Data centers to power 520% rise in AI traffic this US holiday shopping season

The 2025 U.S. holiday shopping season is shaping up to be a watershed moment for data center-powered AI-driven commerce. According to Adobe Analytics, traffic originating from generative AI tools—measured by users clicking on a link—is projected to surge by around 520% year-over-year (YoY)¹.

This follows an extraordinary 1,300% YoY increase from the 2024 holiday shopping season and signals a shift that AI is no longer a novelty.

The biggest spike is expected in the 10 days leading up to Thanksgiving, with Thanksgiving Day alone forecasted to grow by 725%–730% YoY in AI-driven referrals. Nearly eight in ten consumers say they use AI assistants more frequently during or leading up to high-stakes shopping events, such as major sales or holidays².

THE AI-DRIVEN HOLIDAY SEASON

The U.S. holiday shopping calendar—from Thanksgiving to Cyber Monday, and the run-up to Christmas—is set to be the busiest yet for digital retailers.

Adobe’s forecast highlights several other notable trends:

• Retailers are increasingly leveraging AI image generation to create personalized product visuals and marketing assets, driving engagement and conversion.
• Mobile devices set to account for a record 56.1% of online holiday spending, translating to $142.7 billion, up 8.5% YoY.
• Total U.S. online holiday sales are projected to hit $253.4 billion, marking the first quarter-trillion-dollar season for U.S. e-commerce.

But beneath this spending surge lies an even more dramatic transformation, and that’s the rise of AI-assisted discovery and decision-making.

AI Is Changing How Consumers Shop

Generative AI isn’t confined to holiday shopping—it’s reshaping consumer habits year-round.

In August alone, traffic from AI-powered sources to retail sites was up 1,500% YoY, with nearly 300% growth since January³. This isn’t just a spike in clicks; it reflects a fundamental shift in how consumers discover, evaluate, and purchase products online.

Today, three out of four consumers are familiar with AI assistants, and more than one-third have used them for online shopping⁴. Among those who have embraced AI-driven shopping, their verdict is that the experience is overwhelmingly positive: 95% report satisfaction with their purchases, and 72% say they are less likely to return items when guided by an AI assistant.⁵

Consumers rely on AI assistants for a wide range of tasks:

• Product research (53%).
• Recommendations (40%).
• Deal hunting (36%).
• Gift inspiration (30%)⁶.

This explosion of AI-driven activity means every click, query, and checkout is now powered by the world’s most critical infrastructure: the data center.

The Backbone of the Digital Shopping Season

Behind every AI suggestion, product recommendation, and streaming sale alert sits a data center—silently ensuring uptime during peak demand.

A single outage during the peak holiday shopping season could result in million- or even billion-dollar losses in revenue. As AI workloads scale, that risk grows.

By 2030, global data center electricity demand is projected to more than double, reaching around 945 TWh— slightly more than the electricity consumption of Japan today⁷.

While racks for traditional IT applications consume around 5-10kW of power, AI-optimized racks are in the hundreds of kW or even as high as 1MW⁸.

These power-dense environments make uninterruptible power systems (UPS) and battery reliability more critical than ever.

When Internet Traffic Peaks and Pauses

While online shopping dominates headlines during the U.S. holiday season, other traditions also leave a clear mark on internet traffic patterns.

• Thanksgiving Day 2023: According to Akamai, video streaming traffic in the U.S. surged 65% compared to the previous week, driven by families watching the Macy’s Thanksgiving Day Parade and NFL broadcasts.
• Christmas Eve and Day: Data from Cloudflare Radar shows HTTP request volumes dropped as people focused on offline traditions like dinners, gift exchanges, and religious services. Globally, December 25 saw the steepest decline, with traffic down 19% compared to the previous week, while December 24 registered a 14% drop.
• New Year’s Day: Ranked among the lowest traffic days of the season.

These fluctuations underscore the need for flexible and resilient data center infrastructure. Operators must prepare for sudden demand spikes while also optimizing resources during predictable lulls, ensuring efficiency and reliability throughout the holiday season.

WHY UPS BATTERIES ARE MORE IMPORTANT THAN EVER

The holiday season is just one reason which underscores the importance of UPS batteries as the silent guardians of uptime. If the grid were to falter during a peak shopping or streaming time, these systems would kick in immediately, ensuring servers stay online and critical workloads remain uninterrupted.

In today’s AI-intensive environment, that reliability is non-negotiable.

Now Is The Time to Upgrade Your Data Center Reserve Power

With the U.S. holiday moratorium approaching, now is the ideal time to assess your data center’s backup readiness. Proactive planning today prevents last-minute disruptions when uptime matters most.

Three reasons to act now:

• Meet end-of-year project deadlines without last-minute delays.
• Prepare for potential winter outages with extra reserve power.
• Allocate remaining annual budget before it resets in the new year.

Proactive planning today means uninterrupted operations—especially during peak holiday traffic.

REFERENCES
1. Adobe Analytics. (2025, October 6). Adobe Holiday Shopping Forecast 2025. Retrieved from https://news.adobe.com/news/downloads/pdfs/2025/10/100625-adobe-holiday-forecast-2025.pd
2. Adobe Analytics, Holiday Shopping Forecast.
3. Adobe Analytics, Holiday Shopping Forecast.
4. Adobe Analytics, Holiday Shopping Forecast.
5. Adobe Analytics, Holiday Shopping Forecast.
6. Adobe Analytics, Holiday Shopping Forecast.
7. International Energy Agency (IEA). AI is set to drive surging electricity demand from data centres while offering the potential to transform how the energy sector works. Published April 10, 2025. Available at: https://www.iea.org/news/ai-is-set-to-drive-surging-electricity-demand-from-data-centres-while-offering-the-potential-to-transform-how-the-energy-sector-works.
8. DataCenterDynamics. The Impact of AI on Data Center Design. Available at: https://media.datacenterdynamics.com/media/documents/The_Impact_of_AI_on_Data_Center_Design.pdf.
9. Akamai Technologies. How Holiday Season Traditions Affect Internet Traffic Trends. Available at: https://www.akamai.com/blog/edge/how-holiday-season-traditions-affect-internet-traffic-trends.
10. Cloudflare. When the World Logs Off: Christmas, New Year’s, and the Internet’s Holiday Rhythm. Available at: https://blog.cloudflare.com/when-the-world-logs-off-christmas-new-years-and-the-internets-holiday-rhythm/.

Click HERE to contact an EnerSys® sales representative and find a battery backup UPS solution made for your data center.

About EnerSys

Sustainability

Caution Concerning Forward-Looking Statements
EnerSys is making this statement in order to satisfy the “Safe Harbor” provision contained in the Private Securities Litigation Reform Act of 1995. Any of the statements contained in this article that are not statements of historical fact may include forward-looking statements that involve a number of risks and uncertainties. A forward-looking statement predicts, projects, or uses future events as expectations or possibilities. Forward-looking statements may be based on expectations concerning future events and are subject to risks and uncertainties relating to operations and the economic environment, all of which are difficult to predict and many of which are beyond our control. For a discussion of such risks and uncertainties that could cause actual results to differ materially from those matters expressed in or implied by forward-looking statements, please see our risk factors as disclosed in the “Risk Factors” section of our annual report on Form 10-K for the most recently ended fiscal year. The statements in this article are made as of the date of this article, even if subsequently made available by EnerSys on its website or otherwise. EnerSys does not undertake any obligation to update or revise these statements to reflect events or circumstances occurring after the date of the article.