Google on September 28, 2026 folded vertical video from multiple publishers and channels into a single programmatic buying workstream inside Display & Video 360, pointing to an early campaign in which Unilever's mayonnaise brand and its agency reached 24% more individual users while paying 25% less per person. The company attached no availability date, market list, publisher roster or test methodology to the figures.

In Short

Google has made it possible to buy the tall, phone-shaped video ads that run across many different apps and websites in one place, instead of buying each one separately. That matters to brands and agencies because separate buys often show the same ad to the same people again and again while others never see it. In one early test a mayonnaise brand reached more people for less money per person, but Google has not said which apps and sites are included or when every advertiser can use it.

What Google published

The product was set out in a four-paragraph post on Google's Marketing Platform blog dated September 28, 2026. It carries the byline of Marta Martinez, listed as Managing Director, Google Marketing Platforms. When she presented at Google's NewFront event at Pier 57 in New York on March 23, 2026, PPC Land identified her as Managing Director for the Americas at Google Marketing Platform. The September byline drops the regional qualifier; the post does not explain the change.

Google calls the product Vertical Video Unification. According to Google, it makes it possible "to buy and measure vertical video programmatically across channels in one platform." That platform is Display & Video 360, the company's demand-side platform. The post adds that the solution "centralizes vertical video buys from publishers across vertical video into a single workstream." It is a circular sentence - publishers "across vertical video" - and it names none of them.

Two further claims complete the description. Gemini, Google's family of AI models, powers "buying, execution, and full-funnel measurement," according to the post. Display & Video 360 also delivers competitive campaign performance "without overexposing users to the same ads," which Google describes as "benefits that siloed buying strategies can't replicate."

What is missing matters as much as what is there. The post contains no statement on availability, so it is unclear whether the product is open to every account, in beta, or restricted to selected advertisers. It lists no markets, no pricing and no minimum spend. It does not say how the product appears in the interface - as a new line item type, as a deal package, or as a setting on existing video line items - and it links to no Help Center documentation. A quote card in the post's carousel, attributed to a named executive, is rendered as an image and is not legible in the copy of the page reviewed for this article, so it is not quoted here.

The test figures

Every number in the announcement comes from one campaign. "During initial testing, Unilever's Hellmann's and WPP Media achieved a 24% increase in unique reach while reducing cost per unique user by 25%," according to Google. WPP Media is the planning and buying arm of WPP, renamed from GroupM in May 2025.

These are vendor-supplied results from a single advertiser, and the post does not say what they were measured against. The phrase "siloed buying strategies" implies a comparison with separately purchased placements, yet Google gives no test design, no duration, no market, no budget and no count of publishers or channels on either side. Nor does it say whether reach was measured by Google's own modelling inside Display & Video 360 or by an independent party.

The two percentages also interact in a way the post leaves unexplained. Cost per unique user is spend divided by the number of individuals reached. Had spend been held constant, a 24% rise in reach would by itself lower that cost by roughly 19%, because 1 divided by 1.24 is about 0.81. A 25% reduction therefore implies that the test campaign spent around 7% less than whatever it was compared with - 1.24 multiplied by 0.75 gives 0.93 - or that the two figures were calculated against different baselines. Neither reading is implausible. They do, however, describe different outcomes: more people for similar money in one case, more people for less money in the other.

Nor is it clear whether "cost per unique user" corresponds to an existing platform metric. Display & Video 360 added a Cost per reach metric in August 2025, calculated as total media cost divided by impression reach. The names differ and the post does not connect them.

The framing - more people rather than more exposures - echoes a position argued inside Google earlier in the year. In February 2026, Pablo Perez and Jesús Martín Calvo of Google's Marketing Research & Insights team argued that the first exposure to an advertisement carries the greatest value and that consumer behaviour supports maximising reach over repeated contact. In the trade's vocabulary, the Hellmann's claim is essentially about incremental reach: people the campaign added, as opposed to people it reached more often. Google's post does not use the term, and it offers no overlap data that would let a reader check it.

