Marklo AI Marketing Startup Raises $5M, Opens Shopify Beta

A seasoned operator returns with an AI startup

Jessica Herrin built Stella & Dot into a social-selling company that sold more than $2 billion in products. Her new company, Marklo, has closed a $5 million seed round led by Forerunner, with Freestyle Capital also investing.11 Angel investors include former Klaviyo executives and current Shopify leaders.11 The money arrived alongside a public beta that puts Marklo's software in front of Shopify merchants before Black Friday, Cyber Monday and the rest of the holiday rush.1213

The round is small next to the huge checks going to AI model builders. Still, it is a useful example of how a founder with retail experience is shaping an AI product, a business model and a launch plan. Marklo is not pitching a tool that generates more ad copy. It is betting that consumer brands need help deciding what to promote, and that AI is most valuable when it makes that decision using a brand's own sales, inventory and campaign history.

What Marklo actually does

Marklo pulls data from Shopify and from marketing services including Klaviyo, Attentive, Meta and Google. It then lays out past launches, promotions and sales on a single calendar.11 The company says it also connects to Slack, Canva and Figma. Teams get one workspace that combines the calendar, a campaign canvas, a task list and analytics.1314

The software looks backward first. Once connected, it rebuilds a brand's email, SMS, paid media, social activity and revenue day by day, with year-over-year comparisons.12 Herrin told Crunchbase News that this lets teams see things they would miss looking at each tool separately. One example: whether a jump in sales credited to paid ads actually happened at the same time as a product launch or a viral social post.11

Next come recommendations. Marklo adds competitor promotions and social media trends to the brand's own history, checks what is in stock, and suggests what to market next and when.11 Every week it reviews sales, inventory and the existing calendar, then sends proposed campaigns through Slack or email with its reasoning and the products involved.12 Herrin gave one example: the software noticed a color trending on social media, matched it to a handbag the retailer had in stock but hadn't promoted lately, and offered to plan a post modeled on one that had done well before.11

The company stresses that people stay in charge. Marketers review, edit, assign and approve everything Marklo proposes, and the company says it does not use customer data to train AI models.1114

The product thesis: decisions, not content

The most notable part of the launch is how Marklo positions itself against the rest of the AI marketing market. "Every tool in a brand's stack is getting AI, but each one only sees its own corner of the business," Herrin said in the launch announcement.13 The company describes itself as working across the whole marketing stack, unlike AI tools built to improve a single channel.14

Forerunner's Kirsten Green made the same argument. She told Crunchbase News that AI already makes producing marketing content easier, so the bigger opportunity is helping brands decide what to promote, understand the effect on margins and grow in ways that fit their identity.11 "It's one thing to collect the information; it's another thing to understand how to present it and how to recommend an action against it," she said.11

This is the right bet. AI-generated copy and images are quickly becoming cheap and widely available. Marketing tools that only generate content will likely end up competing on price, especially as the platforms brands already pay for add the same features. A layer that connects inventory, past promotions and outside trends is harder to copy, because it depends on integrations, customer trust and judgment about retail. Marklo's "agentic" label is marketing language, but there is a real product decision behind it: the system starts the work instead of waiting for someone to type a prompt.1215

The obvious risk is that Marklo relies on the platforms it connects to. Shopify, Klaviyo and Meta are all adding their own AI features, and any of them could try to build a similar cross-channel view. Having Shopify and Klaviyo people among the angels, plus listings in the Shopify App Store and Klaviyo's integrations directory, suggests Marklo wants to be a partner to these platforms rather than a competitor.1113

Revenue model and the pre-revenue question

Marklo plans to charge a monthly subscription plus fees based on usage. For now, brands can install and try it for free.11 This combination is becoming common for AI software. Usage fees help cover the variable cost of running models, and the flat subscription gives predictable baseline revenue.

The coverage disagrees on Marklo's current revenue. Crunchbase News reports that the company is pre-revenue, with more than 100 brands using the software in a pilot.11 A GetLatka profile lists an estimated $330,000 in 2025 revenue and a three-person team.18 That figure is explicitly an estimate. It also conflicts with Crunchbase's on-the-record reporting of an 11-person team that includes former Shopify and Google engineers.11 The reported interview is the stronger evidence. The safest reading is that Marklo has not yet started charging for its product in any meaningful way. Smaller details also differ. Crunchbase places the company in Burlingame, California, while the funding announcement and a FinSMEs writeup describe it as San Francisco-based.1116

The pilot customers are a credible start: Away, Clare V., Monica + Andy, Maya Brenner and Latico Leathers.11 Herrin has described her pilot brands as businesses with seven-, eight- and nine-figure revenues.3 The main result shared so far is that Latico Leathers planned a month of marketing in one day, a job that used to take most of a week of meetings, spreadsheets and five different tools.11 As one outlet noted, that shows time saved, not higher sales.15 Marklo needs to prove the second to justify a subscription.

Founder strategy: from bootstrapping back to venture

Herrin's history with fundraising matters here. She raised venture capital for her first company, which became WeddingChannel.com, and later concluded it had raised too much. She built Stella & Dot with about $2 million in capital and a focus on profitability.11 In a LinkedIn comment, she wrote that she "over capitalized" her first company and later bootstrapped the second, summing up her view as "Scarcity = profitability = growth capital."3 In a 2012 Stanford interview, she called the heavy venture funding at WeddingChannel her biggest failure and said she didn't want to repeat "glossy, expensive mistakes."7

That background makes a $5 million seed round look deliberate. She told Crunchbase she raised money to hire engineers and build faster, and that she wanted investors with commerce experience.11 The amount covers engineering, the beta and customer success programs.13 It is not large enough to drift away from her focus on profitability.

The company also grew out of her own business. When COVID-19 stopped in-home trunk shows, Stella & Dot's sales fell about 80% almost overnight. Herrin realized that women who might once have joined as sellers could now become influencers or open their own Shopify stores.11 She sold the businesses and brands separately, then bought back the Stella & Dot brand to sell through a different channel and to test Marklo on.11 She has since posted that Stella & Dot is "growing again" and that she is hiring an e-commerce general manager for it.3 Having her own brand as a test customer is a real advantage, and Green said it was central to the investment: Herrin "has been the customer that she's building for."11

Go-to-market timing and a tighter funding climate

The launch timing is a clear go-to-market choice. Opening the beta just before the holiday season puts Marklo in front of merchants when planning pressure peaks and when last year's sales data is most useful.13 Distribution through the Shopify App Store also matches Herrin's stated goal of reaching "every single brand that runs on Shopify" so smaller merchants can "operate like a big brand."1113

The funding environment is less friendly. Crunchbase data shows startups in sales, marketing and CRM raised $7.5 billion across 830 rounds globally this year through mid-September, with most of it going to AI companies. Funding is on pace to come in below 2025's $11.1 billion, and the number of deals may fall for a fourth straight year.11 Investors are making fewer, bigger bets in this category. A founder with a track record, a launch customer she owns and a pitch built around decisions rather than content is the kind of company that still gets funded.

The holiday season will be Marklo's first public test. If brands using the beta can show that its suggestions sold more products or improved margins, not just sped up planning, Marklo will have a strong case for charging subscriptions. If the main benefit stays faster planning, it will be competing with every platform adding AI to its own dashboard.