文 | 影子备忘录
11月1日，一个放在五年前会让电商运营团队通宵盯盘的日子。往年这个时候，朋友圈里刷屏的是凌晨的成交额截图、战报海报上一路跳动的数字、直播间里嘶哑着喊“最后三分钟”的主播切片。
但今天，却异常安静。没有人发战报，没有人晒订单，连那条“双十一首小时XX品牌破亿”的行业快讯，都像是被集体默契地折叠了。
一个跑了十八年的购物节，在今年把音量拧到了最低。
这不是错觉。百度指数的数据摆在那里：2021年10月25日，“双十一”的搜索指数还在1900左右，今年同期只剩380，五年缩水超过七成。
更刺眼的是峰值对比，2021年至今从近万跌到去年的1300上下，降幅超过86%。
晚会没了，跨店满减和定金预售在几个头部平台的规则里同时消失，直播间不再把“全网最低价”印在背景板上，连平台自己都不怎么提GMV了。
但“安静”只是一个表面症状。真正值得拆开来看的，是双十一赖以运转了十几年的那套底层机制，在今年被拆掉了。
往年这个时候，平台在忙什么
要理解今年有多“不正常”，得先回忆一下“正常”是什么样子。
往年的10月下旬到11月初，是双十一最紧绷的冲刺段。天猫和京东会轮番举办预热晚会，明星阵容、舞台特效、平台高管站台，一场晚会的制作成本以千万计，换的是开播前最后一个周末的全网注意力。
预售在10月20日前后就已经打响，定金锁客、尾款提醒、跨店满减的凑单攻略铺满小红书和什么值得买。
11月1日凌晨，战报如期而至——“首小时XX个品牌破亿”“XX品类同比增长XX%”，数字本身就是营销素材。
品牌方的营销预算在这个窗口集中释放，达播排期早在两个月前就被抢空，头部主播的坑位费水涨船高，商家咬着牙往里挤，赌的是一夜之间把半年库存清掉。
那套打法的核心不是“卖货”，是制造一个信息不对称的峰值。
满减规则越复杂，凑单空间越大，消费者越算不过平台；预售锁得越早，沉没成本越高，反悔的概率越低；战报数字越夸张，从众效应越强，犹豫的人越容易被裹进去。
这是一台精密设计的冲动消费机器，双十一不过是它一年一度的全功率运转。
今年，这台机器的核心零件被挨个卸了下来。
信息差生意做不下去了
最先被拆掉的是“算术题”。天猫把跨店满减整体拿掉，改成按店铺维度生效的“官方立减”，分1%、5%、15%三档；京东彻底取消定金预售，10月12日晚8点直接现货开卖；抖音主推“一件直降”，连凑单的入口都不给了。
花旗的研究报告把这件事说得很直白：平台策略从复杂的“凑单规则”全面转向“直接降价”，目的是遏制虚假刷单、降低退货率，把补贴精准投向高终身价值会员。
但“让利”只是表层叙事。更深一层的问题是：当比价成本趋近于零，靠规则制造信息差的生意本身就失去了根基。
过去消费者算不过平台，是因为同一件商品在不同渠道、不同组合下的真实到手价被层层包装。
现在手机上一搜，同一瓶香水的价差从一千二到五百出头摆在那里，任何复杂的满减设计都经不起一次截图比价。
规则越复杂，消费者的不信任反而越深，中消协的提示里“价格不透明、规则复杂化”仍然是投诉最集中的地带。
平台简化规则，与其说是“良心发现”，不如说是在消费者的比价能力面前，那套算术卷已经交不出有说服力的答案了。
预售的退场是同一件事的另一面。京东取消预售、天猫大幅压缩预售窗口，表面上是“体验优化”，实际上是锁客逻辑的失效。
当消费者已经学会把定金当作“可放弃的期权”而非“不可逆的承诺”，预售制造的不再是确定性，而是退款和纠纷。
规则可以一夜之间改掉，但消费心态的变化是更慢、也更根本的变量。
今年一个被反复提及的词是“纯过日子”。这个词从8月开始在微博上发酵，到双十一前夕已经成了一种被广泛认领的身份标签。
它的含义并不复杂：消费者开始用“值不值得用”而不是“便不便宜”来投票。
艾媒咨询针对新生代消费者的调研显示，67.08%的受访者把“消费降级”理解为更合理地分配预算，在不那么重要的地方省一些，把钱花在自己真正看重的需求上。
拉勾招聘的数据更直接：47%的一线城市职场人放弃了在双十一购物，29%的购物者消费低于1000元。
00后的态度可能是最清晰的信号。对这批正在陆续步入职场的年轻人来说，双十一曾经是一场带着社交属性的青春仪式，比如熬夜盖楼、组队做任务、宿舍里商量“喵糖”战略。
但今年，钛媒体的采访里，研究生毕业在即的洋洋和室友们各奔东西，忙着论文和工作，“熬夜组队盖楼的日子只停留在回忆里”。
一个在互联网公司做运营的年轻人说得更干脆：“想买的东西双十一根本没有。”她爱买二次元周边，那些东西平时就买得起，“只会在想买的时候买”。
这不是“消费降级”，也不是“年轻人不花钱了”。这是双十一作为“一年一度的价格窗口”的稀缺性，被日常促销和即时满足彻底稀释了。
每个月都有平台大促，中间还夹着骑行节、美妆节、洗护季、萌宠囤货周，连冰箱洗衣机都要单独办一个“冰洗厨超级品类日”。
“天天低价”已经承接了纸品家清、食品、母婴这些高频补货场景，即时零售承接了临时缺货，双十一能独占的消费场景正在一块一块地被切走。
当一个消费者可以在任何一天以接近甚至相同的价格买到想要的东西，专门等到11月11日下单的理由就消失了。
直播间里“人”的退潮
今年双十一还有一个藏得更深的变化：头部主播集体“噤声”了。
与往年满屏“破亿”“秒罄”不同，2026年双十一首波战报里，平台与直播机构的口径明显收敛，不再把超级主播的数字当作核心叙事。
这背后是一组结构性数据的位移：达人成交占比从2024年的43.5%压到2026年的7.5%，韩束83%的销售额来自1086个自有直播账号；店播在抖音的占比突破62%，达人播跌破40%；达人播的平均ROI只有1比1.2，而品牌自播是1比4.5。
头部主播带动成交的占比，从2022年的26.66%降到了10.66%。
