TikTok Shop brands
Running your own affiliate program? See which free samples turn into sales, and stop shipping the rest.
TikTok Shop brands give away thousands of dollars in free samples, and most go to creators who never sell. SampleROI matches every sample to the sales it drove, blocks the repeat claimers and duds, and sends more samples to creators who actually sell. It tells you the exact fixes to make, and you approve and apply them.
Keep your outreach tool. Works alongside Euka, Reacher or any outreach tool.
Free leak audit · add-on from $99/mo · pays for itself after about 10 blocked samples
Natural Stacks · last 90 days
1.39% quality creators · flat 90 days
Dashboard preview (the web dashboard is in development; today SampleROI runs as reports and inside your AI assistant). Figures are from a real read-only audit of Natural Stacks; creator handles masked; sample costs and the leak are estimates. The recommended-fixes panel is illustrative.
Every free sample matched to the sales it drove, so you see exactly which creators never sell.
Stop repeat claimers and creators who post nothing that sells. SampleROI tells you exactly which rules to set; you approve and apply them.
Send more samples to the creators who actually drive sales, and track the payoff every week.
Who it's for
Running your own affiliate program? See which free samples turn into sales, and stop shipping the rest.
Managing shops for clients? Decide who gets free samples, cut the waste and prove the payoff in a weekly client review.
The problem, in one number
of the free-sample dollars we could trace went to creators who never made a sale.
Natural Stacks, last 90 days (Jul 24 – Oct 22, 2026). Matched cohort of 182 sampled creators; leak and yearly figures are estimates. Details below.
Case study · real data, read-only
An 8‑figure supplement brand7‑figure on TikTok Shop
Last 90 days (Jul 24 – Oct 22, 2026): 15,269 orders, 3,382 sample orders and every shoppable video featuring the brand, audited read-only.
≈ $126K / year of free product to creators who never drive a sale.
Provable floor $970/wk, even if every unmatched recipient was productive.
Almost every sample dollar went to creators with $0 attributed GMV within 5 weeks.
Matched cohort: 182 creators sampled Jul 24 – Sep 17.
Revenue, not profit — attributed video GMV within 5 weeks ÷ est. sample cost, before commissions and ads.
Same matched cohort of 182 creators.
2+ free samples in 7 days to the same account — from ~16/wk to ~152/wk in the last 2 weeks.
Est. sample spend: $2,744/wk in this window.
I assumed more free samples = more videos = more GMV. Seeing that roughly 95% of the sample dollars went to creators who never made a sale changed how I think about the whole program. Now our samples are strategic and incremental to our growth.
Roy KrebsFounder & CEO, Natural StacksDisclosure: Roy Krebs is also the founder of SampleROI.
68 of 6,819 posting creators drove 96.1% of $285,363 in creator-attributed video GMV. Quality beats volume — so samples should follow quality.
Each square ≈ 1% of creators who posted a video featuring the brand. Only the green ones sold anything.
Share of posting creators with ≥$100 GMV in a 4-week block. SampleROI's job is to grow that share — our working target is 3%.
Read-only TikTok Shop Open API data for the last 90 days (13 Fri–Thu weeks, Jul 24 – Oct 22, 2026 PT), cancelled orders excluded. Sample cost is an estimate (landed COGS + fulfillment per SKU). Creator outcomes require matching an order's display name to a video handle; that covers about 14% of free-sample recipients in the judged cohort (182 creators sampled Jul 24 – Sep 17; 176 of them posted within 5 weeks). The est. leak applies that cohort's zero-return rate (94.6% of $2,203 in samples) to the average free-sample spend ($2,563/wk) → ~$2,424/wk, ~$126K/yr; the floor ($970/wk) assumes every unmatched recipient was productive, using the matured monthly cohorts that overlap the window (Jul 1 – Sep 17). GMV is attributed video GMV, not profit.
