The discovery problem no social media operator admits out loud
Every creator I know hits the same wall about six months in. You’ve got a solid content engine—posts are scheduled, engagement is ticking up, and the analytics look like a hockey stick that hasn’t quite launched. Then you need to find someone to collaborate with, a brand to pitch, a guest for your podcast, or a buyer for your productized service. And suddenly you’re back in 2015, staring at a LinkedIn search bar, a spreadsheet of “maybe” accounts, and a vague sense that the right people are out there but you have no reliable way to surface them.
This is the gap that tools like Lev8 are trying to fill—and for social media operators who live and die by the quality of their outreach, it matters more than any algorithm update. Because no matter how good your content is, distribution still depends on knowing who to talk to and why they should care right now. Static databases (Apollo, ZoomInfo, Lusha) give you a list of names that are often six months stale. General-purpose search gives you noise. What we actually need is a live intelligence layer that treats people and companies as dynamic entities, not frozen rows in a CSV.
The team behind Lev8 launched on Product Hunt last week with a pitch that cut straight to my pain: “Finding names is easy—getting real context to reach out right now is the bottleneck.” That’s the sentence I’d tattoo on the monitor of every growth marketer I know. Let me walk through what this tool actually does, where it shines for creators, and where I’d pump the brakes before rolling it into production.
The discovery gap that social media operators feel every day
I run several niche social accounts—one focused on indie SaaS tools, another on creator monetization. When I wanted to find CMOs of subscription businesses with fewer than 50 employees who had recently posted about retention strategies, my existing stack failed. Hunter gave me email patterns but no context. Apollo gave me a list of titles that were probably correct last year. I spent four hours cross-referencing LinkedIn posts with Crunchbase funding data and still ended up guessing.
That’s the exact problem Lev8’s co-founder Tony Zhang described in the launch thread: “Our team spent hours jumping between search tools, databases, spreadsheets, and enrichment services just to answer a simple question: Who should we talk to, and why?” The product solves this by running a swarm of background agents that mine the live web—not a pre-indexed database—and then verify identity, enrich context, and score signals for freshness.
What that means in practice is you can type in plain language: “Find me VPs of Sales at fast-growing voice agent startups in the Bay Area that raised funding recently.” Lev8’s crawler hits public sources (company websites, Crunchbase, GitHub, forums, LinkedIn via authorized APIs) and returns a list of people with evidence attached—not just a name and email but the funding announcement, a recent GitHub contribution, a thread they posted on Reddit. The makers claim the identity system uses cross-model validation across multiple LLMs to reduce hallucinations, accepting data points only when all models agree. That’s a meaningful architectural choice; I’ve tested AI-led enrichment tools before (looking at you, LeadIQ and Snov.io) and the hallucination rate on the “why now” signal was brutal.
Why TikTok creators should care more than LinkedIn ones
Here’s where I’ll split the audience. If your social media operation is B2B and lives on LinkedIn, you already have a relatively structured data lake. Titles, company pages, job change notifications—LinkedIn surfaces a lot. Lev8’s value there is speed and breadth: you can run a query that would take you two hours in thirty seconds.
But for creators who operate on Instagram, TikTok, or YouTube—where the “company” might be a solo creator with a Substack and a Discord—traditional B2B databases are useless. Lev8’s live-web approach is far more relevant because it can find a creator who just launched a Patreon, posted a viral video about camera gear, and hasn’t updated their LinkedIn in two years. The tool’s ability to “find coffee shops in San Francisco with a 4.5+ rating and no website” (one of the examples in the launch) shows how far it goes beyond corporate data. For a creator looking to partner with small businesses or niche brands, that level of discovery is gold.
I’d still want to test how well it handles semi-structured sources like TikTok bios or YouTube channel descriptions—those aren’t traditionally crawled—but the team’s emphasis on “parallel agents exploring the web” suggests they’re prioritizing breadth over depth of any single platform.
How Lev8 actually works (and why the architecture matters)
Most enrichment tools operate on a batch-and-cache model. You upload a CSV, they match against a static database they licensed from a data broker, and you get back whatever was true the last time they ran a refresh. That’s why you often get a “VP of Marketing” who left the company eight months ago.
Lev8’s approach is fundamentally different. Instead of a static database, they run a “live system that mines the open web in real time,” as RichgaLu (a member of the Lev8 team) explained in the comments. The agents monitor sources 24⁄7 for changes—job moves, funding rounds, tech stack updates, even forum posts. Then a three-layer qualification process strips out noise, verifies identity across platforms, and ensures the contact is reachable before you ever hit send.