品牌算的是一笔很朴素的账。达人直播的佣金从40个点抬到60个点，加上投流费用和高达35%左右的退货率，一场合作下来利润薄如纸，甚至净亏。
而品牌自播的退货率只有8%左右，用户进直播间时已经完成了品牌认知和产品筛选，落差被压小，成交的确定性完全不同。
当“流量”变成一次性消耗品，而“留量”可以沉淀进自己的账号矩阵，品牌的选择是理性的，也是不可逆的。
这对双十一热度的打击是釜底抽薪式的。过去几年，超级主播是双十一最大的“话题发动机”——他们的战报、翻车、互撕、复出，本身就是流量内容。
今年小杨哥复出，新账号粉丝停在6.4万，直播间反应平淡。不是他一个人的问题，是以“人”为中心的流量放大器模式，在整个电商系统的效率账本里，已经排不到前面了。
如果把视角拉到平台层面，今年双十一的“安静”还有一个容易被忽略的推手：平台已经不把总成交额当成核心指标了。
一个被反复引用的宏观背景是：2026年前三季度全国网上商品零售额同比增长只有3.8%，前8个月社零增速1.1%，8月单月0.4%。
流量大盘见顶不是新闻，但在双十一这个历来“不惜一切代价冲量”的节点上，平台集体接受低增长、放弃造势，是一个信号。
花旗预计今年双十一GMV只会录得低至中单位数增长，平台自己也清楚，靠一场大促拉动大盘的时代结束了。
取而代之的，是两件更“安静”的事。
一是AI。79.4%的消费者在购物前会先问AI，AI对消费决策的影响力达到43.7%，首次压过传统广告。
平台把AI导购、AI选品、AI拆单摆上货架，入口在从搜索框和直播间向AI对话迁移。但AI给出的“最优解”，到底是在帮消费者省钱，还是在帮平台把库存周转和退货率算到最优，需要多留一个心眼。
二是店播和货架。平台把流量、补贴、搜索优先权往品牌自播和旗舰店倾斜，达人播从“必选项”变成了“可选项”。
货架场在双十一的预算占比据估算能达到60%到70%，比去年高出至少10个点。这些动作都不会产生“战报”，但它们决定了平台下一个阶段的利润结构。
还有一个外部变量不容忽视：监管在“反内卷”上的持续加压。
国家邮政局在双十一首波高峰前召开座谈会，明确持续深化快递行业“反内卷”，稳定末端派费、整治跨区揽件、禁止总部向网点转嫁成本。
市场监管总局在10月下旬召集主要平台开质量提升座谈会，直指假冒伪劣、货不对板、虚假宣传。
从去年开始反复强调的“治理低价竞争”，在今年的双十一规则里变成了可感知的动作：平台连“全网最低价”都不敢印在海报最显眼的位置了。
当“低价”这张牌本身在合规层面变得烫手，造势的语言武器就被缴了械。
安静之后，双十一还剩下什么
写到这里，一个自然的追问是：双十一是不是要死了？
我的判断是：作为“狂欢节”的双十一，确实在退场；但作为“大促基础设施”的双十一，不会消失，只是价值内核发生了位移。
过去的双十一，对平台是秀肌肉、抢新用户的增量战场，对商家是赔本赚吆喝、刷单扩规模的窗口，对消费者是囤货、薅羊毛、情绪释放的出口。
今年的安静，把这套叙事里的水分挤掉了。平台把它当作商业化变现的蓄水池和会员留存的检验场；商家把它当作清库存、回笼资金、推新品的确定性节点；消费者把它当作刚需集中采购和价保兜底的时机。
交易体量还在，参与的品牌和用户数没有崩塌，只是所有人都不再假装这场大促能制造奇迹了。
真正被留下的问题，比热闹和安静重要得多。
当低价不再是最锋利的武器，竞争筹码就转移到了算法分配、供应链效率和综合体验上。
这对品牌来说意味着，靠一场直播冲出来的增长越来越不可靠，能不能把用户沉淀进自己的店播和会员体系，才是能不能在下一个双十一活下来的分界线。
对平台来说，放弃“卷低价”之后，如何在“短兵相接”的战场上组织竞争、分出胜负，是一道还没有答案的题。
对中小商家来说，低价权重下降未必是解脱，如果流量依然主要流向头部品牌和大型直播间，经营困境不会因为规则简化而自动消失。
十八岁的双十一，安静得让人不太习惯。
但一个需要靠嘶吼才能证明自己存在的节日，本身就已经虚弱了。
安静下来之后，该算的账才算得清楚。
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
A shadow memo.
On 1 November, a day was set up five years ago to keep the electric operators on board overnight. At this point in the past year, the circle of friends has been painted with a morning cut-off, a number of moving figures on the newspaper posters, and a major clip of the live broadcast where the noise was shouting “last three minutes”.
But today, it's very quiet. There were no battle papers, no tanning orders, and even the "two-one-hour-two-billion-billion-dollar" industry newsletter, it seemed like a collective fold.
An 18-year-old shopping festival that kept the volume to a minimum this year.
