{"id":10692,"date":"2026-06-09T12:46:19","date_gmt":"2026-06-09T10:46:19","guid":{"rendered":"https:\/\/mio.one\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/"},"modified":"2026-07-09T10:18:15","modified_gmt":"2026-07-09T08:18:15","slug":"everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media","status":"publish","type":"post","link":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/","title":{"rendered":"“Everything’s normal” isn’t the same as “everything’s fine”: the blind spot in anomaly detection in media"},"content":{"rendered":"\n

Estimated reading time: 8 minutes<\/strong><\/sub><\/p>\n

\n
\"\"
Alex Masip
Data & Martech Director
MIO One<\/figcaption><\/figure>\n<\/div>\n

The Blind Spot in Anomaly Detection in Media Data<\/h2>\n\n

We have an anomaly detector monitoring the data. The day our data warehouse stopped <\/em>reflecting reality, it gave the green light. Why the most costly performance failure is also the quietest, and why defining \u201cnormal\u201d is harder than it seems. <\/p>\n\n

I\u2019ve always liked articles with a practical slant\u2014the kind where you can take away knowledge that\u2019s applicable to your own situation. Those who know me and see me ranting on LinkedIn from time to time know that I\u2019ve been particularly persistent lately about the importance of data quality. It\u2019s always been that way, but with AI, it\u2019s even more so, if that\u2019s possible. <\/p>\n\n

I try to lead by example, and as the person ultimately responsible for data quality and its use in AI and other projects, it\u2019s my day-to-day obsession. We have all kinds of controls, tests, monitoring agents, and so on. And you\u2019d like to think that everything is running like clockwork. Obviously, there are always issues\u2026 This is the story of one of those issues. <\/p>\n\n

1. The Cliff That No One Saw<\/h3>\n\n

Suddenly, a performance channel for a lead-generation advertiser went from about 1,900 conversions per month to zero in our data warehouse. Absolutely zero. <\/p>\n\n

The campaigns hadn’t broken down. The advertiser had migrated its tracking to a new CRM (Salesforce) and, in the process, had redefined what constituted a conversion. Our campaign teams were aware of this, were on top of the situation, and continued to optimize using the correct data: the data coming directly from the client’s CRM. In the real world, leads were coming in as usual. The system was working. <\/p>\n\n

What broke was an internal issue\u2014and a much more subtle one. Our data warehouse was still tracking the old conversion metric (the one that the change had stopped triggering). No one on our end had updated which conversion type should be tracked, and the data warehouse that feeds our reports, our dashboards, and our own models ended up, without warning, tracking a phantom. Reality said \u201chealthy\u201d; our data said \u201czero.\u201d <\/p>\n\n

And to catch exactly that, we have an anomaly detector: to detect silent discrepancies between what\u2019s actually happening and what the data reflects. A metric that drops for no apparent reason is a textbook use case. It didn\u2019t catch it. Every morning, it checked the account, and every morning it issued the same verdict: everything normal. <\/p>\n\n

And, technically, he was right. That’s the problem. <\/p>\n\n

Because in performance marketing, a drop to zero is the most expensive failure there is\u2014and, at the same time, the quietest. A spike in spending or a skyrocketing CPL scream out: they stand out on any dashboard, grab your attention, and trigger a notification at nine in the morning. A zero doesn\u2019t. A zero doesn\u2019t raise its voice. The curve flattens out, the cost per lead can no longer be calculated, and the screen remains\u2026 calm. And it\u2019s very easy to mistake that calm for good health. <\/p>\n\n

Let\u2019s be clear, then, about what this story is really about: the campaign management didn\u2019t fail, nor did the client stop selling. The safety net failed\u2014the system that was supposed to alert us that our data no longer reflected reality\u2014and it failed in the worst possible way: silently, by giving the green light. This article is about why building that safety net isn\u2019t, at its core, a problem of statistics, but rather one of defining what \u201cnormal\u201d means; and how that poorly defined concept leaves your system blind to precisely what matters most. <\/p>\n\n

