Seeding is sending free product to creators and hoping. That is not a criticism — it is a legitimate tactic with a measurable conversion rate and a real place in a launch. The trouble is that most seeding programmes are run without anyone establishing what that rate is, which turns a campaign into a shipping exercise.
This covers what it costs per usable outcome, the disclosure obligations that attach the moment a parcel leaves, and the cases where a paid brief simply beats it.
Seeding versus paid creator work
They are different purchases and conflating them is the source of most disappointment.
| Seeding | Paid brief | |
|---|---|---|
| Agreement | None | Contractual |
| Cost per creator | Product plus shipping | Product plus a fee |
| Delivery guaranteed | No | Yes |
| You control the message | No | Yes |
| Posted on | Creator's channel | Usually the brand's |
| Usage rights | None unless negotiated | Whatever the brief specifies |
| Best for | Organic reach and social proof | Specific assets you can run |
Seeding buys a chance at authentic third-party endorsement on someone else's audience. A paid brief buys a file you own the right to use. Wanting both from one budget is where programmes go wrong.
What it actually costs
The per-parcel number is the wrong unit. Cost per post is the real one, and cost per usable post is better still.
Work an example. Two hundred units at $25 cost each, plus $8 shipping, is $6,600 out the door. At a fifteen percent conversion that is thirty posts, or $220 per post. At five percent — which is what cold seeding to an unqualified list produces — it is ten posts at $660 each.
Set that against a paid brief at $120 a clip with contractual delivery, and the arithmetic frequently favours paying. Seeding wins when the product cost is low, the desirability is high, and what you want is presence on other people's channels rather than assets on your own.
Costs people forget to count: the hours spent researching and contacting creators, the parcels that go to dead addresses, the fulfilment overhead, and the follow-up chasing. On a two-hundred-unit programme that labour is usually larger than the product cost.
Conversion rates you can plan against
Honest ranges, assuming a product people would plausibly want:
- Cold seeding to a bought or scraped list: 2–5%
- Targeted seeding, researched fit, personalised outreach: 10–20%
- Warm seeding to people who already mention the category: 20–35%
- Seeding after a prior paid relationship: 40%+
The gap between the first and third rows is entirely research. Sending two hundred parcels to a list costs the same as sending eighty to people who have posted about the category this quarter, and the eighty produce more.
Disclosure, which is not optional
The moment you send free product you have created a material connection, and a creator posting about it has a disclosure obligation. This holds whether or not you asked for a post, whether or not there was any agreement, and whether or not the creator liked the product.
Three practical consequences.
Say it in the outreach. Tell recipients plainly that if they post, they should disclose the gift. This costs you nothing, protects them, and removes the argument that the brand encouraged concealment.
Do not ask for undisclosed posts. Requesting that a creator omit the disclosure, or framing the gift as "no strings so no need to mention it", is the specific thing regulators pursue. Enforcement generally targets the brand rather than the creator.
Do not condition the gift on a positive post. A gift given in exchange for a guaranteed favourable review is not seeding, it is a paid endorsement dressed as one, and it fails both platform rules and consumer-protection law.
The underlying principle is the same one that runs through all of this: a reader has to be able to tell that something changed hands.
Disclosed gifting is a legitimate and widely used tactic. Undisclosed gifting is the thing that produces regulatory letters.
Running one that works
Qualify before you ship
The single highest-return step. Check that the creator has posted about the category, that their audience is in a market where you sell, and that their recent posting is active rather than dormant. Fifteen minutes per creator beats a hundred extra parcels.
Personalise the outreach
Not heavily — two specific sentences proving you looked at their work. Templated outreach converts at the bottom of every range above.
Make declining easy
A creator who feels obliged posts reluctantly and it shows. Explicitly saying there is no expectation improves both the rate and the quality of what appears.
Include what they need, not a brand kit
A one-page note with the product name, the handle to tag, the disclosure reminder, and any claim they should avoid. Nothing else. Long brand guidelines on an unpaid ask get discarded.