It's not a delusion. The data for the Baidu index are there: on 25 October 2021, the “two-xii” search index was still around 1900, leaving only 380 in the same period of the year, with more than 70 per cent in five years.
More striking is the contrast of peaks, which dropped by more than 86 per cent from nearly 10,000 in 2021 to over 1,300 last year.
The party was gone, the store was full of downs and downs and downs, and the rules of several front platforms disappeared, and the "lowest price on the Internet" was no longer printed on the background panel, even the platform itself.
But Quiet is just a symptom. What is really worth looking at is the bottom-up mechanism, which has been in operation for over a decade, which was dismantled this year.
What was the platform doing this time of year?
To understand how “normal” it is this year, it is important to remember what “normal” looks like.
Late in October and early in November of the previous year, the two-xiith most tense sprint. Skycat and Kyodong are organizing pre-heat parties, star formations, stage effects, platform executives, one of which costs tens of millions, in exchange for full attention on the last weekend before the launch.
The pre-sale had already sounded around 20 October, with down payment locks, tail alerts, short cuts across the store, a patch full of REDnote and what was worth buying.
In the early morning hours of 1 November, the war was on schedule — “Billions of first-hour brands” “XX-goods increase by XX per cent” and the numbers themselves were marketing asset.
The brand's marketing budget was released in this window, the broadcasting schedule was looted two months ago, the headmaster's pits were expensive and the merchants squeezed in their teeth, and the bet was to clear six months of stock overnight.
The core of that set is not “sale”, but the peak of creating an information asymmetry.
The more complicated the rules, the larger the room, the bigger the consumer will be the platform; the sooner the locks are locked, the higher the costs of sinking, the lower the probability of repentance; and the more exaggerated the figures in the battles, the more powerful the crowds, the easier those who hesitate to enter.
It's an impulsive consumer machine designed so well that the two elevens operate at full power every year.
This year, the core parts of the machine were unloaded.
I can't do this anymore.
