Portland Leather Goods did not use AI to replace its social team. It used AI to make social media behave more like infrastructure.
Portland Leather Goods had the kind of problem most consumer brands used to treat as a cost of being popular. Its five-person social media team was dealing with roughly 70,000 monthly comments, direct messages, mentions, and tags across Instagram, TikTok, Facebook, YouTube, X, and Reddit, according to Business Insider. Some responses took as long as 48 hours. The brand was not short of attention. It was short of a workable system for turning that attention into service, sales, and learning.
The company found one in Nectar Social, a Silicon Valley startup founded by former Meta employees Misbah Uraizee and Farah Uraizee. The pair built Nectar around a simple observation from the social platforms themselves: more of the commercial action is happening in comments, creator videos, and direct messages, not only on product pages or in paid ads. For Portland Leather Goods, that meant a question under a TikTok Shop video could be as commercially important as an email ticket or a cart abandonment signal.
Business Insider reported that Portland Leather Goods began working with Nectar in July 2025, ran a one-month pilot in August, and moved into full deployment in September. The result was not full automation. Nectar scans incoming social activity, consolidates it in a dashboard, drafts replies, and gives the human team a chance to approve the response. Typical response times during business operations reportedly fell to just over one hour.
That is the useful part of the story. Not because a leather goods brand found a faster inbox, but because social media has become too operationally important to be handled like a stack of tabs in a browser.
Portland Leather Goods is an 11-year-old maker of handbags, wallets, passport covers, and other leather accessories. It has nearly one million followers across Facebook, Instagram, and TikTok, according to Business Insider, and a workforce of about 75 people. Those numbers are large enough to create real pressure, but not so large that the company can throw dozens of people at every channel.
The March 2026 test was a clean example. More than 4,000 TikTok videos mentioned Portland Leather Goods within 36 hours, driven by creators earning commissions by promoting and selling products on the platform. In the old workflow, that would have meant a brand trying to chase questions and comments across many creator posts after the moment had already moved on. With Nectar in place, the team could respond while the activity was still live.
There is a quiet shift inside that detail. A creator post is not just awareness. It is support. It is merchandising. It is attribution. If a customer asks about size, shipping, color, or returns under a video, the answer may decide whether the sale happens at all. The brand that waits two days has not just delivered poor service. It has missed the window where social attention still had commercial value.
Nectar is trying to make that window manageable. Business Insider reported that the startup plugs into Instagram, TikTok, Facebook, YouTube, X, and Reddit through data partnerships and APIs, with Reddit added most recently in April 2026. Its software can identify brand-specific conversations, draft responses, generate reports, and track when engagement leads to a direct sale. That is closer to customer relationship management than community management in the old sense.
The startup wedge is narrow, but real
Nectar had raised $10.6 million in venture funding by June 2025, according to Business Insider, with backing from investors including Google Ventures and True Ventures. In a separate March report on creator-economy startups to watch, Business Insider described Nectar as an agentic AI social commerce platform and noted that it helps brands analyze comments, DMs, and posts while managing communities of creators and fans.
The question is whether that becomes a standalone software category. There is a strong case that it can, at least for now. Shopify, Meta, TikTok, Salesforce, Zendesk, and marketing clouds all have reasons to own pieces of this workflow, but none naturally sees the whole messy surface area at once. A brand may sell on Shopify, advertise on Meta, field questions on TikTok, get mentioned on Reddit, and manage creator relationships somewhere else entirely. The pain lives between the platforms.
Startups often get their chance in exactly that space. Nectar is not asking a brand to move its store, replace its social channels, or rebuild its customer service department. It is sitting over the channels where the work already happens and making the fragmented labor visible. That is a practical wedge, especially for mid-sized consumer brands that cannot staff every channel like a global retailer.
The harder part will be defensibility. If TikTok Shop makes comment-to-purchase workflows more automated inside its own app, or if Meta turns Instagram DMs into a stronger commerce and support layer, brands may not want another vendor in the middle. Customer-service incumbents will also push further into social conversations because their clients already pay them to manage support queues.
Nectar’s answer appears to be breadth and judgment. Portland Leather Goods chose it partly because it covered many types of communication, from DMs to Reels comments to creator videos, and because it kept people in the approval process. That human layer matters. A social reply is public, brand-shaped, and often emotionally charged. A wrong refund answer is one problem. A wrong public reply under a viral video is another.
This is where enterprise AI is moving in many companies: away from the theatrical idea of replacing whole teams and toward smaller systems that sit inside daily work. Portland Leather Goods already used Google Gemini for meeting notes and to help speed marketing assets, according to Business Insider. Nectar went after a more exposed workflow, one where speed changes the business outcome.
For startups selling AI into enterprises, the lesson is plain. The strongest product may not be the one that talks most broadly about intelligence. It may be the one that finds a clogged operational channel, proves it can reduce the backlog, and leaves enough human control in place for the buyer to trust it.
Portland Leather Goods still has people approving responses. It still has to decide what its brand should say when customers are excited, annoyed, or confused. Nectar did not remove that work. It made the work possible at the speed social commerce now demands.
Also read: New research finds AI data center water fears are outrunning the actual science • Apollo's hunt for a second headquarters signals a permanent shift in where private equity capital calls home • Taiwan is using a $200 million data center to keep its last South American ally