2. The real problem isn’t the statistics; it’s defining what “normal” means in media data<\/h3>\n\n

When people think of an \u201canomaly detector,\u201d they usually picture statistics: moving averages, standard deviations, z-scores, and bands that trigger when a number falls outside the expected range. And that part is\u2026 the easy part. The math needed to say \u201cthis is three standard deviations below normal\u201d fits into just a few lines of code. <\/p>\n\n

The hard part\u2014what really determines whether your system works\u2014is answering a much more uncomfortable question: What is \u201cnormal\u201d for this metric? And in media, \u201cnormal\u201d is a moving target. Campaigns launch and pause, budgets rise and fall, there\u2019s seasonality, slow weekends, Black Friday, and platforms that report a day late. Half of what a naive detector flags as an \u201canomaly\u201d is nothing more than just another Tuesday. <\/p>\n\n

And that\u2019s the catch. For a detector to be usable, you have to filter out that noise: if it goes off ten times a day over things that don\u2019t matter, no one will pay attention to the eleventh time\u2014which actually did matter. So you fine-tune it to be accurate, so that it only alerts you when it\u2019s reasonably certain. And in doing so, without realizing it, you build a system that\u2019s deaf precisely to the most serious errors: the ones that don\u2019t look like \u201ca number outside the range.\u201d <\/p>\n\n

It\u2019s also important to distinguish between two tasks that are constantly confused. Monitoring the health of the campaign is something people already do. The detector\u2019s job is different\u2014and more subtle: monitoring the integrity of the data and alerting you when your data warehouse<\/a> no longer reflects reality. These are different things, and the most common mistake is to assume that the former covers the latter. It does not. In our incident, the campaign\u2019s health was perfect; what had broken was the mirror. <\/p>\n\n

What follows is a list of specific ways in which that mirror shatters. Six blind spots, divided into two categories: the things the system fails to see (false negatives: the actual failure occurs, and no one notices) and the things the system invents (false positives: the alarm goes off when nothing is happening). Let\u2019s start with the first ones, which are the ones that cost us the incident. <\/p>\n\n

3. The Anatomy of a Blind Spot in Campaign Monitoring<\/h3>\n\n

Two axes and four cells, with almost all the danger concentrated in a single column:<\/p>\n\n

<\/th>The value decreases (e.g., 100 \u2192 20)<\/th>The value goes to zero \/ disappears<\/th><\/tr><\/thead>
Volume Monitor (Absolute Threshold)<\/strong><\/td>\u2705 Detects: \u201ctoday < threshold\u201d<\/td>\u274c Blind: compares with the current threshold, not with its own history; and a missing row is not visible<\/td><\/tr>
Efficiency Monitor (Ratio: CPL \/ CPA)<\/strong><\/td>\u2705 Detects: \u201cCPL spikes\u201d<\/td>\u274c Blind: CPL = NULL\/\u221e, no data \u2192 no alarm<\/td><\/tr><\/tbody><\/table><\/figure>\n\n

The right-hand column (the drop to zero) is the blind spot that no naive monitor watches. That’s where our incident occurred. <\/p>\n\n

The strength of the chart is that both indicators fail in the same column for different reasons: the efficiency indicator because its ratio becomes undefined, and the volume indicator because it compares today\u2019s value to a fixed threshold rather than to the historical data for that same channel. Let\u2019s go through it cell by cell. <\/p>\n\n

