Skip to content
⌘K
InboxAsk a human

Sample Requests: reviewing requests, allocations and auto-approval

Updated by Alastair Hardie

Sample Requests Management brings every open sample request across your active collaborations into one place, so you can approve, reject or ignore them without moving between collaborations on TikTok.

It does three jobs: reviewing the requests that come in, capping how much product goes out each month, and setting rules so qualifying requests approve themselves.

Reviewing requests

The table groups requests so you can work through them in whatever way suits the decision you are making. Use Group by at the top left to switch between:

- Product, the default. Best for deciding how much of one product to send out

- SKU, when variants matter

- Creator, when you are deciding about a person rather than a product

Search and Filter narrow the list further. Filters include the creator's risk level and their Post Rate, which is the share of a creator's approved samples that resulted in a posted video. Post Rate is usually the single most useful filter on this page, because it separates creators who convert product into content from those who do not.

The tabs

Pending. Requests waiting on a decision from you.

Approved. Everything you have approved, at any stage of fulfilment. A request stays here while it is being prepared, while it is in transit, once it has been delivered and after the creator has posted.

Rejected. Requests you actively declined.

Ignored. Requests you have set aside in Unbox. Ignoring is not a rejection on TikTok, it removes the request from your queue so it stops competing for attention.

Expired. Requests that lapsed rather than being decided, including ones that timed out while pending, ones that could not be fulfilled, and ones the creator withdrew.

The Expired tab is worth watching. A request that sits in Pending long enough will expire on its own, which is a decision by default. If you use letting requests lapse as a soft rejection, expect them to land here rather than in Rejected, and remember that no sample was ever shipped for any of them.

Taking action

Select requests with the checkboxes and the bulk action bar appears with Approve, Reject and Ignore, showing how many rows you have selected.

Ignore works on any request. Approve and reject need the request to still be attached to a live collaboration, so if some of your selection cannot be actioned Unbox will tell you which and process the rest.

Where a sample has got to

Once approved, the sample status on the creator's row tracks fulfilment:

- To Review, no decision yet

- Preparing, approved and awaiting shipment

- In Transit, shipped

- Delivered, arrived with the creator, now waiting on content

Allocations: capping what goes out each month

Allocations let you decide, per product, how many units creators can be approved for in a month. This is the control that keeps sampling inside a budget rather than following demand.

Open Set Allocations from the top of the page. For each product you set a monthly unit figure, and Unbox tracks three numbers:

- Total Allocations, the monthly cap you set

- Approved this month, how many have been approved against it

- Available, what is left

Counts reset at the start of each calendar month.

Setting them

By hand. Search your product list and type a figure against each product. You can also edit a product's allocation inline from its row in the requests table.

By CSV. Import a file with two columns, Products and Allocated Units. Download the template to get your current caps as a starting point, edit it and upload. The review screen shows which rows matched a product, which will overwrite an existing cap, and which failed because the product name did not resolve, was ambiguous, or the units were not a valid number.

Products you have never given an allocation sit at zero. Set caps on the products you actively sample rather than trying to cover the whole catalogue.

Auto-approval rules

You do not have to review every request by hand. Auto-approval is set per collaboration, in the Samples step when you create a collaboration.

Turn on Offer Free Samples, then choose between:

- Auto-approve requests, Unbox decides using the minimums you set

- Manually review, every request comes to you

Setting the minimums

With auto-approve selected, you set the bar a creator has to clear:

- Min. GMV / last 90 days, the GMV the creator has generated in the last 90 days

- Min. Followers

- Min. Post Rate, their rate of posting in the last 30 days

- Min. Avg. Views / Video

- Category Fit, limits requests to creators who sell within a product category you choose

A creator has to clear every minimum you set. Miss one and the request is auto rejected.

Leaving all the fields empty auto-approves everyone who asks. That is the setting's literal behaviour, so if you turn auto-approve on and skip the minimums, you have opened the collaboration to anyone. Set at least one.

Choosing sensible minimums

A few practical points:

Start with Post Rate or GMV, not Followers. Follower count is the weakest predictor of whether a sample turns into a sale. Post Rate directly measures whether a creator converts product into content.

Check what the bar excludes before you commit. A minimum set too high on a small creator pool can decline nearly every request, and those creators are rejected rather than left waiting, so you cannot recover them later.

Newer creators may have no history to score. A creator with no recorded GMV will not clear a GMV minimum, so a strict GMV floor closes the collaboration to creators you have never worked with.

Auto-approval does not check your allocations. They are separate controls. If you are running to a budget, review the allocation on a product before opening it up to auto-approval.

How this connects to the rest of Unbox

Blocking a creator ignores all of their pending sample requests. If you block someone, you do not need to clear their requests separately.

Sample behaviour feeds blocklist rules. Ghosted Samples, meaning creators who received product and posted nothing, and Sample-to-Post Rate are both available as conditions when you build a blocklist rule. Persistent non-posters can be handled by a rule rather than by hand.

Common questions

Why can I not approve some of the requests I selected?

Approve and reject need the request to still be attached to a live collaboration. If the collaboration has ended, only Ignore is available.

What is the difference between ignoring and rejecting?

Rejecting declines the request. Ignoring only removes it from your queue in Unbox. Use ignore to clear noise you do not want to formally decline.

A request disappeared from Pending without anyone touching it.

It expired. Check the Expired tab. Requests do not sit pending indefinitely.

Requests show as approved but nothing has shipped.

Approved covers the whole fulfilment run. Check the sample status on the row: Preparing means approved but not yet shipped.

My allocation says available but requests are still being declined.

Allocations and auto-approval minimums are separate. An auto-approve rule can decline a creator on their metrics regardless of how much product is left.

Do allocations carry over if I do not use them?

No. Counts reset each calendar month, they do not roll forward.

Can I set auto-approval across all collaborations at once?

Auto-approval is configured per collaboration when you set it up.