Cheerful research
Almost every rate guide online publishes what creators ask. Rate cards, tier ranges, survey estimates. This one also publishes what the deals settled at, because the prices moved through our own inbox and we can see both ends.
Measured on 16 September 2026 over 905 creator asks, 298 brand offers and 2,356 itemised rate-card lines from real brand and creator email threads. Every figure carries its sample size. Free to reuse with a link.
What creators asked for, in US dollars, for a single one-off deliverable. Follower count is frozen on the quote itself, so the size and the price are same-dated.
| Followers | Low | Typical | High | Sample |
|---|---|---|---|---|
| 1K – 10K | $250 | $450 | $850 | 220 quotes · 18 brands |
| 10K – 50K | $350 | $500 | $1,300 | 422 quotes · 20 brands |
| 50K – 250K | $500 | $1,500 | $3,000 | 199 quotes · 24 brands |
| 250K+ | $2,375 | $4,250 | $10,125 | 64 quotes · 17 brands |
Low and high are the 25th and 75th percentiles. A quarter of creators that size asked below the first figure and a quarter above the last. The spread is the finding: niche, exclusivity and usage rights move a price further than follower count does, which is why there is no single recommended rate anywhere on this page.
From itemised rate cards, normalised to one unit, so a “3 videos for $2,000” pack counts as $667 rather than $2,000.
| Deliverable | Low | Typical | High | Sample |
|---|---|---|---|---|
| Instagram Reel | $500 | $1,800 | $3,900 | 586 quotes · 30 brands |
| Multi-item bundle | $850 | $2,900 | $7,000 | 389 quotes · 32 brands |
| TikTok video | $500 | $2,000 | $3,500 | 332 quotes · 27 brands |
| Story set | $400 | $1,000 | $2,500 | 287 quotes · 35 brands |
| UGC video (no posting) | $282 | $500 | $1,600 | 211 quotes · 18 brands |
| Long-form video | $2,000 | $5,000 | $10,000 | 195 quotes · 15 brands |
| Video integration | $1,500 | $3,000 | $5,800 | 161 quotes · 19 brands |
| Short / Reel (YouTube, TikTok) | $850 | $2,500 | $4,000 | 129 quotes · 21 brands |
| Feed post | $600 | $1,500 | $3,000 | 42 quotes · 14 brands |
| Carousel | $366 | $2,000 | $5,875 | 24 quotes · 9 brands |
A story figure prices a story set, not a single frame. Around 40% of story lines never say how many frames they include, so dividing would mix two different things into one number. Usage, whitelisting and exclusivity lines are excluded here and covered below.
The part other rate guides cannot publish
Each row compares two prices on the same deal, so these are not two averages divided by each other. The four rows run over different, overlapping sets of threads, which is why they do not multiply together neatly.
The opening gap. A creator typically asks twice what the brand ends up willing to pay.
Once a creator counters rather than opens, the gap collapses from 2x to about 1.2x.
The quartiles are 1.00, 1.00 and 1.00. Deals close at the brand’s number, not above it.
Creators settle at about two thirds of what they opened with — and a quarter of them at 40% or less.
Read the third row twice. Counted deal by deal rather than read off the quartiles, 73% of deals settle at exactly the highest number the brand was willing to say out loud, 98% at or below it, and only 3% above. Your ceiling is the price. The number you put in the first email is not an opening position, it is in practice the deal.
A 5K creator asks three times what brands typically offer at that size. A 500K creator asks 1.4 times. Small creators are guessing. Large ones have a rate card and know what it is worth.
| Followers | Typical brand offer | Typical creator ask | Gap |
|---|---|---|---|
| 1K – 10K | $150 | $450 | 3x |
| 10K – 50K | $200 | $500 | 2.5x |
| 50K – 250K | $650 | $1,500 | 2.3x |
| 250K+ | $3,000 | $4,250 | 1.4x |
Both columns are medians of separate populations, so this gap is a ratio of two typical figures rather than a per-thread measurement. The 2.00x above is the per-thread version and is the one to quote about a single negotiation.
Standard advice is to offer a hybrid: a reduced base fee plus a performance bonus. The quotes do not reflect that. Counting every priced line a creator sent:
| Pricing model | Share of lines | Sample | What it means |
|---|---|---|---|
| Flat fee | 74% | 4,818 · 50 brands | One number for one deliverable. The default by a wide margin. |
| Per post | 11% | 687 · 37 brands | A unit price, quoted when the deal is several of the same thing. |
| CPM | 10% | 640 · 15 brands | Priced against reach. Concentrated: 15 brands, and one is 41% of the lines. |
| Monthly retainer | 2% | 161 · 28 brands | An ongoing arrangement rather than a campaign price. |
| Commission / affiliate % | 2% | 99 · 28 brands | Performance-based pay, quoted far less often than the advice suggests. |
| Product only | 0% | 11 · 8 brands | Gifting with no fee. 11 lines. Creators essentially never quote it as a price. |
Shares are of priced lines, not of revenue: one rate card listing a Reel, a story set and a bundle contributes three lines. CPM is the one row to treat carefully, since 15 brands produced it and one of them is 41% of the sample.
