Aviram Sharon, co-founder and chief executive of the streaming ad technology company InTheGame (ITG), has argued that connected TV publishers could shorten their commercial breaks without giving up revenue by moving some of their advertising into six standardized formats that run outside the break. The argument appeared on IAB Tech Lab's website on October 29, 2026, in a post titled "Rebalancing the Ad Break" and filed under the Tech Lab's member-perspective tag.

It is an opinion piece by the executive of a member company. It is not a new specification, a guideline or a statement of IAB Tech Lab policy, and it creates no obligation for anyone. It is also an argument for formats that ITG itself sells, a point the post does not raise. Its one piece of evidence is a FreeWheel study about badly timed ad breaks; it contains no figures on what non-break formats earn, how viewers react to them, or how much of a break could move into them.

In Short

The boss of a company that makes TV ad products wrote that streaming services put almost all their ads into long breaks, which is exactly when people get up and leave, and that some of those ads could instead show up in smaller spots, such as when you pause a show or in a box beside a live game. If that worked, viewers would sit through shorter breaks while the streaming service kept earning roughly the same money. The post explains how industry rules that already exist describe and sell those ads, but it gives no numbers showing that viewers prefer them or that the money really stays the same.

What the post argues

Sharon starts from habit. Streaming, in his account, took over the broadcast commercial break intact, and with it the practice of placing the entire ad load inside pods. "Streaming inherited the commercial break from broadcast, and with it a habit: put every ad in the pod," he wrote. "On most CTV services the entire ad load sits in a handful of two- to four minute interruptions." The post gives no source for that range. Viewers, in his telling, "know exactly when the break starts, and that is the moment they reach for the remote, the phone, the next show- or simply leave the room."

Publishers, the post says, have had only two ways to respond: fewer breaks or shorter ones. "Both buy experience with revenue," according to Sharon. Carrying fewer ads improves the session but costs money, and nothing in the traditional toolkit escapes that trade.

The single statistic in the post comes next. "A 2024 FreeWheel study reported by eMarketer found that breaks placed at unnatural moments bothered 71% of viewers and cut brand recall by 14%, while programming with well-placed ads was rated as enjoyable as programming with none," Sharon wrote. His conclusion: "The problem is not that ads exist. It is where they are stacked."

The eMarketer account he links to supports the first two figures. According to eMarketer, FreeWheel studied 420 participants exposed to several ad experiences, including no ads, latency problems and unnatural breaks; unnatural breaks bothered 71% of them, raised perceived intrusiveness by 16% and lowered brand recall by 14%. The enjoyment finding is reported slightly differently. eMarketer compares programming with ads against programming without them, both rated 6.3 out of 7, and does not describe the ads in that comparison as well placed. The qualifier is Sharon's.

Seven places, six formats

The core of the post is a map of the viewing session. "A CTV session has seven places an ad can live, and the IAB Tech Lab CTV Ad Portfolio now defines the six outside the break," the post says.

His list runs as follows. Before and between viewing, home screen and menu ads place a tile or banner in an app's rows. During playback, pause ads fill the screen while the viewer has stopped; contextual overlays take a corner or the lower third while content keeps playing; L-bar and double-box squeezebacks shrink the picture rather than cover it; and in-scene ads put a billboard or poster inside the scene. "The seventh place is the ad break: still there, just shorter."

A diagram in the post sets this out in two columns. Under "Today" sits a single ad break, "the only standard slot," with "one-offs, direct-sold" beneath it. Under "With the CTV Ad Portfolio" come seven rows, each paired with an IAB format name: Menu ad, Menu ad, Pause ad, Overlay, Squeezeback, In-scene ad and Linear.

That mapping does not line up exactly with the document it cites. The IAB Tech Lab ad format guidelines that Sharon links to name six new CTV formats outside the traditional break: pause, menu, squeezeback, overlay, in-scene and screensaver. Sharon's six count the menu ad twice, once for home screens and once for in-app menus, and leave out the screensaver ad, which the guidelines define as an ad that starts after a set period of device inactivity. The post does mention a viewer who "leaves the device inactive" as one way a moment opens, but the screensaver never appears among the seven places. The guidelines also state that their list does not cover every CTV experience.