Send something worth posting
Packaging matters here in a way it does not for paid work, because the unboxing is often the content.
Follow up once
A single message two weeks later. Not three.
Track who converted
This is the asset the programme produces, more valuable than any individual post. Creators who posted unprompted are your best candidates for paid work.
Finding the right creators to seed
Targeting is where the conversion rate is won, and it is mostly unglamorous research rather than a tool purchase.
Start with people already in the category. Search the platforms creators post on for recent posts about products like yours and note who appears repeatedly. Somebody who posted about a competitor last month is a far better prospect than somebody with a larger following and no history in the space.
Check recency, not just totals. A profile with good numbers and nothing posted in three months is a dormant account. Follower count is a stock; posting frequency is the flow, and the flow is what you need.
Check the audience location. A creator whose audience is mostly in a market you do not ship to converts into nothing useful, whatever they post. Where a creator publishes this, read it. Where they do not, ask.
Look at their existing gifted posts. Somebody who posts about everything they are sent has an audience that has learned to discount them. Somebody who posts occasionally and specifically is worth three of them.
Size down, not up. Smaller creators convert at higher rates, respond to outreach more often, and produce content that reads as genuine because it usually is. Two hundred parcels to mid-sized accounts outperforms twenty to large ones for most products.
Keep a do-not-send list. People who declined, people whose address bounced, people who took product and never engaged. Re-sending to the third group is the most common waste in repeat programmes.
The outreach message
Short, specific, and explicit about there being no obligation. The structure that converts:
One sentence proving you looked at their work. One sentence on what you are sending and why you thought of them. One line saying there is no expectation of a post and no deadline. One line reminding them that if they do post, it should be marked as gifted. Then the ask for an address.
No brand deck, no attachment, no list of talking points. Every additional requirement in an unpaid ask reduces the response rate, and the disclosure line is the only requirement worth including.
Measuring it
Three numbers, recorded every time.
Conversion rate. Posts divided by parcels shipped. This is the number that tells you whether your targeting worked, and it should improve every round.
Cost per post. Total programme cost, including labour, divided by posts. Compare it against what a paid brief would have cost for the same count.
Cost per usable post. Of the posts that appeared, how many were on-message, well-made, and in a market you sell to? This is usually a fraction of the total, and it is the number that decides whether the next round is seeding or paid.
If cost per usable post exceeds what a paid brief costs, the programme is an expensive way of buying content you do not own. That is a legitimate outcome to discover, and better discovered at eighty parcels than at eight hundred.
The combination that outperforms both
Seed broadly, then pay narrowly.
Use a seeding round to find which creators actually convert and whose audiences respond. That round is cheap per contact and its real output is information rather than posts. Then commission paid briefs from the ones who converted, where you specify the format, the licence and the deadline, and where delivery is contractual.
This is more effective than either alone because it fixes the weakness in each.
Seeding's weakness
You cannot control or rely on the output. Nothing obliges a recipient to post, and nothing shapes what they say if they do.
Paid work's weakness
You are choosing creators from a profile rather than from evidence of how their audience responds.
Running them in sequence removes both.
How to structure the paid half covers brief writing, rates and reviewing at volume.
What we will not run
That rules out a version of seeding that some agencies still sell. It is also the reason a programme run here can go through a brand's legal review, and why the payment rails underneath are not one complaint from termination. the disclosure and audit position in full sets out what is enforced.
Disclosed gifting, and paid creator briefs posted by the brand as its own advertising, are both straightforwardly legitimate. Those are the shapes this supports.
Starting
Eighty creators, researched properly, in one or two markets. Personalised outreach with the disclosure note included. One follow-up.
Measure the three numbers. Then decide whether round two is more seeding, a paid campaign to the converters, or both — with the answer coming from your own conversion data rather than from a benchmark in an article.
Where the asset you want is an existing customer speaking rather than a seeded creator posting, what a testimonial requires and where the disclosure line falls covers a different set of obligations entirely.