more informative. Auriti suggests examples such as competitor position, citations appearing or disappearing, and new crawler user agents in logs. These are proposed directions from his account, not independently tested recommendations; whether they suit a product depends on its users and purpose.
Best Value
- Wiley
- Language: english
- Book - storytelling with data: a data visualization guide for business professionals
Adding more signals also changes what the product is responsible for explaining. A useful signal mix should connect to decisions recipients can make, rather than simply increasing the volume of data.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Decide whether stability is a result worth reporting
Some monitoring products exist to find problems; others may also promise reassurance that a monitored system remains stable. Those promises produce different expectations for a “no change” result. If stability itself is valuable, state that clearly and make the confirmation informative. If recipients expect discoveries or actions, a stream of unchanged reports may be a sign that cadence or signal selection needs attention.
A practical sequence is to define report value, measure it, check whether the underlying signal changes, and only then adjust thresholds. Auriti says he reversed that order and describes the audit query as a cheap check to make first.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errorsQuick Recap
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.
