Back
Back

Ad spend at scale: inside adtech’s algorithmic era

Ad spend at scale: inside adtech’s algorithmic era

This year, for the first time ever, global ad spend is projected to exceed $1 trillion.

It may come as a surprise, because no more than five years ago, ad spend was actually on the decline; the post-covid era sent advertisers into a period of tightening budgets and cautious spend as brands grappled with economic uncertainty.

But amid AI-driven automation and the rise of new media formats, 2025 marked a tipping point: making algorithms a core driver of the industry’s renewed momentum. Enter, the algorithmic era.

Algorithms are shaping everything we do: the content we watch, the food we eat, and most crucially for advertisers, the things we buy. 35% of Amazon’s annual revenue is generated from product recommendations – that’s $200 billion coming directly from an algorithm.

“In the algorithmic era, the brands that win will be the ones that understand how discovery and decision making are shaped by algorithms and use media as a strategic engine to earn attention and build long term advantage” says Will Swayne, Global Practice President of Media and Integrated Solutions. “2026 rewards the marketers who innovate with intent, design for outcomes and meet people in the moments that matter.”

The conversation has shifted from ‘how do we survive budget cuts?’ to ‘how do we compete in a performance driven, always-on market?’ The result, inevitably, is a total market value that is set to nearly double since the pandemic.

What this means for ad spend, and why

Not just a single breakout channel or sudden surge in brand confidence, algorithms are a structural change to the advertising landscape. They touch almost every corner of the internet, from your social media feed to your check-out basket recommendations. For a product or service to be seen on an algorithm, money has to be spent on advertising.

Reports vary from 5.7% to 9.1% YoY increases into 2026, but the fact remains that the ad spend trajectory is on course to reach heights it’s never seen before. This success largely comes down to three main reasons.

Advertising is now closer to the transaction than ever before

Retail media networks, social commerce, and shoppable video have turned discovery into a data signal and conversion into a feedback loop. Algorithms do far more than merely deciding which ad appears - they learn what leads to a sale, adjust in real time, and reinvest budget where performance is proven. That connection has given brands more confidence to scale what works.

Automation has raised the ceiling on scale

Campaigns used to be limited by human capacity: how many creatives a team could produce, how many bids a trader could manage, and how many markets a platform could realistically support. Today, algorithms generate, test, and optimise thousands of variations simultaneously. When the systems and infrastructure can handle the load, budgets can grow with them.

Attention has become monetizable  

Streaming platforms, gaming environments, creator feeds, and retail marketplaces are now algorithmic ecosystems that shape what people see. For advertisers, this creates something closer to a living system than a static inventory, where data pulled by ‘moments’ (for example a search, product view, or scroll stop) can be tested, refined, and scaled in real time.

Together, these shifts have changed how growth actually happens. Campaign data flows into models, models adjust bids and creative, and those decisions shape the next wave of investments. The result is a feedback cycle that rewards platforms and brands able to learn faster than the market around them.

Same expectations, new demands

Advertising hasn’t suddenly rewritten its playbook. The fundamentals still matter: reach the right audience, prove impact, and scale what works. What’s changed is the environment those expectations now live in. As algorithms take on more of the decision-making, performance is being shaped by automated systems. The result is a market that still wants the same outcomes, but demands different methods to get there.

For 2026, companies in the adtech ecosystem need to be aware of four key expectations as we move further into the algorithmic era. 

1. AI is now impossible to ignore

If the last couple of years were about understanding where AI sits in the industry, 2026 is the year it cements itself into the adtech ecosystem.

“We’re moving from a position in the last 18 months of [AI] being experimental to actually becoming an embedded core capability for most of the adtech ecosystem” says Bradley Lewington, Adtech Sales Executive. “We’re seeing increased automation for bidding strategies, budget, planning and pacing, as well as increasing the efficiency across the entire creative process.”

60% of advertisers are now using AI for targeting and algorithm personalization, and a further 78% of brands use AI to optimize ad creative elements. This trend not only achieves a speed of output that is impossible with human hands, but targets adverts to the right audiences at the microscopic level. Clicks, shares, and watch times - just to name a few - are tracked using machine learning to put the right ad in front of the right person.

“The economics of advertising are being transformed. As the costs of production fall, the opportunities for advertisers multiply,” says David Cohen, CEO of the Interactive Advertising Bureau, reflecting how rapidly AI tools are already reshaping where and how brands create, optimise, and scale campaigns. “The pool of potential advertisers is growing, as it is easier than ever to plan, buy, optimize, and creatively connect with consumers utilizing new technologies across all forms of media.”