Why does the count matter? Because the post's commercial claim rests on it. "Because every format is now standardized, a buyer can plan across all seven moments in one campaign, the same way they buy the pod today," Sharon wrote.

The standards work behind that claim is recent. IAB Tech Lab released the six formats for public comment on December 11, 2025, with feedback due by January 31, 2026, and the signaling side followed when IAB Tech Lab finalized signaling guidance for six CTV formats on July 22, 2026, covering pause, menu and home screen, overlay, screensaver, in-scene and squeezeback units. Under that guidance, pause, screensaver, overlay, squeezeback and in-scene placements travel in the video object of an OpenRTB request with nonlinear VAST delivery, while menu and home screen ads use the native object. The AdCOM placement subtype list now carries five values for the in-player formats: 5 for pause, 6 for screensaver, 7 for overlay, 8 for squeezeback and 9 for in-scene.

A shared vocabulary is not the same as buyable supply, though. Whether a buyer can actually plan "across all seven moments in one campaign" depends on which publishers expose the units, which supply-side platforms pass the signals and which demand-side platforms bid on them. The post names none.

What the guidelines set for each unit

The guidelines date back to 2008 and were revised in 2015, before a 2022 update brought digital video and CTV formats into scope; the current version adds the six non-break formats and sets specifications for each. Pause ads run full screen at 1920 x 1080 or as a 600 x 600 partial-screen unit, and static files should stay under 350kB. Menu ads are specified by aspect ratio rather than fixed pixel size, as headline banners across a row or tiles inside the navigation.

Squeezebacks are the most detailed. The content shrinks over one to two seconds; in the L-shape and frame layouts it keeps 60% of the 1920 x 1080 screen, while in the double box layouts the content and the ad each take 25%. A fourth variant adds a branded background between the two boxes. Squeezebacks and overlays run for at least 10 seconds, with the actual duration signaled in the bid request. Corner overlays take 25% of the screen and lower-third overlays 30%.

In-scene ads are narrower than the post's wording might suggest. The guidelines currently standardize only out-of-home style insertions, such as billboards and posters composited into a scene, with no sound, no interactivity and a minimum brand exposure of three seconds. AdCOM's definition goes further and limits the current standardization to non-live content. In a live match, then, the in-scene slot is not one of the standardized moments. Linear spots, for comparison, are recommended at 6, 15, 20 or 30 seconds.

The stoppage slot

Sharon illustrates the argument with live sport. "In the example below, a double box runs during an injury stoppage in a live match, the moment a full break would have sent viewers away," he wrote.

The mock-up shows a football match, marked live and flagged as an injury stoppage, shrunk into a box on the left, with a video ad for a fictional coffee brand on the right and the brand's artwork filling the background. The ad carries a QR code offering free delivery and a line telling viewers that play resumes in a moment. Its caption describes one 1920 x 1080 underlay with a cut-out for the content, signaled as placement subtype 8 with a position value for the layout. In AdCOM's list of placement positions, value 12 is a double box and value 13 a double box with background, which is the layout shown.

"Content and ad each take about a quarter of the screen, side by side, and here the advertiser also brands the background between them," according to the post. "Sound stays off unless the publisher decides the stoppage is long enough for the ad to become the viewer's focus. The slot, the layout and the sound rules are all defined in the IAB spec, so a buyer can book it like any other standard unit."

The guidelines match that description. They make audio optional by default, with the publisher signaling in the bid request when it is required, and give a lull in sports content as the typical case. The caption adds that the content box is resized but never covered, and that after a minimum of 10 seconds the content returns to full screen.

The example also carries a tension the post does not address. Shrinking a live match to a quarter of the screen does not stop play, but it changes what the viewer sees far more than a corner overlay would. Sharon wrote that moving load into such moments "adds no interruption; it removes some." That is a definition of interruption based on whether content stops, not on how viewers perceive it, and the post offers no viewer data either way.