What the System Doesn’t See<\/h4>\n\n

The blind spot of zero.<\/strong> Most of the metrics that really matter in performance are ratios: CPL, CPA, cost per acquisition. And a ratio hides a trap: it divides by the result. As long as there are conversions, the CPL goes up or down, and a tracker follows it without any problem. But when conversions drop to zero, the CPL doesn\u2019t become \u201cbad\u201d\u2014it becomes undefined: a division by zero, a NULL, an infinity. A system monitoring \u201cHas the CPL skyrocketed?\u201d doesn\u2019t see an alarming value; it sees a lack of data. There\u2019s nothing to compare it to. The problem isn\u2019t that the alarm doesn\u2019t go off\u2014it\u2019s that it has no way to go off. That\u2019s exactly what happened to us: conversions in the store were zero, so the CPL disappeared, and with it, any signal. <\/p>\n\n

The trap of \u201cthe present, not the past.\u201d<\/strong> \u201cOkay,\u201d you think, \u201cI\u2019ll keep an eye on the volume: if today\u2019s conversions fall below a certain threshold, I\u2019ll send an alert.\u201d Better, but not enough. A fixed threshold looks at today\u2019s value in isolation and can\u2019t distinguish between a channel that has plummeted to zero and one that has always been close to zero because it barely generates leads\u2014and in that case, that\u2019s normal. For today\u2019s zero to mean anything, you have to compare it to that same channel\u2019s past: 1,900 \u2192 0 is an emergency; 3 \u2192 0 is just another Tuesday. The signal isn\u2019t in the number itself; it\u2019s in the difference from its own history. <\/p>\n\n

“Disappearance” is not the same as “low numbers.”<\/strong> There\u2019s a more subtle distinction. Sometimes the problem isn\u2019t that a row is zero, but that the row ceases to exist. If the channel disappears from the feed, there\u2019s no \u201c0\u201d to watch for\u2014there\u2019s nothing there. And almost all detectors reason based on what\u2019s present; they scan through rows and evaluate them; what isn\u2019t there isn\u2019t evaluated. Detecting an absence requires reframing the question: no longer \u201cIs this number an outlier?\u201d but rather \u201cWhat did I expect to see that isn\u2019t there?\u201d This requires maintaining a list of what should appear and monitoring streaks of zeros or silence. <\/p>\n\n

The deployment gap.<\/strong> The most mundane of all\u2014and therefore the most common. A detector only catches what happens while it\u2019s looking. If you build it\u2014or fix it\u2014after an incident, it won\u2019t find it: it\u2019s already happened. Looking ahead isn\u2019t enough; you also need a retrospective scan of the history, searching for spikes and changes in levels that occurred before the system itself existed. Monitoring without memory is blind to its entire past. <\/p>\n\n

What the system comes up with<\/h4>\n\n

The false positive due to freshness.<\/strong> Now, the opposite scenario. Suppose you get everything above right and set up a detector that\u2019s sensitive to drops to zero. Morning comes, a connector is halfway through syncing\u2014which is common: some integrations take hours, or fail and retry\u2014and half a day\u2019s worth of data still hasn\u2019t come in. To your detector, a perfectly healthy channel appears to have crashed. If you raise the alarm, you\u2019re crying \u201cwolf!\u201d; and the third time the wolf doesn\u2019t come, you lose credibility. The only solution is to model the data\u2019s freshness: determine whether a day\u2019s data is complete before judging it, and keep quiet\u2014or mark it as provisional\u2014about what\u2019s still loading. <\/p>\n\n

Not every zero is a mistake.<\/strong> And finally, the most subtle of false positives. Sometimes a zero is\u2026 correct. For example, attribution is passed to the ad server, and the platform\u2019s native conversion rate drops to zero by design. A line is intentionally paused. A metric is no longer measured because it no longer applies. A detector that treats every zero as an emergency will end up triggering alerts in response to deliberate decisions, and every false positive erodes that trust that is so hard to build. The system doesn\u2019t just have to detect zeros\u2014it must have a way of knowing which zeros were expected. <\/p>\n\n

The Six Blind Spots, at a Glance<\/h4>\n\n
\n
\"\"<\/figure>\n<\/figure>\n\n

Four of the six are things the system doesn’t see; two are things it makes up. And the timeliness of the data lies at the heart of both\u2014it’s what distinguishes a “broken campaign” from “data that hasn’t come in yet.” <\/p>\n\n

