Kochava today said StationOne Workspaces are available for Google, Meta, Reddit, Snap and TikTok for Business, routing campaign build, bulk edits, audience work and reporting through a chat interface wired to platform connectors including Meta's Ads MCP Server and the TikTok Ads Model Context Protocol Server.

The announcement went out at 09:30 Eastern Time on Thursday from Sandpoint, Idaho, distributed through GLOBE NEWSWIRE. It extends StationOne, the orchestration layer Kochava first shipped as a desktop application in late 2025, from a general-purpose workspace for large language models into a set of destination-specific surfaces aimed squarely at performance advertising operations.

According to Kochava, the Workspaces combine guided AI workflows with curated connectors so teams can create, manage, optimize and govern campaigns across accounts while reducing context switching and standardizing operational best practices. Five platforms are named. Two connectors are named. Nothing else in the release identifies which capabilities are live on which platform, when each Workspace became available, or what the product costs.

What the release specifies

The announcement groups the functionality into five categories.

Campaign build and launch workflows cover the creation and management of core campaign structures and settings through guided steps intended to reduce setup time and rework. Bulk changes apply updates across campaigns and accounts, with pacing-related adjustments and status changes given as the examples. Audience and creative operationsare described as repeatable workflows for audience management, creative organization and quality assurance, and performance review without constant tool switching. Reporting, diagnostics and change history pull reporting outputs, troubleshoot delivery problems and maintain visibility into actions taken. Governed collaboration at scale covers role-aware execution and standardized workflows across agencies and multi-team organizations.

Kochava frames the underlying problem as an operational burden that repeats regardless of channel: launching and updating campaign structures, refreshing audiences, coordinating creative quality assurance, pulling reports, troubleshooting delivery, and managing permissions and billing. The company lists ecommerce and retail, gaming, subscription apps, marketplaces and lead generation as the use cases where faster iteration and fewer manual steps could translate into measurable improvements in speed-to-launch and optimization cadence.

That last formulation is conditional. No throughput data, launch-time reduction, error-rate figure or customer count accompanies the release.

"As AI becomes part of day-to-day marketing operations, teams need a platform that scales across a diverse partner set, not one-off automations tied to a single destination," said Charles Manning, Founder and CEO at Kochava. "StationOne is built for enterprise-grade governance and privacy objectives, while giving organizations a consistent operating layer across Google, Meta, Reddit, Snap, and TikTok for Business so teams can move from insight to execution faster, with clear guardrails, as budgets, accounts, and partner mixes grow."

David Kaufman, Global Head of Monetization Product Partnerships at TikTok, is the only platform executive quoted. "Advertisers and agencies want simpler ways to execute campaign strategies and workflows," Kaufman said. "TikTok is excited to partner with Kochava on practical and impactful AI that will streamline operations for performance marketing teams and improve results."

No representative from Google, Meta, Reddit or Snap appears in the release, and the announcement does not describe the Workspaces as joint products with those companies.

The connector layer this depends on

The two connectors named in the release are the visible part of a build-out that has been running for more than a year. Model Context Protocol originated at Anthropic in November 2024 and was later donated to the Linux Foundation. It defines a consistent way for AI systems to query external tools and data without bespoke integration code for each provider.

Advertising platforms adopted it in sequence. Google's Ads API team released an open-source MCP server on October 7, 2025, limited to read-only reporting and diagnostics. Amazon Ads ran a closed beta from November 13, 2025 and moved to open beta on February 2, 2026. Meta went further than either: its Ads AI Connectors and companion command line interface, announced on April 29, 2026, carried write access from the first day, letting external agents create and edit ads, ad sets and full campaigns. On July 16, 2026, Meta opened that same ads MCP server to any developer holding a Meta app, removing the custom integration step that had previously stood in the way.

TikTok launched its Ads MCP server and developer toolkit at TikTok World on May 13, 2026, then packaged agent access into a product surface with Agentic Hub, which shipped on July 28, 2026 alongside seven other product sets. Snap announced its own MCP server on June 18, 2026 as part of a wider AI suite spanning conversational campaign setup and agentic Dynamic Product Ads, exposed through the Snapchat Marketing API.

Reddit is the outlier in the group. Its 2026 advertising development has run through AI-assisted creative and shopping tools presented at Cannes Lions on June 22, 2026, not through a publicly documented agent protocol server. The Kochava release does not say what the Reddit Workspace connects through, and no Reddit MCP endpoint has been disclosed in the way Meta, TikTok and Snap have disclosed theirs. The gap is worth noting because the value of a governed operating layer depends on what each underlying connector actually permits.

That variance is not incidental. Governance postures across advertising MCP implementations diverge sharply. Google and Amazon launched read-only. Meta launched with write access. Adform shipped 29 read-only skills for its FLOW DSP in July 2026, with the read-only limitation stated repeatedly as a design choice. A chat instruction that edits a Meta campaign directly may, on another platform, produce nothing more than a report.

Where StationOne sits in Kochava's product line

Kochava introduced StationOne on November 25, 2025 as a desktop client for Windows, macOS and Linux, connecting Anthropic's Claude, OpenAI's GPT, Meta's Llama and custom models through one interface. Manning described the target problem then as tech fatigue among marketers juggling AI-backed tools that force work into silos.

The platform has accumulated layers since. Atlas Performance, a supply performance system for premium publishers, arrived on February 11, 2026. On March 25, 2026, Kochava removed the invitation code requirement and opened StationOne to public beta with an IAB Tech Lab AAMP workspace containing 19 agentic skills across eight functional areas, running on the IAB Tech Lab reference implementation MCP server. In July 2026, the company's quarterly bulletin detailed the Kochava AI Workspace, which connects a customer's own account data through the Kochava MMP Connector and is offered at no additional cost to Kochava customers.