Who opens the moment

The post's most practical section concerns timing. Some moments, Sharon wrote, "are inherent to the format: a viewer pauses, enters a menu or leaves the device inactive. Other units can be activated manually by a live producer, at scheduled intervals, from production or event signals, or through AI-assisted detection of a timeout, replay, substitution or other natural events." For on-demand content, "scene context, emotional sentiment, chapter markers and pre-identified transitions can play the same role."

He presents these as working methods. "These are not theoretical approaches. AI and video telemetry are already being used to identify contextual advertising opportunities in live sports, while manual and scheduled activation allow publishers to retain editorial control over when an experience appears." No deployment, publisher, sport or result is named.

Each trigger, the post argues, yields a different kind of inventory. "Time-based rules provide predictable volume. Manual activation preserves editorial judgment. Event- and scene-based triggers make the placement more contextual." Publishers can combine them, coordinate pacing and frequency across a session, and sell the result "for direct sponsorships, programmatic demand or both." The division of labor, as Sharon puts it: "The standard gives buyers and sellers a common way to describe and transact the unit. The publisher's orchestration determines when that unit earns its place in the experience."

Live timing is where the industry's standards work is most active. A day before the post appeared, IAB Tech Lab gave the market until November 23 to review its Live Event Ad Playbook implementation guide, which covers forecasting audiences and inventory before an event and counting concurrent streams during it. The 2026 World Cup provided a large test of naturally timed breaks: FIFA approved ads in 208 hydration-break windows of roughly two minutes each, and iSpot later reported that hydration break pods retained viewers better than halftime or extra-time slots. Those were full breaks, not squeezebacks. They suggest that when a break falls matters, which is the same point the FreeWheel figures make.

Three levers, one of them additive

The post's title implies a transfer: ads leave the break and reappear elsewhere. Its own text is broader. "Rebalancing is not simply moving an impression from one slot to another," he argued. "It gives publishers more than one revenue lever. They can use formats outside the break to shorten pods while preserving monetization, create controlled incremental supply around moments that previously carried no advertising value, or package particularly relevant moments as premium sponsorships."

Only the first of those levers reduces time in breaks. The second adds advertising to parts of the session that carried none, and the third is a packaging choice that can go either way. A publisher could adopt the formats, keep its pods exactly as long as they are, and sell the new moments as extra inventory. Nothing in the standards prevents that.

Sharon anticipates the objection. "The goal is not to monetize every available second. It is to make more of the session monetizable without making more of it feel interrupted." That is a statement of intent. How publishers actually use the formats will decide whether viewers end up with shorter breaks or simply more ads.

What the evidence covers

The closing section restates the thesis. "The ad break is not the problem; its monopoly is. A session with shorter pods and standard formats between them can carry the same load with fewer interruptions, and the IAB Tech Lab standards that make it sellable are already published," Sharon wrote. "The publishers who treat ad load as a balance to set, not a count to cut, will be the ones whose breaks viewers stop leaving."

The claim that shorter pods plus outside formats "can carry the same load" is not backed by any number in the post. There is no revenue comparison, no CPM for pause or squeezeback inventory against linear spots, no sell-through rate, no completion or attention figure, no viewer survey on overlays or double boxes and no case study. The FreeWheel figures concern ad breaks placed at unnatural moments. They support better-timed breaks as much as they support moving ads out of breaks.

Other published research points in several directions. A FreeWheel and MediaScience study of more than 700 viewers found breaks of two minutes or less were the most preferred, below the two- to four-minute range Sharon describes. Magnite's US survey found respondents would accept an average of 8.7 minutes of ads per hour of live content. And in Magnite's South Korean survey of 905 streamers, 34% were aware of pause ads and 31% said they were likely to engage.

YouTube took a different route on television screens. Rather than shortening pods, it moved to fewer, longer ad breaks on connected TVs, saying in August 2024 that viewers could watch 50% longer before the next break. Consolidation and dispersal are opposite answers to the same complaint, and the post does not engage with the first.