4. What We’re Changing: Changes That Are Being Rolled Out Widely<\/h3>\n\n

The good thing about having a blind spot map is that the solutions are no longer improvised: each fix addresses a specific issue. None of them are particularly sophisticated; the hard part was figuring out what needed to be fixed. <\/p>\n\n

Compare against the high for the day (or all-time high), not against a fixed threshold.<\/strong> A direct response to the blind spot of zero and the \u201ccurrent\u201d trap: if you measure the drop using the greater of today\u2019s value and the channel\u2019s baseline as a reference, a 1,900 \u2192 0 triggers the alarm, while a 3 \u2192 0 stays where it should\u2014silent.<\/p>\n\n

Explicit detection of missing values \/ runs of zeros.<\/strong> Change the question from \u201cIs this number anomalous?\u201d to \u201cWhat value did I expect to see, and it hasn\u2019t appeared for N days\u2014or is it zero?\u201d Without this, missing values (the quietest kind of error) are never detected. <\/p>\n\n

Modeling data freshness.<\/strong> Before evaluating a day, check to see if it is complete. Data that is still being loaded is marked as provisional and does not trigger an alert. This is what distinguishes a \u201cbroken campaign\u201d from \u201cthe connector hasn\u2019t finished,\u201d and it prevents false alarms. <\/p>\n\n

Retroactive scanning upon deployment.<\/strong> When starting up or replacing a detector, don’t just look ahead: go back through the history to identify the steps that led up to its existence. Monitoring should be built with memory from the start. <\/p>\n\n

A person at the end of the loop.<\/strong> No flood of alerts: a daily summary, with severity levels, that arrives before the team starts their day. The true KPI of a monitoring system isn\u2019t how many anomalies it detects, but how much trust it inspires (and trust is earned by being accurate and honest about what it doesn\u2019t know). <\/p>\n\n

5. What We Take With Us (and this applies to anyone)<\/h3>\n\n

Beyond our specific case, here are a few ideas that can be applied to any system that monitors media data\u2014or data in general:<\/p>\n\n

“Everything’s normal” doesn’t mean “everything’s fine.” A quiet dashboard doesn’t prove that everything is healthy; it may just mean you’ve stopped tracking. <\/p>\n\n

Watch for the absence of what’s good, not just the presence of what’s bad. The most costly error is almost never a bad number\u2014it’s a number that should be there but isn’t. <\/p>\n\n

Every ratio hides a denominator that can trip you up. Keep an eye on the numerator and denominator separately\u2014not just the quotient. <\/p>\n\n

Compare everything to its own past and context, not to a universal standard: what is \u201cnormal\u201d in one channel is abnormal in another.<\/p>\n\n

A monitor fails in two ways (not seeing what’s happening and seeing what isn’t happening), and data freshness lies right in the middle of the two.<\/p>\n\n

Treat your monitoring system like a product: it has its own bugs, regressions, and blind spots. The day you assume it\u2019s \u201cdone\u201d is the day it starts lying to you. <\/p>\n\n

6. The safety net, and why it matters more and more<\/h3>\n\n

It’s tempting to read all this as a technical anecdote: a conversion mapping that didn’t propagate, a detector that didn’t catch on. But the issue is bigger than that, and it’s only getting bigger. <\/p>\n\n

We\u2019re automating more and more layers of media operations: bidding, budget allocation, optimization, and even quality control for the setup. And all of these automated processes draw from the same source: the data in the data warehouse. The moment that data ceases to reflect reality\u2014silently, as happened to us\u2014it\u2019s not just a report that\u2019s wrong: it\u2019s every automated decision that depends on it. The monitoring layer is the safety net for everything else, and its blind spots cease to be a local problem and become a systemic risk. <\/p>\n\n

