A San Francisco startup spent Friday selling a simple pitch: skip the bots, skip the VPNs, and hire real people in the United States to flood TikTok and Instagram with your videos. By Saturday morning, the same launch was being called a trap for founders who do not know what distribution actually costs.
That whiplash is the story of HumanPost, a “human social posting” service that went public this week and immediately ran into a fight over pricing, math, and what organic reach is worth in 2026.
HumanPost does not make your content. It tries to industrialize the last mile of posting it.
Customers design TikTok or Instagram profiles, upload videos or carousels, and let the company assign U.S.-based operators to create the accounts, warm them up in a chosen niche, and publish by hand. The company says posters have no creative control, that every post is done manually, and that operators are verified so the network is not just another proxy farm. The homepage currently advertises more than 12,000 people ready to post and an average Scaling-plan CPM of $0.67 over the last 90 days.
Founder Mehak Vohra framed the launch as infrastructure for distribution, not another influencer marketplace. The Friday video put it in blunt terms: one brand, hundreds of accounts, thousands of posts. “You send us the content. We turn it into a distribution army.”
That line traveled. The launch post crossed roughly 150,000 views in a day. The replies did not stay congratulatory for long.
HumanPost’s public pricing is split into a test tier and a scale tier.
The Learning plan is listed at $180 a month. The catch is in the fine print. It is billed $540 every three months, plus a $100 prepaid view balance. Customers get 200 posts a month across up to five accounts, then pay $2.50 per 1,000 views, capped at $250 per post. The plan includes five comment tasks and one Spark Ads code request.
The Scaling plan is $1,125 a month, billed $3,375 quarterly. It includes 250 posts, up to eight accounts, views with no extra CPM charge, 30 comment tasks, and three Spark code requests. Enterprise is custom. Unused monthly credits do not roll over.
On paper, that is a familiar SaaS pattern: pay more while you learn, lock in a flat rate once a format works. In the replies, it read like something else.
People saw “$180 a month” and then found a quarterly invoice, a prepaid CPM wallet, and a $250 ceiling that sounded like a per-post fee. One commenter asked whether they were paying $180 and then $250 every time a video went out. Another called $540 plus $2.50 CPM “ridiculously expensive.”
The sharpest attack came Saturday from an X user who works in TikTok ads. The post called the starter plan a disguise: $540 for three months, a $2.50 CPM on content the customer already produced, and a product that amounts to renting warmed U.S. accounts. The advice was blunt. Hire a creator. Or pay clippers. Do not buy this if you already understand social distribution.
That critique spread because it named the comparison set HumanPost was hoping people would ignore.
TikTok distribution in 2026 is not one market. It is three, stacked on top of each other.
Paid ads are the cleanest benchmark. U.S. TikTok campaigns often land somewhere in the mid-single to low-double-digit CPM range depending on the objective. Against that, HumanPost’s claimed $0.67 Scaling CPM looks cheap, assuming the views are real and the accounts stay alive.
Influencer posts are the expensive benchmark. A micro-creator video can cost a few hundred to more than a thousand dollars. HumanPost is cheaper per upload than that. It is also selling something different: new or low-follower accounts pointed at For You distribution, not an existing audience.
Clipping and account rental are the dirty benchmark, and that is the one critics used. In that world, people talk about CPMs closer to a dollar, sometimes lower, with the clipper doing the account work, the posting, and the infrastructure. A public rental index from TokPortal puts small accounts in the low hundreds of dollars a month. Once operators compared HumanPost with that stack, $2.50 CPM on customer-supplied creative looked less like a premium and more like a markup.
There is a reason the company can still argue it is not the same product. HumanPost is selling U.S. operators, no VPNs, dedicated brand accounts, a dashboard, an API, warm-up, and a compliance story that a random clipping Telegram chat does not offer. Whether that package is worth 2x to 4x the gray-market rate is the actual pricing question. The launch page did not answer it clearly enough.
Vohra has been posting results for months, well before the polished launch video.
