Automated Impatience Is Now a 2.5-Second Product Spec

Mujeeb Jaihoon, an Indian-born writer based in the UAE, calls modern gadgets automated impatience. The three-second click-away he describes is already a search ranking rule, and a grocery picker’s clock.

In a June 2026 essay, Jaihoon treats restlessness as a private vice that apps learned to feed. The machinery that grew around that vice is public, scored, and paid out to someone else.

Jaihoon Names the Habit Automated Impatience

Jaihoon is an author and social critic whose books have been translated into Italian, Arabic, Tamil, and Malayalam. He argues that every major upgrade in human history tried to shrink the gap between wanting a thing and having it, and that gadgets now automate the itch they were built to soothe.

Technology, at its core, is automated impatience.

Mujeeb Jaihoon, author, June 2026 essay

He remembers being thrilled to watch a dial-up page render over two minutes, then clicking away if a site takes more than three seconds. A 30-minute wait for a meal, he writes, now feels like a crisis. He is describing a habit. Product teams, search engines, and dark stores wrote that habit into targets.

The cognitive costs of engineered impatience show up in attention and mood. The second bill arrives in ranking reports and in the minutes a picker is given to clear a tote.

Google Grades Pages on a 2.5-Second Clock

Google Search Central tells site owners to have Largest Contentful Paint, the moment the biggest visible element renders, occur in the first 2.5 seconds of load. Interaction to Next Paint should stay under 200 milliseconds. Cumulative Layout Shift should stay under 0.1. Those three Core Web Vitals, Google says, align with what its ranking systems seek to reward.

That is softer than the SEO copy that treats a miss as an automatic penalty. Page experience still functions as a tiebreaker when relevance is close. The commercial clock is harder. Google and SOASTA, in 2017 research posted on Think with Google, found that as mobile load time goes from one second to three, the chance a visitor bounce rises 32 percent.

GOOGLE’S CORE WEB VITALS THRESHOLDS

Google classifies a page at the 75th percentile of real visits, so three of four loads must clear the good line. HTTP Archive’s 2025 Web Almanac still found only 62 percent of phone sites with a good LCP, against 74 percent on desktop. Across all three vitals, 48 percent of mobile sites passed and 56 percent of desktop sites did. The 2.5-second habit Jaihoon describes as personal is the line most of the mobile web is still chasing.

Why the Threshold Is 2.5 Seconds, Not One

Human-factors work never said three seconds was a moral limit. Jakob Nielsen, writing for Nielsen Norman Group, restated his 1993 response-time limits in 2010: 0.1 seconds feels instant, 1 second keeps thought seamless, and 10 seconds is about as long as attention holds. Those caps, he wrote, come from older laboratory research and do not move when networks do.

Google’s own Core Web Vitals threshold research, by Bryan McQuade and Barry Pollard, started from that literature. Newell’s range for an “immediate” response runs about 0.3 to 3 seconds. Miller warned that machine delays past about two seconds change the character of a task. A one-second LCP “good” line would have sat at the fast end of that range. It also failed a second test: enough sites had to be able to hit it.

Chrome User Experience Report data from April 2020 showed only 3.5 percent of phone origins meeting a 1-second LCP good line, and 13 percent at 1.5 seconds. At 2.5 seconds the share rose to 42 percent of phone origins and 51 percent of desktop origins. At 3 seconds it reached 55 percent on phones. Two and a half seconds was the fastest line that top-performing sites could hit consistently at the 75th percentile. The ranking spec is a compromise between what minds will bear and what the existing web could actually ship.

HOW THE WAIT LINE GOT WRITTEN

1. 1993: Nielsen publishes the 0.1 second, 1 second, and 10 second response-time limits from older human-factors studies.
2. 2010: He repeats the same three limits, arguing that broadband did not rewrite attention.
3. 2017: Google and SOASTA report that bounce probability rises 32 percent as mobile load time goes from one second to three.
4. May 2020: Core Web Vitals set LCP good at 2.5 seconds, with a poor line over 4 seconds.
5. March 2024: Interaction to Next Paint replaces First Input Delay, with a good line of 200 milliseconds.
6. June 2026: Jaihoon describes clicking away after three seconds as a private failure of patience.

Shop-floor SEO posts in September 2026 still recite 2.5 seconds as if it were a hard demotion. One auditor put a 5.9-second mobile LCP next to a 96 desktop performance score and noted that paid clicks arriving on a slow 4G phone bounce straight back to the results page. The docs reward a good field score. The ads budget punishes a slow one either way.

Inside the 30-Minute Grocery Promise

The meal that feels late at 30 minutes is assembled on a much tighter clock. Foodthink’s Yakou spent part-time shifts as a packer at a Xiaoxiang Supermarket, Meituan’s fresh-grocery format, in a Guangdong city, and wrote a field report on the 30-minute delivery warehouse clock. Xiaoxiang then ran nearly 1,000 outlets in 20 cities. Dedicated green-uniform riders, the report said, make the “30-minute lightning delivery” claim possible. Inside the depot, the picker is given 3 minutes.