Samples need about 5 weeks before outcomes can be judged, so the newest weeks are “immature”: this case study therefore shows the last 90 days, judging outcomes on creators sampled through Sep 17. Creator handles are masked. Results are for one brand — your numbers will differ, which is what the free audit is for.
The fix
Block the waste and send samples to creators who actually sell. SampleROI recommends each fix, down to the exact thresholds to set in TikTok Seller Center's sample auto-approval rules, and you apply it. Nothing in your shop changes unless you do it.
Repeat claimers: the same accounts take 2+ free samples in 7 days — from ~16/wk to ~152/wk in the last 2 weeks.
One open sample per affiliate creator until they post — proven creators are exempt.
~$482/wk (≈ $25K/yr) of last-90-day sample spend would have been blocked, with no proven creator blocked.
Sample requests were approved automatically, and some creators made throwaway low-view posts just to unlock more free product.
Manual review for 2-packs, 1 SKU per creator per week, minimum 30-day views (e.g. ≥ 500): the exact thresholds to set in Seller Center's sample auto-approval rules.
Closes the “post anything, get more product” path. Impact is tracked in your weekly review rather than promised up front.
Samples went everywhere, but the top 1% of creators drove 96.1% of video GMV.
Every pending request scored 0–100 on views, GMV, GPM and history; target-collaboration invites go to creators who convert.
~$727/wk (≈ $38K/yr) of sample budget redirected if 30% of the leak moves to proven creators — the same assumption as the calculator.
No weekly number for the leak, so it grew unnoticed while spend nearly tripled and quality stayed flat.
A Monday review with leak $, repeat claims, blocked orders and quality-creator rate — a trend line you can hold an agency to.
Goal: lift the quality-creator rate from 1.39% toward a 3% working target.
Two separate estimates on last-90-day data — they overlap, so don't add them.
Rule: block a repeat claim within 30 days unless the creator posted since or is proven. Applied to past orders — a simulation, not a deployed result.
MCP connection
SampleROI connects as an MCP server to Claude, ChatGPT, Cursor and Grok Bot. Ask which creators are wasting samples and get the exact fix in the same chat, then apply it in Seller Center.
MCP endpoint https://mcp.sampleroi.com
{
"mcpServers": {
"sampleroi": {
"url": "https://mcp.sampleroi.com"
}
}
}
The hosted endpoint isn't live yet. Founding customers get set up directly: [email protected].
Which creators should I stop sampling this week?
154 of 441 free-sample requests last week came from creators who still have an open, unposted sample. That's $1,854 in product.
| Creator | Open samples | Posted since | Cost |
|---|---|---|---|
| Creator A | 3 | none | $33 |
| Creator B | 3 | none | $33 |
| Creator C | 2 | none | $24 |
Decline these 154 requests in Seller Center. Proven creators are exempt. Approve-from-chat is coming once TikTok grants write scopes.
Example from a real read-only audit of Natural Stacks' requests, Oct 16–22, 2026. Creator handles masked.
Works with your stack
Outreach tools help you find and invite creators. SampleROI decides who actually gets free product — and proves whether it paid off.
Find creators, send invites, run follow-ups and CRM.
Blocks hoarders & post-rate gamers, scores requests 0–100, tracks sample ROI weekly.
Fewer wasted samples, more GMV per sample dollar.
Based only on what each vendor publishes; see “Side by side, as published” below. Trademarks belong to their owners. SampleROI is not affiliated with any of these vendors.
Pricing · add-on
SampleROI makes sure your free samples go to creators who actually sell. Outreach tools start at $199–$350/mo; SampleROI starts free.
One-time · last 90 days · 1 shop
Free shows you the size of your leak. Paid shows you who — and fixes it.
Get my free auditHow to start
What you get
Leak in dollars needs your Sample analytics or Affiliate export as well as the Orders export. Orders alone shows sample spend and repeat claims.