The honest answer to “what should I pay” is a band, and it is worth being specific about why rather than waving at it. Three reasons, in order of how much they cost you if you ignore them.
At 10K to 50K followers the middle half of asks runs $350 to $1,300. The typical ask one tier up is $1,500. So knowing a creator’s follower count tells you far less than knowing where they sit inside their own tier, and two creators the same size routinely quote four times apart. That is why we tested a single recommended rate and abandoned it: predicting a real quote from the median of its peers lands within a factor of two only 37% to 75% of the time.
Every rate guide says to adjust up for strong engagement. We tested it against the asks themselves, splitting each tier at its own median engagement rate. The relationship is weak everywhere and it does not even point the same way in every tier.
| Followers | Above-median engagement | Below-median engagement | Correlation | Sample |
|---|---|---|---|---|
| 10K - 50K | $500 | $625 | -0.15 | 110 per side · 13/12 brands |
| 50K - 250K | $2,000 | $1,500 | +0.12 | 67 per side · 18/10 brands |
| 250K+ | $3,500 | $2,000 | +0.23 | 15 per side · 10/8 brands |
Correlation is log-log, inside the tier. At 10K to 50K the more engaged half asks for less. The 1K to 10K row is missing on purpose: its above-median cell is 67% one brand, so it fails our share test, and half a comparison is not a comparison.
This does not mean engagement is irrelevant. It means engagement tells you whether an ask is good value, not what the ask will be. A 15K creator whose audience actually buys is worth more than a 50K creator whose audience does not, and neither of them knows that when they send you a number. Use engagement to judge the quote in front of you. Do not use it to predict the quote, and do not let anyone sell you a formula that multiplies one by the other.
Location moves rates, and we have no data to tell you by how much. The snapshot frozen onto each quote carries follower count, engagement, average likes and comments, posting cadence and sponsored share. It carries no country, region or language, so there is no honest way to split any table on this page by market.
One thing worth not misreading: these tables are filtered to US dollars, and 21% of quotes are in another currency and stored unconverted. That is a currency filter, not a country filter. A creator in London quoting in dollars is in these numbers and we cannot pick them out. Treat the tables as dollar-denominated rates skewed toward US direct-to-consumer campaigns, which is what Cheerful mostly runs, rather than as a US benchmark.
Nearly every rate guide states that usage rights add somewhere between 50% and 150% to a base rate. None of them sources it. We have 940 rate-card lines that explicitly name paid usage, out of 12,076 lines in total, so we tried to check it.
Comparing lines of the same format with and without usage named, the ones that name paid usage come out lower on three formats of the four we tested, and two of the four fail our publication gate outright. The reading that fits is that a line naming paid usage is usually the add-on priced on its own row rather than the deliverable with usage folded in. That measures the price of the rider, not a premium on the base rate.
So we cannot confirm or refute the 50% to 150% figure, and we are not going to publish a multiplier we would have had to invent. What the data does support is narrower and more useful: usage is normally quoted as its own line, not as a markup. Ask for it itemised and you will be reading the same thing the creator is.
Since 73% of deals land on exactly the brand’s highest number, opening low does not save money, it costs you the reply, and opening high does not buy goodwill, it just becomes the price. Pick the real number, say it once, and spend the negotiation on scope.
A first ask is typically 2.00x your number and is mostly an anchor. A counter is 1.19x and is close to final. When someone counters you are usually one small move from done, and a quarter of counters already sit at or below the brand’s figure.
Usage is quoted as its own line, which makes it the cleanest thing to trade. Drop the exclusivity window or shorten the usage term rather than moving the fee. Practical wording for each of these is in our guide to negotiating rates with influencers.
Measured on 16 September 2026. The source is an append-only log of prices stated in real brand and creator email threads on Cheerful. Each price carries who said it and what kind of statement it was, so a creator’s ask and a brand’s offer are never averaged into one figure.
Only one-off fees in US dollars between $50 and $50,000 are included. Monthly retainers and per-subscriber rates are not comparable to a per-post price, and amounts outside that window are almost always an extraction artefact rather than a rate anyone quoted. Non-USD prices are kept in their own currency rather than converted, so they are out of scope here.
No figure is published unless it survives three tests: at least 15 quotes, from at least 5 different brands, with no single brand making up more than 50% of them. The third test is the one that does the work. A segment can have a perfectly healthy sample and still be essentially one company’s price list, and publishing its median would hand that company’s pricing to its competitors. Segments that fail are dropped, never rounded or merged into a neighbour.
There is no single recommended rate on this page because we tested one. Predicting a real quote from the median of its peers lands within a factor of two only 37% to 75% of the time depending on the segment. The spread is genuine information, so we show it instead.
These are Cheerful’s own campaigns, which skews toward direct-to-consumer brands working with nano through mid-tier creators in the US. It is not a census of the influencer market.
Tables are free to reuse under CC BY 4.0 with a link back to this page.
Related: how to negotiate rates with influencers for the wording, Instagram engagement benchmarks for judging whether a rate is justified, and outreach for the message that starts the thread.
From first hello to signed agreement — see how much of the work disappears.