Integral Ad Science yesterday announced episode-level pre-bid optimization for programmatic audio on Spotify, making granular brand suitability controls available to advertisers via The Trade Desk ahead of a broader rollout to other demand-side platforms later this year.

The announcement, dated June 16, 2026, marks a significant technical step in programmatic audio advertising. For the first time, buyers on The Trade Desk can apply contextual classification at the individual podcast episode level - not merely at the show or app level - before placing a bid on Spotify inventory. The launch builds on an expanded partnership between IAS and Spotify, extending work that had previously been introduced for the Spotify Audience Network.

The gap that pre-bid audio classification fills

Podcast advertising in the United States has grown sharply. According to IAS, U.S. podcast ad revenue surged more than 17% to a record $2.9 billion in a recent measurement period. That growth is happening alongside a persistent structural problem: contextual classification for audio has long lagged behind the tools available for display and video.

The core issue is granularity. Show-level or app-level classification means a campaign targeting, say, a broadly suitable true-crime show could inadvertently appear in an episode covering subject matter that conflicts with a brand's specific suitability settings. The inverse is also a problem - an episode that would pass a brand's filters gets excluded because the show as a whole does not. Both outcomes represent waste, one reputational and one financial.

Episode-level classification has been discussed in the industry for some time as the logical next step for audio. The Trade Desk already supports contextual controls in display and video environments, and the programmatic infrastructure to handle pre-bid segment activation was in place. What was missing was audio-specific classification at sufficient depth to make episode-level targeting viable.

What IAS has built

The product is called IAS Context Control for Spotify podcasts. According to IAS, it allows advertisers to apply contextual controls before a bid is placed, maximising media quality performance across Spotify podcast impressions.

Four specific capabilities are described in the announcement. First, the system allows the same brand suitabilitystandards used across display and video campaigns to be applied to Spotify podcast inventory before bidding. Second, contextual classification runs at the individual episode level rather than at the show or app level. Third, marketers can choose from 33 contextual avoidance segments, spanning 11 industry-aligned categories, each available at three distinct risk levels: high, medium, and low. Fourth, the IAS Context Control pre-bid segments can be applied to Spotify campaigns directly within The Trade Desk's interface.

The 33-segment, 11-category taxonomy is not arbitrary. It mirrors the kind of industry-standard framework - aligned with GARM-adjacent classification structures - that brand safety vendors have used in display and video for years. The three-tier risk system (high, medium, low) gives buyers more precision than a simple block-or-allow binary. A brand that wants to avoid high-risk violence content but remains comfortable with, say, moderate-risk conflict coverage can set that distinction explicitly rather than blocking entire topic areas.

Lisa Utzschneider, CEO of Integral Ad Science, described the intent: "Programmatic audio has always allowed brands to reach a large and growing audience, and now, with this launch, it also allows for greater control. By launching episode-level pre-bid optimization for programmatic audio for Spotify, via The Trade Desk, we are enhancing brands' ability to confidently drive their podcast ad spend, aligned to their own brand suitability profiles."

The Trade Desk as launch partner

The choice of The Trade Desk as the first DSP to carry this capability is notable. Spotify launched its Ad Exchange in April 2025, naming The Trade Desk as a supported partner alongside Google Display and Video 360, Magnite, Yahoo DSP, and Adform. Spotify subsequently expanded automated podcast buying to 170 million listeners across 12 markets in July 2025, with The Trade Desk among the DSPs providing direct access to podcast ad inventory.

The Trade Desk's Kokai platform is now the primary interface through which the IAS Context Control pre-bid segments can be activated for Spotify. According to IAS, these segments will be available via The Trade Desk starting in July 2026, with a rollout to other major DSP platforms expected in the months after that. No specific additional DSPs or timelines were named in the announcement.

The Trade Desk's centrality here reflects both its reach in the open programmatic ecosystem and its existing relationship with both IAS and Spotify. The Trade Desk integrated Bunny Studio for audio ad creation in April 2025, signalling growing investment in the audio advertising infrastructure on its platform.

Context: how IAS arrived here

Today's announcement does not emerge from nothing. IAS has spent several years building out its contextual classification capabilities, beginning with the April 2022 launch of Context Control - a suite of brand safety and suitability tools built on IAS's patented Ozone semantic technology. That technology was originally developed through IAS's 2019 acquisition of ADmantX and uses natural language processing to classify content by meaning, sentiment, and emotional tone rather than by simple keyword matching.

The company then extended those capabilities into new environments in succession. IAS expanded brand safety measurement to Facebook and Instagram in February 2024. IAS and DoubleVerify introduced pre-screening content controls for Meta platforms in late 2024. IAS partnered with Snap for AI-powered brand safety measurement in mid-2024. TikTok-specific misinformation reporting followed in October 2024.

The pace accelerated further in 2025 and 2026. In December 2025, IAS launched IAS Agent, an AI-powered campaign analytics tool. In January 2026, automated supply path optimisation via Total Visibility produced quality spend improvements from 88% to 97% in campaigns from Digitas and Kinesso. In March 2026, IAS published Context Control Targeting performance data showing a 300% higher click-through rate for Samsung compared with alternative targeting approaches. In late March 2026, IAS and Mastercard announced a Sales Outcomes product linking media quality signals to anonymised purchase data. In early April 2026, IAS opened its Low-Quality GenAI Avoidance feature to open beta. And later in April 2026, IAS launched Total TV, giving CTV buyers show, genre, and rating-level content visibility across Disney, NBCUniversal, Paramount, and Prime Video inventory.