An order lands on a handheld PDA. The worker weaves through ambient, cool, chilled, and frozen zones, scans rack and pack codes, and must finish before the next ticket appears. A top-shelf item on racks about 1.8 metres high would take about 30 seconds with a stool, a sixth of the whole window, so pickers climb the frames even where signs say not to. More than about 1.5 minutes in a cold store at minus 20 degrees, Yakou wrote, and the job starts to feel like it might end there.

Pay Drops If the Hour Misses 100 Items

A recruitment app advertised 20 yuan an hour for part-timers. That rate holds only if the picker clears 100 items in an hour. Miss it, and pay falls to 13 yuan. A customer complaint took 10 yuan with no appeal. Full-timers around Yakou worked six days on and one off, shifts of at least 12 hours, with one meal break. On the third day, after six hours, Yakou ran a fever. The station manager had planned a run from 4:30 p.m. to 10:30 p.m. with no break.

METRICS ON THE STATION DISPLAY

• Site scoreboard: A large overhead display runs the station’s on-time rate and 30-minute delivery rate in real time.
• Personal output: Average handling time per order, items processed per hour, picking on-time rate, and an “effective working hours” ratio sit beside a live leaderboard and a watchlist for stragglers.
• Idle trap: Time between tapping complete and accepting the next order counts as non-productive, so pickers hit complete before the bag is actually packed.
• Peak window: Part-time shifts cluster on weekends and from 4:30 p.m. to 7:30 p.m., when the floor is told to run.

Foodthink opened with a January mix-up at Hema, Alibaba’s grocery chain, in which a packer put narcissus bulbs into a lily order and a 71-year-old woman and a 12-year-old were hospitalized. Hema blamed a concurrent pick. The report treats those errors as what a three-minute clock produces, not as a one-off lapse.

Priority Tiers Sell the Same Wait Twice

Once the warehouse and the rider are already racing a 30-minute promise, the app still has a second product to sell: a more expensive wait. Delivery platforms split Saver, Standard, and Priority as if impatience were a menu. People who buy the top tier often watch the cheap order from the same kitchen arrive first. A Standard drop from a restaurant less than a kilometre away can still take two hours. The fee does not buy a faster kitchen. It buys a rank in a queue that was already late.

That is the consumer face of the same design Jaihoon wants individuals to resist. The interface offers a button for people who cannot sit with a spinner, then bills them for a promise the depot’s own board may not keep. Restlessness becomes a line item, billed to the customer and worked off by the picker whose hourly rate depends on clearing 100 items.

Amazon’s 1-hour grocery pitch, framed around impatient customers, follows the same pattern at a larger scale. The wait is not removed. It is reassigned, repriced, and given a badge.

A Pause at the Spinner Leaves the Spec Intact

Jaihoon asks readers to treat a slow app as a reminder to practice patience, and to protect learning, relationships, and creative work from the upgrade itch. That counsel still holds for a person staring at a lock screen. It does not rewrite Search Console, and it does not add a minute to a PDA.

The three-second click-away is no longer only a mood. It is a 2.5-second LCP target at the 75th percentile, a bounce curve that steepens 32 percent between one second and three, and a 3-minute pick in a depot whose wall shows the 30-minute rate. Waiting a few extra seconds in a browser does not change the 100-item rung that sets a packer’s wage, and it does not make Priority mean anything if Saver is already on the bike.

The bag that arrives in 29 minutes was picked on a three-minute clock, by someone whose pay drops if the hour falls short of 100 items.

Frequently Asked Questions

What Is a Good LCP Score for Google Search?

Google calls Largest Contentful Paint good at 2.5 seconds or faster, then still requires Interaction to Next Paint at 200 milliseconds or faster and Cumulative Layout Shift at 0.1 or less, because Search Console grades a URL group by its worst of the three; a page with a fast LCP and a poor layout-shift score is still Poor. Interaction to Next Paint replaced First Input Delay in March 2024, so any checklist that still lists FID is out of date.

Why Does Google Use the 75th Percentile?

McQuade and Pollard wrote that a 95th-percentile rule on 100 visits would let five odd slow loads swing the grade, while the 75th percentile needs 25 outlier visits before the score moves, which is why Google accepted a line that describes three of four real visits rather than the worst tail.

How Long Will People Wait for a Page to Load?

Nielsen’s cap for holding attention is 10 seconds, after which people start thinking about something else, and even an 8-second slideshow widget that filled 23 percent of a page drew only 1 percent of one tester’s viewing time, against 20 percent for a tester who looked after it had finished rendering.

Do Faster Delivery Windows Raise Rider Stress?

Borzo, an intracity delivery firm, surveyed 10,900 gig workers across 20 cities and found 57.3 percent reported stress on 15-minute jobs, 80.6 percent said a 60-minute window was safer to ride, and 56.5 percent said the hour-long window let them earn more without the same rush.