The first one to be removed was the arithmetical question. The Skycat removed the total loss from the store and changed it to an “official reduction” effective at the level of the shop, with a third tranche of 1 per cent, 5 per cent and 15 per cent; Kyoto completely cancelled the advance and sold directly at 8 p.m. on 12 October; and Douyin pushed “a straight drop” and did not even give the billing entry.
Citicorp’s study makes this clear: the platform’s strategy has shifted from a complex “single rule” to a “direct price reduction” aimed at curbing false billings, reducing the rate of refunds, and channelling subsidies precisely to high-value members.
But Jeanly is just a surface narrative. The deeper problem is that when the cost of comparison is close to zero, the rule-making poor-informed business itself loses its roots.
In the past, consumers had been unable to calculate the platform because the same commodity had been packaged in layers at a real price under different channels and combinations.
The price difference for the same bottle of perfume is now between 1,200 and 500 hours away from the cell phone, and no complex full-down design can afford a single cut-off.
The more complex the rules, the deeper the mistrust among consumers, and the “transparent prices and complicated rules” in the hint of disharmony remain the areas of the most frequent complaints.
The platform ' s simplified rules are not so much a “conscience discovery” as a consumer ' s bargaining power, and the calculator book has no convincing answer.
The pre-sale exit is the other side of the same thing. The cancellation of the pre-sale in Kyoto and the significant compression of the pre-sale window by the Skycats, ostensibly “experative optimization”, are in fact the failure of the lock-in logic.
When consumers have learned to treat down payments as “renounceable options” rather than “irreversible commitments”, pre-solding is no longer a matter of certainty, but of refunds and disputes.
The rules can be changed overnight, but changes in consumption mindset are slower and more fundamental.
One of the words repeatedly mentioned this year is “to live in peace”. The term had been fermented on microblogging since August and had become a widely accepted identity label on the eve of the twentieth anniversary.
Its meaning is not complex: consumers start voting with “no worth” rather than “no convenience”.
According to a study conducted by the Aids Counsel for New Generation Consumers, 67.08 per cent of respondents understood “consumption downgrade” as a more rational allocation of budgets, and in less important locations, spent money on their real needs.
The data on recruitment are even more straightforward: 47 per cent of front-line urban workers give up shopping in two elevens and 29 per cent consume less than $1,000.
The attitude after 0000 is probably the clearest signal. For this group of young people who are moving on into the workplace, it was a youth ceremony with social attributes, such as overnight building construction, tasking teams, and discussing a “meat” strategy in the dormitory.
This year, however, in the titanium media interviews, graduate students traveled around Yoyo and roommates, busy with essays and work, “the nighttime team building only in memory”.
A young man who operates in an Internet company says it all the more straightforward: "There's nothing to buy." She likes to buy around the second dimension, which is usually affordable, “only when she wants to buy”.
This is not a “consumer downgrade” or “young people don't pay”. This is the scarcity of the two elevens as an “annual price window”, which is completely diluted by daily promotions and immediate satisfaction.
Each month, there is a huge platform, and in the middle there are riding festivals, cosmetics festivals, babysitting seasons, hoarding hoarding weeks, and even refrigerator washing machines run a single “Ice-Cooking Super-Cook Day”.
The “low daily price” has taken over the high-frequency feed scenes of paper household cleaning, food, mother and child, as well as the temporary short supply of instant retail supplies, and the consumption scene of the two-decalogues is being cut off in one piece.
When a consumer can buy what he wants on any day near or even at the same price, the reason for the 11 November bill is gone.
It's the downfall of "people" on the air.
This year, there is a much deeper change in the two decades: the collective “sniffing” of headmasters.
Unlike the “billions” of “seconds” to reach the screen, in 2026, the platforms and the live broadcasters experienced a marked loss in calibre and no longer treated the numbers of the supercasts as a central narrative.
Behind this is a shift in a set of structural data: the ratio of people entering into deals was down from 43.5 per cent in 2024 to 7.5 per cent in 2026, and 83 per cent of the sales of Korean beams came from 1,086 self-owned live accounts; the share of shops broadcasting in Douyin exceeded 62 per cent and fell by 40 per cent; the average number of people reaching ROI was only 1 to 1.2, while the brand was one to 4.5.
The proportion of transactions facilitated by front-runners decreased from 26.66 per cent in 2022 to 10.66 per cent.
The brand is very simple. The commission was raised from 40 points to 60 points, plus the cost of the drop and the return rate of up to 35 per cent, and a cooperative effort resulted in a low or even net loss.
With a return rate of only around 8 per cent for brands, users have completed their brand recognition and product screenings when they enter the live booth. The gap has been reduced and the certainty of the transaction is completely different.