How one workstream changes the mechanics

Frequency across sellers

The technical case for pooling vertical buys is a counting problem. When a brand purchases vertical placements from several sellers through separate deals or separate platforms, each purchase keeps its own exposure counter. A frequency capping rule of three impressions a week then applies per buy rather than per person, and the same viewer can hit the cap several times over. Inside Display & Video 360, caps sit at line item, insertion order or campaign level, with the tightest applicable cap winning.

Routing every vertical purchase through one workstream lets one set of counters see all of them. That is what Google appears to mean by not overexposing users, at least on paper. It depends on deduplication, the step of recognising that impressions served on two different apps or two different sites belong to the same individual. The post does not describe how Google does this for the new product. On Google's own properties, signed-in signals carry much of the load; across third-party apps and publisher sites the same task rests on device identifiers and modelling, where coverage is patchier.

Display & Video 360 already has reporting that exposes overlap. In April 2026 Google added a campaign-level dimension to Unique Reach Overlap reports, describing it as a way to refine frequency caps and move budget away from redundant reach. Google has also made the frequency argument with an advertiser on stage before. At the March NewFront, ESPN said that unifying its media in Display & Video 360 lifted reach by 28% and saved more than $1 million through frequency capping - self-reported figures, presented at a Google event, much like the Hellmann's numbers.

Where Gemini sits

Google credits Gemini with buying, execution and measurement without saying what the model decides. At NewFront, the company presented Gemini as the engine behind bidding across its marketing platform and made Ads Advisor, its Gemini-based campaign assistant, generally available to Display & Video 360 users. PepsiCo told the same audience that Video Reach Campaigns 2.0, an AI bidding strategy that shifts spend across YouTube formats including Shorts in real time, had produced higher reach and lower CPMs in its tests before it scaled the approach across its divisions.

The September post supplies nothing comparable. It names no bidding strategy, describes no manual controls and says nothing about whether advertisers can see how budget was split between sellers. For a product sold partly on measurement, that last gap is not trivial: an allocation that cannot be inspected cannot be tested.

What "across channels" might cover

The post never defines a channel. Within Google's own inventory, YouTube Shorts is the largest source of vertical supply; at NewFront Google put Shorts at 200 billion daily views and confirmed the inventory was accessible in Display & Video 360. Programmatic Shorts buying in the platform moved to Demand Gen line items after Google stopped the creation of new YouTube video action line items from March 2025, with Demand Gen supporting vertical video in-feed, in-stream, on YouTube Search, on Shorts and on Google video partners. Beyond YouTube itself, a significant share of Google video partner inventory is vertical-focused, gaming and video apps in particular.

Outside Google, vertical supply has multiplied over the past year. JWX, formerly JW Player, released a swipeable vertical video product for publisher sites on February 19, 2026, installed with one line of JavaScript; Connatix, now part of JWX, took the JWX name in Display & Video 360 reporting on September 21, 2026. Media.net paired its own vertical product, Bytes, with a survey of more than 1,000 US consumers in which 90% said they wanted vertical video across publisher sites - a finding from a company with a commercial interest in the format. Disney began rolling out Verts, a vertical feed in the Disney+ mobile app, to US subscribers on March 12, 2026, and NBCUniversal said Peacock's vertical video format would open to advertisers in 2026.

The largest pools of vertical inventory are the hardest to reach this way. Instagram Reels and TikTok's main feed are generally bought inside Meta's and TikTok's own tools, which is how most social video is sold. Google's post does not say whether any of that supply sits inside the unified workstream, and nothing in it points to a new arrangement with either company. If the scope is limited to YouTube, Google video partners and publishers already trading through Display & Video 360, the change is one of management - one workstream in place of several - rather than access to new supply. Google has not said which it is.

Timing inside the DV360 roadmap

The product arrives during a crowded migration window for video buyers on the platform. Google's Q3 2026 roadmap for Display & Video 360, last updated on August 21, 2026, sets October 26 as the date when remaining legacy maximum cost-per-view and Video reach 1.0 line items, active or paused, are archived automatically and stop serving. Active campaigns have to be moved by hand to Video reach 2.0 and Video view line items. Teams rebuilding video structures over the next four weeks will be doing so while the vertical workstream's own setup remains undocumented.