This matters operationally because it changes the risk profile of cold outreach. If you’re a creator emailing a brand manager about a sponsorship, and you reference a campaign they just launched last week, your response rate goes way up. Lev8’s signal scoring system claims to only “trigger outreach when there is true urgency and ICP fit.” In my experience testing similar AI outreach tools, the biggest failure mode is generating personalized-sounding messages that are actually based on stale or wrong information. The cross-model validation—running the same data point through multiple LLMs and only accepting it when they agree—is a clever defense against the “confidently written spam” problem that Andras Czeizel rightly flagged in the launch thread.
That said, the team acknowledged that this approach sacrifices some recall for precision. If you need a massive, broad list of every possible contact in a category, Lev8 may return a smaller set than a traditional database. But for a creator or small team sending a dozen highly targeted outreach messages per week, precision is far more valuable than volume.
Where the math breaks: delivery, compliance, and the obvious caveats
I’m not going to pretend this is a perfect tool for every creator. The launch thread surfaced three major questions that any operator needs to answer before adopting Lev8.
Compliance and account risk. Omri Ben-Shoham asked a sharp question in the comments: “If it’s finding someone’s personal email/phone/social from public sources and then messaging them across several channels automatically, what’s stopping that from tripping spam filters or violating CAN-SPAM/GDPR?” Lev8’s response was measured: they don’t blast across all channels automatically. They use a paced sending strategy with controlled time intervals, and compliance is ultimately the user’s responsibility. They explicitly state they don’t position the product as a way to bypass spam protections. That’s honest and correct, but it also means you can’t just turn Lev8 loose on a list of 500 prospects and walk away. You still need to understand platform policies (LinkedIn’s anti-automation rules, Instagram’s DM limits) and handle opt-outs yourself. For solo creators without legal support, this is a real burden.
Data source transparency. Janez Novak asked what data sources Lev8 uses and how it handles name ambiguity. The team replied that they built their own search engine to discover and analyze publicly available web information, while integrating data from “a range of established data platforms.” They also cross-check company info and work history to disambiguate similar names. But they didn’t name the specific data partners or the full set of sources. For trust, I’d want to see a documented list—is it scraping LinkedIn user profiles via authorized APIs (they confirmed they use licensed LinkedIn data APIs, not scraping through user accounts) and then layering on Crunchbase, PitchBook, and public social profiles? That’s helpful, but I’d still want to verify the freshness on a sample.
Pricing and scale. The launch offered Product Hunt users 500 free credits, but as of writing, full pricing is not disclosed. The team hasn’t said what happens after those credits run out. For a creator running a small operation, a $50/month tool might be fine; for a social media manager handling multiple brand accounts, the cost could escalate quickly. I also don’t know how many searches or contacts a “credit” buys. This is the biggest open question for me.
What creators and small teams can borrow from Lev8’s approach (even if you never sign up)
You don’t have to buy the tool to learn from its philosophy. For any social media operator, the most valuable takeaway is the shift from static list-building to live signal discovery. Here are three practices I’m immediately applying in my own workflow:
Define your “why now” before you search. Lev8’s success depends on framing the query with intent and recency. Instead of “find me fitness influencers,” ask “find me fitness influencers who have posted about recovery tools in the last 30 days and have engagement rates above 3%.” That specificity forces you to articulate the relevance criteria you’d otherwise skip.
Use multiple verification steps. Lev8’s three-layer qualification and cross-model validation are overkill for a manual search, but the principle is sound. When I find a potential collaborator via a search, I now cross-reference their recent activity across three platforms (LinkedIn, X, and their blog or YouTube) before drafting an outreach message. It takes 10 minutes per person but has doubled my reply rate.
Track signal freshness manually. I keep a simple Airtable with columns for “last public activity” and “signal type” (funding, product launch, job change, personal milestone). Lev8 does this automatically, but a lightweight version can be built in minutes. The key insight: if the last public activity is more than three months old, that person is probably not a “right now” target.
What I’d watch / test next
If I were a creator or social media operator evaluating Lev8, I’d take the 500 free credits (still available per the launch) and run three specific tests:
Test a niche query outside tech. Use the healthcare example from the comments: try to find therapists in a specific state by license type, then see if the tool surfaces a named human with a working contact method. That’s the hardest test case and will tell you whether Lev8 actually solves for the “long tail” of discovery or just works for SaaS leads.
Evaluate signal freshness for a known contact. Search for yourself or a colleague who recently changed jobs. Check whether Lev8 surfaces the new role, the date of the change, and a source link. Do this weekly for a month to see how fast the system picks up updates.
Assess the outreach execution on a small test list. Connect one social account (say, a secondary Twitter/X account) and send a handful of outreach messages. Monitor deliverability and reply rates. The team says pacing is built in, but you’ll want to see whether it feels human or robotic—and whether any platform flags the activity.
Ultimately, Lev8 is not a magic wand. It won’t replace the judgment needed to write a message that doesn’t feel like spam. But if the live-crawler architecture works at scale, it could become the default layer between a creator’s content engine and the people they need to reach. For a solo operator or a small team, that’s worth 500 credits of your time to find out.