When “flow” becomes a one-off expendable, and “retention” can sink into its own account matrix, brand choices are rational and irreversible.
This blow to the double 11th heat is a bottom draw. Over the past few years, supercasts have been the two largest “thematic engines” — their battles, rollovers, tearing each other apart, re-emerging — and are themselves the flow content.
Little Yang came back this year, and the new account fan stopped at 64,000, and the reaction was flat. It was not his own problem, but his flow amplifier model, which was “people-centred”, was out of line in the efficiency books of the entire electrical system.
If the perspective is brought to the level of the platform, this year's double 11 “quiet” has an easily overlooked pusher: the platform no longer considers total turnover as a core indicator.
A macro background that has been cited repeatedly is that, in the first three quarters of 2026, there was only a 3.8 per cent increase in online merchandise retail sales throughout the country, while in the first eight months the Society grew at a zero rate of 1.1 per cent and in August, 0.4 per cent.
While high-volume flows are not news, the Platform's collective acceptance of low growth and abdication of momentum is a signal at the historical “all-cost” node of the two elevens.
Citigroup expects that this year ' s double 11 GMV will only record growth in medium-sized units, and the platform itself knows that the time has ended with a big push.
Instead, there are two things more quiet.
The first is AI. 79.4 per cent of consumers ask AI before shopping, and AI has 43.7 per cent influence on consumer decisions, pressing traditional advertising for the first time.
The platform places AI-guided purchases, AI selections, and AI-dismantling lists on the shelf, and the portal moves from search boxes and live to AI dialogue. But the “best solution” given by AI is whether it saves consumers or helps the platform to calculate the best stock turnover and return rates.
The second is shop and shelf. The platform tilted its flow, subsidies, search priorities towards brand and flag stores, moving from "necessary options" to "options".
The ratio of the shelf to the budget in the eleventh was estimated to reach 60 to 70 per cent, at least 10 points higher than last year. None of these actions will produce the "war papers", but they determine the profit structure for the next phase of the platform.
There is also an external variable that cannot be ignored: the constant pressure on the “anti-inner volume” of regulation.
The National Post Office held a colloquium in advance of the two-first wave, which clearly and continuously deepened the express delivery industry's “anti-inner volume”, stabilized end-of-pipe fees, overhauled trans-regional packages and prohibited the transfer of costs from Headquarters to Internet sites.
In late October, the Directorate-General of Market Supervision convened a quality improvement seminar on the main platform, which referred directly to hypocrisy, miscalculation and false propaganda.
The “governance of low-price competition”, which began to be repeatedly emphasized last year, has become a visible move in this year's double 11 rule: the platform is afraid to put “the lowest price on the Internet” in the most visible position in the poster.
When the “low price” card itself becomes hotter at the compliance level, the aggressive language weapon is turned in.
After the silence, what's left of the elevens?
And here's the natural question: Is the double 11 dying?
My judgment is that, as the "marriage" is, indeed, retreating; but as the "grave infrastructure" is not going to disappear, but the kernel has shifted.
In the past, for both, the platform was an incremental battleground for muscles and new users, for merchants it was a window for profit and scale-up, and for consumers it was an export of hoarding, grab freebies, and emotional release.
This year's silence has squeezed the water out of this narrative. The platform uses it as a commercialized reservoir and as a testing ground for member retention; as a firm node for the clearing of stocks, for the return of funds, for the promotion of new products; and as an opportunity for consumers to focus on procurement and price preservation.
The volume of the trade is still there, and the number of participating brands and users has not collapsed, except that everyone no longer pretends that this big push is a miracle.
The real problems left behind are much more important than the excitement and silence.
When low prices are no longer the most aggressive weapons, competitive leverage shifts to algorithmic distribution, supply chain efficiency and integrated experiences.
For brands, this means that growing from a live broadcast is increasingly unreliable, and it is the line between survival and survival in the next 11 years that can sink users into their own shop and membership systems.
For the platform, it is an open question of how to organize competition and divide victory on the battlefield of the “short-to-long” after abandoning the “roll price”.
For small and medium-sized businesses, falling low-price weights may not necessarily be a relief, and if flows continue to flow mainly to head brands and large-scale broadcast outlets, the business dilemma will not automatically disappear as a result of simplified rules.
Eighteen-year-olds and elevens, quieter than ever.
But a holiday that needs to be proved by shouting is already weak in itself.
It's not until we settle the score.
Auto-fetched and translated by AI Atlas. All rights belong to the original author.