That\u2019s why the work doesn\u2019t end with detecting zero-hour outages. The next step is to make the network smarter: baselines by entity, thresholds that account for seasonality, alerts that not only indicate what went wrong but also pinpoint the cause, and\u2014on the horizon\u2014detection linked to an automated response. <\/p>\n\n

For now, all it takes is a change in mindset. The next time a dashboard looks normal\u2014with everything in the green and everything calm\u2014it\u2019s worth asking the uncomfortable question: Is everything okay\u2026 or are we missing something? <\/p>\n\n

<\/p>\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

Estimated reading time: 8 minutes The Blind Spot in Anomaly Detection in Media Data We have an anomaly detector monitoring the data. The day our data warehouse stopped reflecting reality, it gave the green light. Why the most costly performance failure is also the quietest, and why defining \u201cnormal\u201d is harder than it seems. I\u2019ve […]<\/p>\n","protected":false},"author":4,"featured_media":10694,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[232,305],"tags":[224],"class_list":["post-10692","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-data-management-en","category-martech","tag-artificial-intelligence"],"acf":[],"yoast_head":"\nThe Blind Spot in Media Anomaly Detection<\/title>\n<meta name=\"description\" content=\"Our detector passed the test on the day the data warehouse stopped reflecting reality. The six blind spots that no naive monitor can detect.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"The Blind Spot in Media Anomaly Detection\" \/>\n<meta property=\"og:description\" content=\"Our detector passed the test on the day the data warehouse stopped reflecting reality. The six blind spots that no naive monitor can detect.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/\" \/>\n<meta property=\"og:site_name\" content=\"MIO One\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/SomosMIOGROUP\/\" \/>\n<meta property=\"article:published_time\" content=\"2026-06-09T10:46:19+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-07-09T08:18:15+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/mio.one\/wp-content\/uploads\/sites\/2\/2026\/06\/image-1781001744160-2-scaled.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"2560\" \/>\n\t<meta property=\"og:image:height\" content=\"1429\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"comunicacion\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@SomosMiogroup\" \/>\n<meta name=\"twitter:site\" content=\"@SomosMiogroup\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"comunicacion\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"13 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/\"},\"author\":{\"name\":\"comunicacion\",\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/#\\\/schema\\\/person\\\/d535a756e871c6347b8cecf3a9282a1b\"},\"headline\":\"“Everything’s normal” isn’t the same as “everything’s fine”: the blind spot in anomaly detection in media\",\"datePublished\":\"2026-06-09T10:46:19+00:00\",\"dateModified\":\"2026-07-09T08:18:15+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/\"},\"wordCount\":2687,\"publisher\":{\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/mio.one\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2026\\\/06\\\/image-1781001744160-2-scaled.jpg\",\"keywords\":[\"Artificial intelligence\"],\"articleSection\":[\"DATA\",\"Martech\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/\",\"url\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/\",\"name\":\"The Blind Spot in Media Anomaly Detection\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/mio.one\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2026\\\/06\\\/image-1781001744160-2-scaled.jpg\",\"datePublished\":\"2026-06-09T10:46:19+00:00\",\"dateModified\":\"2026-07-09T08:18:15+00:00\",\"description\":\"Our detector passed the test on the day the data warehouse stopped reflecting reality. The six blind spots that no naive monitor can detect.