Today's release does not state whether the platform Workspaces follow that pricing model.

The named platforms also overlap with Kochava's certification roster. The Partner Certification Program launched on December 15, 2025 with Meta, Google Ads, Snapchat, TikTok, Liftoff and YouAppi as its initial members, and added six more organizations on March 3, 2026, including LG Ad Solutions, InMobi and Digital Turbine. Four of the five Workspace platforms were already certified partners.

Governance is the claim under the most pressure

Manning's quote leads with enterprise-grade governance and privacy objectives, and the release repeats variations of the theme: clear guardrails, role-aware execution, accountability with visibility into actions taken. Change history is listed as a feature rather than a compliance artifact, but audit trails are precisely what regulators have been asking about.

Four United Kingdom regulators warned in April 2026 that agentic systems are already operating in production and that oversight frameworks are needed immediately, citing action bundling, where a single agent instruction triggers several consequential steps at once, as a risk specific to this class of software. Usercentrics acquired MCP Manager on January 14, 2026, extending consent tooling into agentic workflows as EU AI Act enforcement began. Security researchers flagged validation weaknesses in MCP implementations as early as July 2025.

The protocol itself has moved underneath its adopters. Its largest specification rewrite since launch shipped on July 28, 2026, removing sessions from the transport layer entirely, hardening OAuth handling and formalizing an extensions track. Vendors running sticky-session infrastructure face rework.

Kochava's own regulatory history is part of the record here. The Federal Trade Commission closed a nearly four-year case against the company with a stipulated order signed on May 4, 2026, barring Kochava and its subsidiary Collective Data Solutions from selling or disclosing precise location data without affirmative express consent tied to a consumer-requested service. The Commission vote was 2-0, and the order runs for ten years from entry.

The efficiency claim meets the efficiency data

The commercial argument for chat-driven operations rests on time saved. Recent industry research complicates it.

Typeface's Signal Report, published on June 22, 2026 from a survey of more than 200 marketing leaders at vice president level or above, found campaign timelines lengthening rather than shortening despite AI adoption. According to Typeface, 93 percent of respondents reported increased pressure to move faster, 92 percent said campaigns now require ten or more stakeholders, 34 percent needed one to two months to ship a campaign, and 16 percent described their organization as prepared. Content generation was not the bottleneck; approval was, with 88 percent saying teams can produce quickly but struggle to get sign-off.

The cost side has drawn similar findings. KPMG research covered on August 8, 2026 documented that 49 percent of organizations scaled back agentic AI rollouts when costs outran value, alongside TransUnion data from August 5, 2026 showing 53 percent of senior United States marketing leaders reporting meaningful return on AI investment while 89 percent expected to increase spending anyway.

None of that invalidates the Kochava product. It does establish the standard against which it will be judged: whether standardizing execution across five destinations reduces coordination overhead, or adds another layer to a stack already accused of adding layers.

What advertising teams gain and what stays unresolved

For agencies running multi-platform performance accounts, the practical appeal is narrow and real. Bulk status changes, pacing adjustments and reporting pulls across Google, Meta, Reddit, Snap and TikTok currently require separate interfaces, separate permission models and separate export formats. A single governed surface that executes those tasks through instruction rather than navigation removes clicks from a workflow that is mostly clicks.

The unresolved questions are equally concrete. Whether each Workspace supports write operations or stops at reporting is not stated. Whether the Reddit Workspace runs on a protocol connector or a conventional API integration is not stated. Pricing, availability timing, supported account structures and the handling of billing permissions are absent from the release. The Trade Desk's closed beta placing Claude inside campaign creation, reported in March 2026, surfaced the same underlying issue from the trader's side: expertise displaced at the creation stage is harder to audit than expertise displaced at optimization.

The industry has also been working on the vocabulary problem that sits beneath all of this. IAB Tech Lab research published in April 2026 argued that taxonomy identifiers, not natural language, are what keep multi-agent advertising workflows deterministic. A chat instruction that means one thing to a media buyer and another to a model is a governance problem before it is a performance problem.

Kochava is not claiming to have solved that. The company is claiming a consistent operating layer across five destinations, built on connectors two of those destinations have already published. Whether performance teams treat it as consolidation or as one more window remains an open question, and the release offers no metrics to settle it.

Timeline

Summary

Who: Kochava, the Sandpoint, Idaho measurement and data solutions company led by founder and chief executive Charles Manning, with David Kaufman, Global Head of Monetization Product Partnerships at TikTok, the only platform executive quoted in the announcement.

What: StationOne Workspaces for Google, Meta, Reddit, Snap and TikTok for Business, bringing chat-driven execution to campaign build and launch, bulk changes, audience and creative operations, reporting and diagnostics, and role-aware collaboration. The Workspaces use curated connectors including Meta's Ads MCP Server and the TikTok Ads Model Context Protocol Server. No pricing, availability dates, write-access scope or performance data accompany the release.

When: August 13, 2026, at 09:30 Eastern Time, distributed through GLOBE NEWSWIRE.

Where: Across the five named advertising platforms, through the StationOne application Kochava launched in November 2025 for Windows, macOS and Linux, with the company headquartered in the United States and operating offices globally.

Why: Performance teams repeat the same operational tasks across every channel, and each platform has been opening its own agent access on its own terms and its own timetable. Kochava is positioning a single governed layer over that fragmentation. The claim runs against recent research showing AI adoption lengthening campaign timelines and nearly half of organizations pulling back on agentic rollouts over cost, which makes measurable operational gain, rather than connector coverage, the test this product will face.