Smaller shops getting started with samples.
or $69/mo billed annually
Up to 300 free-sample orders/mo
1 shop
7-figure TikTok Shop brands sampling every week.
or $139/mo billed annually
Up to 2,000 free-sample orders/mo
1 shop
8-figure shops and heavy samplers.
or $279/mo billed annually
Up to 6,000 free-sample orders/mo
Up to 3 shops (regions or brands)
TikTok Shop agencies running client shops.
or $349/mo billed annually
Up to 10,000 free-sample orders/mo (pooled · 2,000 per shop)
5 client shops
+$99/mo per shop, shops 6–10
+$79/mo per shop, shops 11+
Each extra shop adds 2,000 to the pool.
For the first 25 customers who connect read-only API access and allow anonymized stats.
30-day money-back guarantee on every paid plan.
Plans are sized by free-sample orders per month. Going over never blocks anything — if you're over for 2 months, we'll suggest an upgrade.
MCP / AI-agent access on every plan (Claude, Cursor or any MCP client).
Connect through the official API, get your number and the exact fixes. SampleROI never changes your shop; you make the call.
Sample ROI calculator
Move the sliders to match your shop. The cost default comes from the Natural Stacks audit; example rows elsewhere are illustrative.
Estimate only, based on your inputs. It is not a guarantee of results.
Go deeper
Three steps, no extension to install and nothing changes in your shop until you say so.
Authorize your TikTok Shop through the standard Seller Center OAuth flow. We start with read scopes only.
Every sample is matched to the creator's videos, views and GMV — then scored.
You get the exact before → after changes, like the thresholds for Seller Center's sample auto-approval rules, and you apply them. Approval-gated actions are coming once TikTok grants the write scopes.
Eight read & intelligence tools today, plus four approval-gated actions coming once TikTok grants the write scopes — in a dashboard, or inside your AI assistant over MCP.
Landed sample cost vs. videos posted, views and GMV — per creator, per SKU, per week. See exactly which samples paid back and which vanished.
30-day shoppable views, GMV, GPM, Promote Performance Score and creator level, with anti-hoarding guardrails.
Catches multi-SKU claims in the same week and 5-second, low-view videos posted only to unlock more free product.
triage_pending_sample_queueRanks creators by GMV per video and GPM so samples and Target Collaborations go to the people who convert.
search_niche_creatorsVideos contracted vs. delivered, cost per video, ROAS and net contribution — before the next payment goes out.
track_retainer_and_blitz_deliveryOpen vs. Target Collaboration rates, overlapping invites and inverted tiers that quietly overpay.
audit_commission_tiers_and_conflictsOne-call WBR: shop sales, video GMV, sample spend, leak trend and top-10 concentration. Perfect for owners with an agency.
generate_weekly_agency_wbrSurfaces unread threads from your top-GMV creators first. Approval-gated 1:1 replies (capped at 10/hour) are coming once TikTok grants write scopes.
summarize_creator_inboxApproval-gated by design
Today SampleROI only reads your shop and recommends fixes you apply yourself. When TikTok grants the write scopes, actions stay approval-gated: SampleROI won't change your shop without a human confirming an exact diff.
SAMPLEROI_READ_ONLY=true blocks every write tool server-side. Perfect when an agency runs your shop.
stage_* returns a before → after table and a batch token. Only confirm_staged_action executes it.
Tokens expire after 30 minutes and work once. Batches max out at 50 reviews; creator replies at 10 per hour. No mass DMs.
# Preview of a coming action (needs TikTok write scopes) status STAGED_AWAITING_HUMAN_CONFIRMATION batch_token stg_7f3a… expires 30:00 · single use
| Rule | Before | After |
|---|---|---|
| 2-pack · sample mode | OPEN_AUTO_APPROVE | MANUAL_REVIEW |
| SKUs / creator / week | unlimited | 1 |
| Min 30-day views | — | ≥ 500 |
Use SampleROI on its own or next to the outreach tool you already have. Here's how the focus differs, based only on what each vendor publishes.