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/#primaryimage\",\"url\":\"https:\\\/\\\/mio.one\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2026\\\/06\\\/image-1781001744160-2-scaled.jpg\",\"contentUrl\":\"https:\\\/\\\/mio.one\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2026\\\/06\\\/image-1781001744160-2-scaled.jpg\",\"width\":2560,\"height\":1429,\"caption\":\"The Blind Spot in Anomaly Detection in Media Data\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Portada\",\"item\":\"https:\\\/\\\/mio.one\\\/en\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"“Everything’s normal” isn’t the same as “everything’s fine”: the blind spot in anomaly detection in media\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/#website\",\"url\":\"https:\\\/\\\/mio.one\\\/en\\\/\",\"name\":\"Mio One\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/mio.one\\\/en\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/#organization\",\"name\":\"Mio One\",\"url\":\"https:\\\/\\\/mio.one\\\/en\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/mio.one\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2023\\\/09\\\/mioone-en.svg\",\"contentUrl\":\"https:\\\/\\\/mio.one\\\/wp-content\\\/uploads\\\/sites\\\/2\\\/2023\\\/09\\\/mioone-en.svg\",\"width\":782,\"height\":782,\"caption\":\"Mio One\"},\"image\":{\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/www.facebook.com\\\/SomosMIOGROUP\\\/\",\"https:\\\/\\\/x.com\\\/SomosMiogroup\",\"https:\\\/\\\/www.instagram.com\\\/somosmiogroup\\\/\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/miogroup\\\/\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/mio.one\\\/en\\\/#\\\/schema\\\/person\\\/d535a756e871c6347b8cecf3a9282a1b\",\"name\":\"comunicacion\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/6b0a338c590839724a3da0b111c410569242af145ffefee60fcaf7117a667ecd?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/6b0a338c590839724a3da0b111c410569242af145ffefee60fcaf7117a667ecd?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/6b0a338c590839724a3da0b111c410569242af145ffefee60fcaf7117a667ecd?s=96&d=mm&r=g\",\"caption\":\"comunicacion\"},\"url\":\"https:\\\/\\\/mio.one\\\/en\\\/blog\\\/author\\\/comunicacion\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"The Blind Spot in Media Anomaly Detection","description":"Our detector passed the test on the day the data warehouse stopped reflecting reality. The six blind spots that no naive monitor can detect.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/","og_locale":"en_US","og_type":"article","og_title":"The Blind Spot in Media Anomaly Detection","og_description":"Our detector passed the test on the day the data warehouse stopped reflecting reality. The six blind spots that no naive monitor can detect.","og_url":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/","og_site_name":"MIO One","article_publisher":"https:\/\/www.facebook.com\/SomosMIOGROUP\/","article_published_time":"2026-06-09T10:46:19+00:00","article_modified_time":"2026-07-09T08:18:15+00:00","og_image":[{"width":2560,"height":1429,"url":"https:\/\/mio.one\/wp-content\/uploads\/sites\/2\/2026\/06\/image-1781001744160-2-scaled.jpg","type":"image\/jpeg"}],"author":"comunicacion","twitter_card":"summary_large_image","twitter_creator":"@SomosMiogroup","twitter_site":"@SomosMiogroup","twitter_misc":{"Written by":"comunicacion","Est. reading time":"13 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/#article","isPartOf":{"@id":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/"},"author":{"name":"comunicacion","@id":"https:\/\/mio.one\/en\/#\/schema\/person\/d535a756e871c6347b8cecf3a9282a1b"},"headline":"“Everything’s normal” isn’t the same as “everything’s fine”: the blind spot in anomaly detection in media","datePublished":"2026-06-09T10:46:19+00:00","dateModified":"2026-07-09T08:18:15+00:00","mainEntityOfPage":{"@id":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/"},"wordCount":2687,"publisher":{"@id":"https:\/\/mio.one\/en\/#organization"},"image":{"@id":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/#primaryimage"},"thumbnailUrl":"https:\/\/mio.one\/wp-content\/uploads\/sites\/2\/2026\/06\/image-1781001744160-2-scaled.jpg","keywords":["Artificial