2. Click-Through-Rate (CTR) is on the decline

As AI pushes advertising forward, it’s also thinning out the ground beneath some of adtech’s older benchmarks. Click-Through-Rate is one of them. 

Google’s integration of AI-generated answers into search results – specifically AI Overviews and AI Mode – is causing disruption to traditional advertising revenue models. Despite ad-spend’s growth, search users are receiving their answers without having to click through to a website or resource, coining the phrase ‘zero-click searches’.

Pew Research Center tested this with an analysis of 68,879 google searches: those who weren’t met with an AI summary clicked on a search result nearly twice as often as those who were (15% vs 8%). This has led to a 15-64% decline in organic traffic, with roughly 60% of searches yielding no clicks at all.

What this signals for adtech isn’t a drop in demand, but a shift in where value is being created. As more discovery happens through algorithms - from AI-generated answers to retail platforms and social feeds - budgets are moving away from visibility on search pages and towards environments further down the customer journey. Global ad spend continues to grow because brands are finding performance in new places, using systems that connect exposure to outcomes.

3. Protection against ad fraud is as important as ever

As more budget flows into automated systems, fraud follows the same path. The algorithmic layer that now decides where ads appear and how money moves also creates new surfaces for exploitation. From spoofed traffic and bot-driven engagement to manipulated inventory in CTV and in-app environments, fraudulent data is now distorting the very information that algorithms rely on to make their next decision.

“Global ad fraud losses are projected to grow from $100 billion in 2024 to $172 billion by 2028, meaning the problem will reach roughly $131 billion in 2026” reports Tapper. “As budgets expand, fraudsters will follow the money, making it harder for advertisers to trust platform-reported performance.”

The scale of those numbers is already shaping how the industry behaves. Brands are tightening verification standards, agencies are building fraud checks into planning and reporting, and platforms are under pressure to prove the quality of the environments they sell, not just the volume of impressions they deliver.

The rise of ad fraud is tied, in part, to the scale of the market itself. As digital advertising has expanded into a global, automated economy, it has created the kind of volume that attracts exploitation. Where attention and budget move quickly, so do those looking to divert them. In a market increasingly steered by automated decisions, the ability to protect the integrity of performance data is part of how long-term credibility, and long-term growth, is earned.

4. Agentic AI is just getting started

We’re still in the early days of agentic AI, but already we’re starting to see its abilities bleed into the advertising space.

Unlike traditional automation which follows pre-set rules, agentic systems can observe performance, make decisions, and act across campaigns with minimal human input. Today, you can create agents that prioritize areas such as performance, brand suitability or contextual alignment using agents that “evaluate context, apply brand guidance, and make real-time decisions at the impression level, which leads to safer, smarter, and more sustainable campaigns” – Scope3.

Over time, the role of these systems is likely to expand from execution to orchestration - coordinating planning, buying, measurement, and iteration as a continuous loop. For advertisers, that points toward a future where strategy sets the direction, and software handles much of the day-to-day movement of spend, turning media operations into an always-on, self-adjusting layer beneath the brand.

Tracing this growth back to its physical footprint

Infrastructure sits at the core of algorithm-driven advertising. Every bid request, performance signal, and model update depends on systems that remain available can handle compute-intensive workloads efficiently. The reliability of these systems directly affects how quickly platforms respond to market shifts, and how confidently brands can scale their campaigns. With global ad spend continuing to rise, the physical hardware underpinning it all has never been more critical.

For continuous, 24/7 workloads, consistency is paramount, where costs aren’t being eaten up by unnecessary resource, and the risk of downtime is at a minimum. At the same time, campaigns generate bursts of demand, such as product launches, regional rollouts and holidays, where extra capacity is needed to flatten out the curve.

And yet AI is adding a new layer of complexity. Training and running models create compute-heavy workloads that place distinct demands on infrastructure, namely GPUs. In practice, this has made flexibility a core requirement, so teams can match their resources to the real-world needs of their platforms.

“If you’re a company who wants to spin up huge amounts of GPU clusters, you really have to factor in the cost/benefit ratio to that” comments Bradley Lewington. “Having dedicated machinery that runs the ‘non-AI’ 24/7 workloads, while using the hyperscalers exclusively for GPUs, will make the difference for companies who want to be as efficient as possible when running AI algorithms through their platforms.”

If you want to find out more about how servers.com can help you navigate a landscape of algorithms and growing ad-spend, you can reach out to Bradley on LinkedIn, or you can get in touch here.

Author: Nathan Jollands

Nathan Jollands, Content Writer

Nathan studied Creative Writing at Bath Spa University, including a six-month Erasmus scheme at Stockholm University in 2020. Outside of work, Nathan is both a film buff and car enthusiast.

Related articles