AI Search: Product & Packaging Moves

Product and packaging moves are the clearest signal of where an AI-search visibility vendor thinks its value sits. A pricing page tells you what a vendor charges; a shipping cadence tells you what it believes buyers will pay for next. Across the 14 companies tracked here, the last several months have been dense with launches that shift products from measurement toward action, from dashboards toward agents, and from single-surface tracking toward multi-model coverage.

For marketing and SEO teams evaluating vendors, this matters because the category is no longer a set of interchangeable rank trackers for AI answers. Packaging decisions now determine whether a tool reports on AI visibility, remediates it, or sells the remediation as a managed service. Reading the moves tells you which vendors are building depth in a narrow job and which are assembling suites. That distinction should drive shortlisting far more than feature-count comparisons.

Pattern name: From dashboards to agents and actions. The most consistent thread is vendors converting passive reporting into executable workflows. Profound launched Profound Ads in beta and rolled out a next-generation conversational agent builder, while Peec AI launched Peec Agent and expanded Actions for AI search. Goodie AI added Skills and Artifacts to its Goodie Agent, and Conductor introduced AgentStack for agentic AEO workflows. The interpretation is straightforward: buyers have had enough time to look at AI visibility charts and now want the tool to do something about the gap. Vendors are packaging agency-like labour into software, which raises the stakes on output quality and makes evaluation about trust in automated action, not just data coverage.

Pattern name: Attribution and revenue proof are being pulled into the core product. Several vendors are moving beyond mention tracking toward connecting AI visibility to traffic and money. Scrunch AI expanded AI Referrals with GA4 engagement and conversion data, Peec AI shipped AI Referrals to track traffic from AI assistants, and Brandlight lists Impact Tracking and Attribution as a platform product. Evertune launched AI Retargeting Segments in The Trade Desk and DV360, and its ChatGPT Ad Agent pushes further into paid activation. This is a response to a familiar budget question: marketing leaders want to know whether AI visibility changes pipeline. Packaging attribution as a first-class feature signals that vendors expect to be judged on commercial outcomes, and it means SEO teams should test whether referral and conversion data actually reconciles with their own analytics.

Pattern name: Model and surface coverage is becoming a packaging battleground. Vendors are competing on how many AI surfaces and models they cover, and how visibly they expose that coverage. AthenaHQ expanded its Starter tier model coverage to 11, up from 9, and shipped a feature drop including Shopify Shopping Pages and ChatGPT Ads reporting. Goodie AI added Muse and Instinct to tracked AI models, AirOps added Codex and Claude Code tracking, and Peec AI opened a beta for AI Shopping Tracking plus SKU-level product tracking. Ahrefs Brand Radar added entity extraction and AI response visibility inside Site Explorer. The signal is that AI search is fragmenting across assistants, shopping surfaces, and coding tools, and buyers increasingly treat breadth of coverage as a proxy for reliability. The risk is that coverage counts become a vanity metric unless the vendor can show depth on the surfaces that matter to a given brand.

Pattern name: Packaging is splitting between self-serve software and done-for-you service. The category is differentiating on delivery model, not just features. GetIntel added a done-for-you managed service at $1,000/Month, which explicitly sells outcomes and labour rather than seats. At the other end, Knowatoa added period-over-period comparison to its MCP analytics tools and exports competitor sentiment data to NinjaCat, Looker and BI tools, positioning itself as plumbing inside existing stacks. Profound operates an Agents Marketplace with over 50 workflow templates, and Ahrefs doubled API and MCP limits in a roundup of 26 product updates. This split matters because it maps to team maturity: lean teams may prefer managed service, while in-house analytics teams want exports and APIs they can wire into their own reporting.

What this means for marketing and SEO teams evaluating AI-search visibility vendors: Shortlist by the job you actually need done, not by feature count. If you need remediation and content output, scrutinise the agent and action features from Profound, Peec AI, Goodie AI and Scrunch AI for quality control and reversibility. If you need to prove impact, test attribution claims from Scrunch AI, Peec AI, Brandlight and Evertune against your own GA4 and CRM data before committing. And decide early whether you want software, managed service, or data piped into your BI stack, because vendors are increasingly packaging toward one of those three models.