Audio is the final major programmatic channel where IAS had not yet delivered pre-bid classification at this level of granularity. Today's announcement closes that gap, at least for Spotify inventory via The Trade Desk.

What episode-level classification means technically

The distinction between episode-level and show-level classification matters more than it might initially appear. A show like a general-interest news podcast could include hundreds of episodes, each covering different subjects, formats, and emotional registers. Show-level classification assigns a single suitability label to the whole show. That label will tend to reflect the show's overall character - which means it will misclassify individual episodes that deviate from the norm in either direction.

Episode-level classification requires analysing the audio content of each episode independently. IAS has not publicly detailed the specific method - whether it uses transcript-based natural language processing, acoustic analysis, or a combination - but the classification output feeds into the same Context Control pre-bid segment framework the company already uses for web and video inventory. That framework, which has earned MRC accreditation in other environments, operates pre-bid: the segment data reaches the DSP before a bid is submitted, allowing the buying system to either bid or abstain based on whether the episode meets the advertiser's suitability criteria.

The 33 avoidance segments across 11 categories at three risk thresholds is a specific number worth noting. In display advertising, Context Control has operated with a comparable taxonomy. Extending that same structure to audio means buyers can, in principle, apply a single unified suitability profile across display, video, and podcast inventory - the same criteria evaluated consistently across formats. That cross-format consistency is what IAS describes as "confidence to scale."

Why the Spotify Audience Network matters

The announcement specifically references IAS's earlier work with the Spotify Audience Network, the ad network that extends Spotify's podcast inventory to third-party apps and publishers beyond Spotify itself. The AI-driven brand safety and suitability solutions previously introduced for that network formed the technical foundation on which episode-level pre-bid optimisation is now built.

The Spotify Audience Network broadens the reach of this capability. Advertisers buying Spotify podcast inventory via The Trade Desk are not limited to ads served within the Spotify app alone - the network extends to Spotify's publisher partners. Episode-level pre-bid classification applies across that full scope of inventory, which is what the announcement means when it refers to podcast ads "on and off Spotify."

Spotify expanded automated podcast buying to 12 markets in July 2025, and launched a Partner Program that introduced dual revenue streams combining audience-driven payouts with advertising monetisation. That infrastructure - the Ad Exchange, the Spotify Ads Manager, and the programmatic connections to The Trade Desk - is what makes today's announcement technically possible. Without the underlying programmatic pipes already in place, episode-level classification data would have nowhere to go in the buying workflow.

What this means for podcast advertisers

The practical implications depend on how advertisers are currently buying podcast inventory. Those who rely on broad topic targeting or show-level blocklists are most likely to see a change in their workflows. The new system allows a more surgical approach: instead of blocking an entire show because one episode covered a topic that conflicts with brand guidelines, a buyer can now exclude specific episodes while continuing to buy the rest of the show's inventory.

According to IAS, the IAS Context Control pre-bid segments for Spotify podcasts are available to apply directly within The Trade Desk's platform - meaning no separate workflow or third-party activation step is required for buyers already using both IAS and The Trade Desk.

PPC Land has tracked the structural imbalance between audio consumption and audio advertising investment for some time: despite digital audio commanding roughly 20% of all time spent with digital media, audio investment represented only approximately 2.9% of total digital advertising revenue as of mid-2025. A key part of that gap has been the absence of the verification and classification tools that display and video buyers take for granted. Today's announcement addresses a specific part of that gap directly.

The rollout begins in July 2026 via The Trade Desk. Additional DSP availability is described as coming "in the coming months," though IAS has not named specific platforms or provided a more precise schedule.

Timeline

Summary

Who: Integral Ad Science (IAS), a global digital media measurement and verification company owned by private equity firm Novacap following a $1.9 billion acquisition in 2025, in partnership with Spotify and The Trade Desk.

What: IAS launched episode-level pre-bid optimization for programmatic audio advertising on Spotify. The product, called IAS Context Control for Spotify podcasts, gives advertisers 33 contextual avoidance segments across 11 industry-aligned categories at three risk levels (high, medium, low), applied at the individual episode level before a bid is placed. It is the first time IAS has delivered this depth of pre-bid classification for audio inventory.

When: The announcement was made on June 16, 2026. IAS Context Control segments for Spotify podcasts will be available via The Trade Desk starting July 2026, with additional DSP platforms to follow in subsequent months.

Where: The product is available to advertisers buying Spotify podcast inventory - both within the Spotify app and across the Spotify Audience Network - through The Trade Desk's Kokai platform. Activation occurs directly within The Trade Desk's interface.

Why: Contextual classification for podcast advertising has historically operated only at the show or app level, making it difficult for advertisers to align campaigns with precise brand suitability profiles. Episode-level pre-bid classification resolves that limitation, allowing buyers to exclude unsuitable individual episodes while continuing to buy the rest of a show's inventory - addressing a gap that has contributed to the structural underinvestment in audio advertising relative to its share of consumer media time.