Brand suitability shifts in the same window. From October 1, 2026, inventory mode and content theme exclusions cover all YouTube and programmatic inventory in the platform, and sensitive categories and digital content labels are deprecated for programmatic buys in favour of three tiers - Maximum, Moderate and Limited - and ten content themes. Any third-party vertical placements pulled into the new workstream will therefore be governed by the tiered model from the outset. YouTube's implementation of that model received Media Rating Council brand safety accreditation in June 2026, covering in-stream and Shorts inventory bought through Google Ads and Display & Video 360. The post does not say how vertical placements from other publishers are classified.

Why this matters for the marketing community

Vertical video is no longer a single platform's format, and the money around it is large. IAB projections put 2026 US social video spend at $31.9 billion against $29.3 billion for CTV, with online video at $20.7 billion and total digital video above $80 billion. Much vertical viewing sits inside the social figure; the publisher players and streaming feeds described above sit in the other two. A brand that wants to fund vertical as a format, rather than platform by platform, has had to stitch purchases together and reconcile reach by hand.

Google's answer is to do the stitching inside its own buying tool. That fits a pattern visible across 2026. In March the company moved YouTube Mastheads and Pause Ads into Display & Video 360 and added Kroger to its Commerce Media Suite. Albertsons Media Collective joined the suite on April 27, 2026, and Walmart Connect routed its purchase data into Display & Video 360 for YouTube campaigns on June 11, 2026. Each step brings more inventory, data or workflow into one platform, and each is measured largely by that platform.

The scale of that platform is not incidental. Guideline data discussed on May 6, 2026 put four buying platforms at roughly 85% of global programmatic spend in the first quarter of 2026, up from about 75% in 2022, with Display & Video 360 alone near 41%. A product that makes pooling vertical buys easier inside the largest of them adds another reason to concentrate spend there, and the reach figures that justify the move are, for now, Google's own.

Rival platforms are making the same kind of argument about their own vertical placements. TikTok's TopReach, which merges its TopView and TopFeed slots into a single booking, was pitched with internal 2025 data claiming 59% incremental reach and 3X lower cost per reach than TopView alone. Combined buying, fewer duplicated exposures, cheaper reach: the claim is converging even as the inventory behind it remains divided between companies that do not share identity data.

For agencies, the arrangement cuts two ways. WPP Media is the only agency named in the post, supplying the proof point for a tool that moves the orchestration of cross-publisher vertical buys from an agency trading desk into Google's software. Whether that saves agency labour or narrows agency bargaining power depends on details Google has yet to publish.

Several of those details decide how much the product changes. Which publishers and apps are inside the workstream, and are Reels or TikTok among them? Is the product generally available, and in which markets? Can reach and overlap be verified by an independent measurement provider rather than by Google's reporting alone? And how does the 25% cost figure reconcile with the 24% reach figure? Until Google answers those questions in documentation rather than a blog post, the Hellmann's result is a single, self-reported data point.

Timeline

Summary

Who: Google, through its Display & Video 360 buying platform, in a post bylined by Marta Martinez, Managing Director, Google Marketing Platforms. The only named test participants are Unilever's Hellmann's brand and WPP Media. The product affects advertisers, agencies and programmatic teams buying vertical video across YouTube, Google video partners and third-party publishers.

What: Vertical Video Unification, which Google says lets advertisers buy and measure vertical video programmatically across channels in one platform, centralising purchases from multiple publishers into a single workstream, with Gemini powering buying, execution and measurement. In initial testing, Hellmann's and WPP Media recorded a 24% increase in unique reach and a 25% lower cost per unique user, according to Google, with no methodology, baseline or duration disclosed.

When: The post is dated September 28, 2026. Google gave no availability date. The product arrives weeks before Display & Video 360's October 1 brand suitability change and its October 26 archive of legacy video line items.

Where: Display & Video 360. Google named no markets, publishers or channels, and did not say whether social platforms such as Instagram Reels or TikTok fall within scope.

Why: Google argues that pooling vertical video buys in one platform avoids showing the same ad to the same people too often, which separate buys cannot prevent. For the market, the product extends a 2026 pattern of Google drawing inventory, retail data and workflow into Display & Video 360, while the evidence offered for the reach gains remains self-reported.