| SampleROI | Euka | Reacher | Cruva | HiveHQ | |
|---|---|---|---|---|---|
| Core focus | Sample ROI audit + exact fix recommendations | Creator discovery, outreach, CRM, GMV reporting | Creator discovery, outreach automation, CRM | Outreach, CRM, sample management, analytics | API-driven outreach + profit dashboard |
| Entry price (as published) | $0 audit · $99/mo add-on | $199/mo per workspace | $199/mo | $199/mo per shop | $350/mo outreach |
| Sample cost vs. creator GMV ledger | Core feature | Not listed | Not listed | Not listed | Shop-level COGS & profit tracking |
| Sample request review | 0–100 score + anti-hoarding | Not listed on pricing page | Sample review (Pro+) | Request management; AI approval (Scale) | Creative brief per request (Standard) |
| Multi-SKU hoarder & post-rate gamer flags | Yes | Not listed | Not listed | Not listed | Not listed |
| Read-only mode by default | Yes — server-enforced | Not listed | Not listed | Not listed | Not listed |
| Every write shown as a diff & human-approved | Yes (read-only today; writes coming) | Not listed | Not listed | Not listed | Not listed |
| Mass outreach automation | Not our job — keep your outreach tool (replies are 1:1, 10/hr cap) | Yes | Yes — up to 10,000 creators/day | Yes | Yes — API-driven |
| Use inside AI assistants (MCP) | Yes — any MCP client | Not listed | Not listed | Enterprise API with Claude integration | Not listed |
Competitor details are as published on each vendor's pricing and feature pages, checked Oct 23, 2026 (source links in the page source). “Not listed” means we didn't find it published, not that it can't be done. Trademarks belong to their owners.
The fastest way to lose a TikTok Shop is a bot that misbehaves. SampleROI is engineered the other way around.
OAuth 2.0 shop authorization and HMAC-SHA256 signed requests to TikTok Shop's Open API. No browser extensions, scraping or session cookies.
Read scopes only, plus a server-side write lock. Actions come later, once TikTok grants write scopes and you turn them on.
Read-only today. Once TikTok grants write scopes, every change is staged as a diff with a single-use, expiring token, and rate caps are enforced on the server.
We use order amounts, SKUs, sample flags and creator attribution. No shopper shipping addresses or phone numbers are stored.
SampleROI is an independent product of Voraus LLC. It is not affiliated with, endorsed by, or certified by TikTok.
Built on standards, not hacks: TikTok Shop Open API · OAuth 2.0 + HMAC-SHA256 · Model Context Protocol · read-only scopes first.
FAQ
Free product shipped to creators that never turns into sales: samples that never get a video, accounts that claim several SKUs at once, and throwaway low-view posts made only to unlock more free product. SampleROI puts a dollar figure on it every week.
No. It runs read-only by default with a server-side lock, so it can't change samples, commissions or messages. Many owners use it as a silent scoreboard alongside their agency.
We match sample orders to each creator's shoppable videos, views and GMV. Several free samples in a week, no posts, or posts with almost no views get flagged — and pending requests from those accounts are scored down.
No. Today SampleROI is read-only: it recommends fixes and you apply them in Seller Center. When TikTok grants write scopes, every action will be staged as a before → after diff and run only after you confirm. Creator replies will be 1:1 and capped at 10 per hour. There's no mass outreach.
Through the official API: orders and sample flags, products, shop video and product analytics, and — once you enable them — read access to affiliate samples, collaborations, creators and inbox counts. We don't store shopper addresses or phone numbers.
SampleROI ships as a Model Context Protocol (MCP) server, so it works with MCP clients such as Claude Desktop and Cursor. You can also just read the dashboard and weekly reports.
SampleROI is an independent product built on TikTok Shop's official Open API. It is not affiliated with, endorsed by, or certified by TikTok.
A read-only leak report for one shop: sample spend vs. GMV, the accounts behind the leak, your top-10 creator concentration, and the specific rule changes we'd recommend first.