intelligence"],"articleSection":["DATA","Martech"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/","url":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/","name":"The Blind Spot in Media Anomaly Detection","isPartOf":{"@id":"https:\/\/mio.one\/en\/#website"},"primaryImageOfPage":{"@id":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/#primaryimage"},"image":{"@id":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/#primaryimage"},"thumbnailUrl":"https:\/\/mio.one\/wp-content\/uploads\/sites\/2\/2026\/06\/image-1781001744160-2-scaled.jpg","datePublished":"2026-06-09T10:46:19+00:00","dateModified":"2026-07-09T08:18:15+00:00","description":"Our detector passed the test on the day the data warehouse stopped reflecting reality. The six blind spots that no naive monitor can detect.","breadcrumb":{"@id":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/#primaryimage","url":"https:\/\/mio.one\/wp-content\/uploads\/sites\/2\/2026\/06\/image-1781001744160-2-scaled.jpg","contentUrl":"https:\/\/mio.one\/wp-content\/uploads\/sites\/2\/2026\/06\/image-1781001744160-2-scaled.jpg","width":2560,"height":1429,"caption":"The Blind Spot in Anomaly Detection in Media Data"},{"@type":"BreadcrumbList","@id":"https:\/\/mio.one\/en\/blog\/everythings-normal-isnt-the-same-as-everythings-fine-the-blind-spot-in-anomaly-detection-in-media\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Portada","item":"https:\/\/mio.one\/en\/"},{"@type":"ListItem","position":2,"name":"“Everything’s normal” isn’t the same as “everything’s fine”: the blind spot in anomaly detection in media"}]},{"@type":"WebSite","@id":"https:\/\/mio.one\/en\/#website","url":"https:\/\/mio.one\/en\/","name":"Mio One","description":"","publisher":{"@id":"https:\/\/mio.one\/en\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/mio.one\/en\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/mio.one\/en\/#organization","name":"Mio One","url":"https:\/\/mio.one\/en\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/mio.one\/en\/#\/schema\/logo\/image\/","url":"https:\/\/mio.one\/wp-content\/uploads\/sites\/2\/2023\/09\/mioone-en.svg","contentUrl":"https:\/\/mio.one\/wp-content\/uploads\/sites\/2\/2023\/09\/mioone-en.svg","width":782,"height":782,"caption":"Mio One"},"image":{"@id":"https:\/\/mio.one\/en\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/SomosMIOGROUP\/","https:\/\/x.com\/SomosMiogroup","https:\/\/www.instagram.com\/somosmiogroup\/","https:\/\/www.linkedin.com\/company\/miogroup\/"]},{"@type":"Person","@id":"https:\/\/mio.one\/en\/#\/schema\/person\/d535a756e871c6347b8cecf3a9282a1b","name":"comunicacion","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/6b0a338c590839724a3da0b111c410569242af145ffefee60fcaf7117a667ecd?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/6b0a338c590839724a3da0b111c410569242af145ffefee60fcaf7117a667ecd?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/6b0a338c590839724a3da0b111c410569242af145ffefee60fcaf7117a667ecd?s=96&d=mm&r=g","caption":"comunicacion"},"url":"https:\/\/mio.one\/en\/blog\/author\/comunicacion\/"}]}},"_links":{"self":[{"href":"https:\/\/mio.one\/en\/wp-json\/wp\/v2\/posts\/10692","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mio.one\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mio.one\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mio.one\/en\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/mio.one\/en\/wp-json\/wp\/v2\/comments?post=10692"}],"version-history":[{"count":2,"href":"https:\/\/mio.one\/en\/wp-json\/wp\/v2\/posts\/10692\/revisions"}],"predecessor-version":[{"id":10703,"href":"https:\/\/mio.one\/en\/wp-json\/wp\/v2\/posts\/10692\/revisions\/10703"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/mio.one\/en\/wp-json\/wp\/v2\/media\/10694"}],"wp:attachment":[{"href":"https:\/\/mio.one\/en\/wp-json\/wp\/v2\/media?parent=10692"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mio.one\/en\/wp-json\/wp\/v2\/categories?post=10692"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mio.one\/en\/wp-json\/wp\/v2\/tags?post=10692"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}