Who is speaking

The byline lists Sharon as chief executive of InTheGame (ITG). According to ITG's website, the company sells advanced CTV units for streaming apps, among them pause, double box, L-band, shoppable, interactive, animation, banner and home screen units, together with an SDK, an ad router, a signaling system, an ad server and a demand product that brings brand budgets to publishers' inventory. The site describes triggering for live and on-demand content, including SCTE-35 support, presents the units as complementing existing ad breaks, and lists offices in Tel Aviv, London and Philadelphia. It also carries the company's own claims of higher attention than linear TV ads and more than 100 million devices running its technology, neither of which appears in the post.

The company started elsewhere. In 2020 Calcalist described Inthegame as an Israeli startup adding interactive layers of polls and quizzes to sports broadcasts, a graduate of the Techstars Israel accelerator that had raised $1.2 million at the time.

The post itself does not mention ITG's products, clients or results. It names no company other than FreeWheel and eMarketer. Its argument for squeezebacks, pause ads and home screen units is, in practice, an argument for categories of inventory ITG supplies, a commercial interest the post leaves unstated. The member-perspective tag marks the piece as a member's view rather than the Tech Lab's own.

Why this matters for the marketing community

For buyers, the post is a reminder that the CTV inventory map now extends well beyond the 30-second spot, and that the plumbing to describe those placements in bid requests exists. Among the in-player formats, pause ads have moved furthest into programmatic channels. Magnite rolled out pause ads across DIRECTV, DISH Media and Fubo in August 2025; Philo launched pause ads with eight supply partners in June 2026; and in July The Trade Desk opened its OpenPath direct route to CTV pause ad supply. On the interface side, LG opened its Home Screen ads to Magnite's AI agent in a three-year deal announced on October 14, 2026, one of several moves that make home screen placements a programmatic product rather than a one-off sale.

Squeezebacks, overlays and in-scene units have far less public evidence behind them. Standard dimensions and signals make them describable; they do not establish what they are worth, how often they will be available or how viewers will tolerate them. A buyer comparing a double box in a live stoppage against a 30-second spot in a hydration break has no common currency for the comparison in Sharon's post.

For publishers, the question Sharon raises is real. In his account the pod carries nearly all streaming ad revenue, and it is also where the audience is most likely to leave. Much of the industry's recent work has gone into making the pod itself more efficient, from pod-level auctions that cut A+E Global Media's inbound ad server requests by 84% to the live-event forecasting work now under comment. Standardized advanced formats were listed among the barriers to programmatic CTV at the IAB Tech Lab's August 2025 workshop. The format guidelines and signals have since been published, so that barrier is narrower than it was.

What remains open is the trade Sharon says can be avoided. If moving ads out of the break holds revenue while shortening pods, the evidence will have to come from publishers who try it and publish the results. A vendor's essay on a standards body's blog, however well argued, is not that evidence.

Timeline

Summary

Who: Aviram Sharon, co-founder and chief executive of InTheGame (ITG), a company that sells pause, double box, home screen and other CTV ad units and the software to trigger and serve them. The post was published on the website of IAB Tech Lab, the digital advertising standards body, under its member-perspective tag. It is addressed to streaming publishers and the buyers of their inventory.

What: An opinion post, "Rebalancing the Ad Break," arguing that streaming publishers should shorten ad pods and shift part of their ad load into six formats outside the break: home screen and menu ads, pause ads, overlays, squeezebacks and in-scene ads. It cites a FreeWheel study (71% of viewers bothered by unnaturally placed breaks, 14% lower brand recall) and illustrates a double box ad during an injury stoppage in a live match. It provides no revenue, pricing or viewer data for the non-break formats, counts the menu ad twice while omitting the screensaver format, and does not disclose that ITG sells the formats it recommends.

When: Published on October 29, 2026. It follows IAB Tech Lab's public comment draft of six CTV formats on December 11, 2025 and the finalization of their bid request signals on July 22, 2026.

Where: IAB Tech Lab's website, with reference to the IAB Tech Lab ad format guidelines for digital video and CTV on GitHub. The argument applies to connected TV services generally, with live sport as the main example.

Why: Streaming services carry most of their advertising in commercial pods, which is when viewers most often leave. Sharon argues that standardized formats outside the break let publishers cut pod length without losing revenue, add new inventory or sell sponsorships. Whether that trade holds has not been shown with published data, and one of the three levers he describes adds advertising rather than moving it.