She has claimed a new app went from zero to about 8 million views for under $700 in three weeks, all on accounts created for the brand by people in the United States. Other posts cited 10 million views at a $0.07 CPM, and a consumer app doing 4 million views a month at $0.23 CPM. A site testimonial from Favorited’s Farid Adibi describes the platform as a way to get more content out and test creative without adding operations overhead.
Those numbers, if repeatable, explain the product thesis. TikTok still rewards volume and format testing more than follower counts. A network of warmed accounts can, in theory, find a winning video faster than a single brand profile. Once that format is proven, a flat Scaling plan with views included makes the effective CPM fall as reach rises.
A reply from Miles Feldstein tried to put that funnel in buyer language: Learning is roughly a dollar a post plus $2.50 CPM on a dedicated U.S. account while videos are still averaging under 1,500 views. After a format works, move to the higher flat plan, where he said people average around $0.60 CPM.
That is a coherent offer. It is also easy to miss when the first number a visitor sees is $180 and the second is $2.50 per thousand views.
Price started the fight. Structure kept it going.
New accounts do not have audiences. HumanPost is betting that warm-up plus TikTok’s recommendation graph can manufacture distribution from scratch. That is how a lot of faceless and clipping accounts work. It is also how a lot of them get limited, banned, or stuck in a 200-view jail. One reply under the launch post asked the obvious question: how effective can this be if the accounts are brand new?
Spark Ads access is thin on the cheap plan. One code on Learning is not a media-buying workflow. For brands that want to take an organic winner and put spend behind it, that limitation matters as much as CPM.
Then there is the platform-risk problem. TikTok’s rules are hostile to coordinated inauthentic behavior, account farms, and undisclosed commercial networks. HumanPost’s pitch — real Americans, real devices, no VPNs, accounts built to a brand’s spec — is cleaner than a warehouse of phones behind datacenter IPs. It is still a multi-account posting network publishing brand content at volume. Similar paid-persona operations have already been mapped on TikTok. Customers are not just buying posts. They are buying the chance that the network stays inside the platform’s tolerance.
The presentation problem may be the most avoidable part. “$180 / month” next to “billed $540 every 3 months plus $100 prepaid view balance,” then “$2.50 per 1,000 views,” then “$250 maximum per post,” is a pricing page that invites the worst reading. In a launch that asked strangers to trust a new distribution army, confusion was a cost center.
HumanPost did not invent human posting. Agencies, clip farms, and “human APIs” have been selling versions of this for years: software in front, a person on a phone at the end. TokPortal and others already market programmable operators on real devices. HumanPost’s twist is positioning — American labor, brand-owned accounts, self-serve packaging — and a public launch aimed at startup Twitter rather than closed agency Slack channels.
The backlash is a referendum on that packaging.
If you believe TikTok still has cheap organic inventory that only a human on a real U.S. phone can unlock, the Scaling plan is the product, and Learning is an expensive on-ramp. If you believe clipping already commoditized that inventory, HumanPost is charging software-margin prices for a labor market that got priced in public group chats years ago.
Both things can be true at once. The company’s case studies describe unusually cheap reach. The launch-day critics describe a price card that punishes anyone still looking for a winning format, which is exactly when a startup is most likely to buy.
Two questions will decide whether this was a messy launch or a broken offer.
First, do the $0.07 and $0.23 CPM stories repeat for customers who are not already in the company’s screenshot thread? A few winning accounts can make any network look like a machine.
Second, do the accounts last? Volume posting across many new profiles is a classic enforcement pattern. The whole pitch collapses if “dedicated accounts just for your brand” become a rotating set of banned handles.
Until those answers show up in more than founder posts, the Saturday consensus on X is the one that will travel: the idea is easy to understand, the site looks expensive, and the Learning plan asks people to pay performance prices for content they already made.
HumanPost wanted the launch to say distribution is the moat. For now, the internet is arguing that the invoice arrived before the proof.