Multiply Raises $9.5M to Build Self-Learning Ads That Rewrite Themselves Weekly Using Sales Call Data

Every B2B marketing team knows the feeling. A campaign launches. The creative is sharp, the targeting feels dialed in, the first week of metrics looks promising. Then the slow rot sets in. Click-through rates slide. Engagement flatlines. The agency gets a call for a "creative refresh," and the whole expensive cycle starts over.

Multiply, a San Francisco-based startup, emerged from stealth on March 18 with $9.5 million in funding and a thesis that this cycle is structurally broken. The round was led by Mayfield, with participation from Sorenson Capital, Instacart co-founder Max Mullen, and Josh Woodward, Google's VP of Labs and head of Gemini, the executive behind NotebookLM.

Executives from HubSpot, Braze, Brex, Sierra, and Common Room also invested. For a seed round, that's an unusually dense roster of operational expertise in exactly the channels Multiply plans to disrupt.

The company was founded by CEO Matt Jayson, formerly at Google in user acquisition and later Head of Product at Brex, and CTO Ashish Warty, who spent five years as SVP of Product and Engineering at HackerOne and held senior engineering roles at Dropbox and Airship.

Their pitch: B2B companies are already sitting on the data they need to run dramatically better advertising. They just aren't using it fast enough.

The core insight isn't complicated. Sales call recordings, CRM pipelines, closed-won deal notes, and customer emails contain precise information about why buyers actually choose one product over another. The specific objections. The competitor comparisons. The language that resonates.

Almost none of it makes its way into ad campaigns quickly enough. By the time a traditional agency absorbs that intelligence and ships updated creative, the market has moved.

Multiply's platform plugs directly into those internal data sources and deploys a suite of five specialized AI agents to translate them into continuously improving campaigns on Google Search and LinkedIn:

The company reports early customers seeing 300% to 500% increases in pipeline from ad campaigns, according to Multiply's own data. That's a dramatic claim, though it comes without independent verification or sample size disclosure.

Multiply isn't the first company to promise AI-powered ad optimization. The martech landscape is cluttered with platforms claiming to automate creative, targeting, and bidding. What distinguishes Multiply's approach is the integration depth. Rather than optimizing ads based on platform signals alone, it's pulling from proprietary first-party data that competitors don't have access to.

The company positions itself not as pure software but as an "AI-native media agency" where human media buyers provide oversight, brand safety review, and strategic direction while the AI handles experimentation velocity. It's a hybrid model that acknowledges something the pure-play automation vendors often gloss over: brands still want a human checking the work.

Patrick Salyer, Partner at Mayfield and a Multiply board member, frames the opportunity in market terms.

That $50 billion figure comes from Mayfield's own framing and hasn't been independently cross-referenced. The broader B2B digital ad market is estimated between $30 billion and $60 billion depending on which segments you include, so the number is plausible but worth flagging as an investor's characterization rather than a neutral benchmark.

Several well-funded players have staked out adjacent territory. Mega, which raised $11.5 million in March 2026, is building AI agents to replace agency functions for B2B brands. Plurio, backed by $3.5 million in seed funding, targets performance marketing automation. Jasper and Writer focus on AI-generated marketing content at the creation layer.

But Multiply's closest analogue might be what a performance marketing agency does today, just compressed into software that learns on a weekly cycle instead of a quarterly one.

Here's where Multiply's roadmap gets genuinely interesting. The company says its infrastructure is already being positioned for ChatGPT advertising, the emerging format that OpenAI has signaled it intends to scale.

OpenAI began testing ads in February on ChatGPT for logged-in adults using the free and Go tiers, with initial advertiser buy-ins reportedly priced at a minimum of $200,000. Google remains open to advertising on Gemini. Anthropic has stated publicly it won't put ads on Claude.

Multiply's argument is that the same campaign learning systems built for search and social can extend into conversational ad formats as they emerge. If B2B buyers increasingly discover products through AI assistants rather than Google searches, the companies with systems already translating real customer language into ad copy will have a head start.

That's forward-looking to the point of speculative. Nobody yet knows what ChatGPT ad units will look like at scale, what the targeting options will be, or whether B2B audiences will engage with them the way they do with search ads. But the strategic positioning is clear.

A few things deserve scrutiny. First, the 300-500% pipeline improvement numbers lack independent validation. Self-reported metrics from a company announcing its launch should be taken as directional, not definitive.

Second, the "self-learning" framing, while compelling, depends heavily on the quality of the input data. Companies with messy CRM data, inconsistent sales call recording practices, or thin deal history may not see the same results as those with clean, abundant datasets. Multiply's model is only as good as the data it ingests.

Third, the hybrid agency model creates an inherent tension. As the AI gets better, the human oversight layer either becomes a bottleneck or a cost center that clients will eventually question. Multiply will need to define clearly where the human value persists as automation improves.

Multiply enters a B2B advertising market that's simultaneously consolidating around AI and fragmenting across new channels. The next twelve months will test several assumptions.

Can the company's agent architecture maintain quality as it scales beyond early design partners? Will the ChatGPT advertising ecosystem materialize in a form that rewards Multiply's infrastructure investments? And can a seed-stage startup compete for enterprise deals against incumbents like 6sense, Demandbase, and the major agency holding companies that are all racing to bolt AI onto existing offerings?

The investor roster suggests confidence. When Google's head of Gemini and executives from HubSpot and Braze write personal checks into a B2B ad startup, it signals something beyond standard venture enthusiasm.

Whether Multiply can convert that signal into sustained market traction is the question its second year will answer.