influencer marketing

Influencer Hero has launched a new version of its Influencer Finder, the creator discovery tool inside its platform, now powered by a proprietary search engine built in-house that indexes roughly 100 million Instagram profiles. The company says the move brings AI-led research to the most manual part of influencer marketing: finding the right creators.

Most of the industry searches the same way. The company estimates that around 95% of influencer marketing tools run on the same handful of third-party data sources, and those sources index what a creator writes about themselves: the bio, plus a keyword or two. What a creator actually posts never enters the search. Influencer Hero's own previous search worked the same way. Its new engine turns that around. An AI has reviewed the bio and the recent posts of every profile, so each of the four search methods – including the profile search that replaces the old keyword lookup – matches a query against what the creator publishes, not just how they describe themselves.

The new Influencer Finder adds four capabilities: a profile search that reads a creator's full profile and recent content rather than bio keywords; a visual search that finds creators based on the content they post; stacked lookalike searches that work from up to ten reference creators, or an entire campaign board, at once; and a 'deep research' agent that plans up to 20 searches and then reviews every candidate one by one against a written brief.

Why Discovery Is Still the Bottleneck

Most discovery tools search bios and hashtags. A brand looking for vegan personal trainers in the US gets anyone who has typed 'vegan' into their profile, and the team then scrolls through hundreds of accounts by hand to check whether the content actually matches.

The issue is not going away. Influencer Marketing Hub estimates the global influencer marketing industry reached $32.55 billion in 2025, nearly double its 2022 size, and its 2026 Benchmark Report found that creator discovery is now the most common use of AI among influencer marketers, cited by 36.67% of respondents. Brands are spending more and expecting software to do the sifting.

Like most platforms in the category, Influencer Hero had built its search on top of licensed data, which limited which filters it could add and how precisely it could search. The new proprietary search engine gives the company room to ship new filters and tools directly, including the multi-keyword and brand-safety filters users had been asking for.

1. Profile Search

A vision-enabled AI has reviewed every profile in the index -bio, profile image and the last 12 posts- so a query is matched against the whole profile rather than a keyword list. Searching 'vegan personal trainers in the US with more than 25,000 followers' returns only accounts whose recent content combines both, and the images that matched the query surface first, so a reviewer can confirm fit without scrolling.

Keyword filters have also been reworked. The previous version required every keyword to match, so each term added narrowed the results and building a large list meant running one search per keyword. Searches can now match any of several keywords, so 'cooking', 'recipes' and 'kitchen' expand the list in a single query, while must-match and exclusion filters remain available to tighten it.

2. Visual Search

Users type a description -'chef making sushi', 'morning running selfie', 'woman holding a bulldog'- or drop in a reference image, and the engine returns creators whose posts match the scene, lighting and style. The company positions it for brands with a very specific visual identity, and as the clearest break from keyword-based discovery.

3. Stacked Lookalikes

Lookalike search previously worked from one creator at a time. It now takes up to ten reference handles, or an entire campaign board, in a single run. In 'expand' mode it widens the list and ranks creators who resemble several references higher; in 'refine' mode it isolates the common thread between imperfect examples- three creators who each post about a niche sport half the time produce a list focused only on that sport. A brand with 500 approved creators on a board can generate the next 500 in one step.

4. Deep Research

This is the headline feature, which the company compares to the research modes in consumer AI assistants. A user writes a plain-language brief -for example, 'find perfume creators in Germany who compare or review fragrances, exclude anyone with their own perfume brand, exclude face tattoos for brand safety.'

The agent drafts 10 to 20 searches across the other three modes, including translated keywords for the target market, and returns five to eight sample profiles for feedback. It then reviews every candidate it found – typically 2,000 to 20,000 accounts depending on the market – looking at posts, profile images and bios, and scores each one against the brief with a one-line reason for accepting or declining it.

In one run for the German perfume brief it executed 18 searches, found around 4,300 profiles, rated 2,300 unique accounts and accepted 330 as matches. The process takes between 20 minutes and two hours, against days of manual review for a comparable list.

Part of a Plan To Automate Influencer Marketing End To End

The release is one piece of a wider strategy at Influencer Hero to automate influencer marketing with AI, from discovery through to campaign execution. Recent releases point the same way. The company has shipped an MCP (Model Context Protocol) server with 123 endpoints: 67 for reading data and reporting, and 56 that take action directly in the platform, such as creating deals and tasks, sending emails and products, managing payouts and building automations.

Where most influencer platforms have so far exposed reporting endpoints only, the company says, this lets AI assistants run a campaign rather than just analyse it. Influencer Hero has also built AI automations into its CRM, added a native Airtable integration, provisioned WhatsApp numbers for creator messaging, and made AI profile review configurable per brand.

The AI does the heavy lifting, but the company has not automated people out of the picture. Brands on managed plans get 24/7 access to a human account manager, so a campaign question, a payout issue or a search that needs a second opinion gets a real person at any hour.

The new Influencer Finder is available for Instagram, TikTok and YouTube.

Influencer Hero is an all-in-one influencer marketing software for D2C brands and agencies, covering creator discovery, outreach, CRM, campaign management, affiliate tracking and payments, and UGC management across Instagram, YouTube and TikTok. Brands including Gymshark and Optimum Nutrition run automated creator programmes on the platform, which holds a 5.0/5.0 rating on Capterra and 4.8/5.0 on G2.