Three brands in this category file their marketing spend with the government. Here is what it says.
Every free benchmark on supplement and beverage marketing spend traces back to a vendor whose data pages no longer load. So we stopped looking for benchmarks and read three annual reports instead. One supplement brand, one hemp supplement brand and one beverage brand, all public, all audited, all filed with the Securities and Exchange Commission for the year ended 31 December 2025. Three companies is not an industry. It is three companies whose numbers are true.
The benchmark everyone quotes has no source behind it.
Search for what a supplement brand should spend on marketing and you will find the same handful of figures repeated across agency blogs, usually a cost per purchase to two decimal places. We went looking for the primary source. It is a benchmarking vendor that has since changed business, and its benchmark pages no longer resolve. The figure survives only in the articles that copied it.
That is the normal state of this subject. The numbers in circulation are one agency’s own client book, or a dataset nobody can open, and in both cases the sample and the definitions are unstated. A figure you cannot check is not a benchmark, it is a rumour with a decimal point.
A public company has to publish audited numbers on a schedule, in a document it can be sued over. So this page uses three of those instead. The trade is honest and worth stating: filings are true and few, benchmarks are many and unverifiable. Three companies cannot tell you what your category spends. They can tell you what three real businesses in it actually did last year, which is three more than most published benchmarks can.
Three income statements, side by side.
All figures are for the year ended 31 December 2025, in thousands of US dollars, taken from each company’s consolidated statement of operations. Percentages are ours, computed from those lines. Where a company states a margin itself, our arithmetic matched it before we published.
| Company | What it sells | Revenue | Gross margin | The marketing line | As a share of revenue |
|---|---|---|---|---|---|
| Niagen Bioscience | Tru Niagen, a nicotinamide riboside supplement | $129,423 | 64.3% | Sales and marketing, $35,506 | 27.4% |
| Charlotte’s Web | Hemp derived CBD supplements | $49,897 | 43.5% | Advertising, $8,127 | 16.3% |
| Jones Soda | Craft soda and canned beverages | $25,303 | 26.7% | Selling and marketing, $5,254 | 20.8% |
Read the marketing column carefully, because the three companies are not counting the same thing. Niagen’s line is a full sales and marketing function. Jones Soda’s is selling and marketing, which includes the cost of selling. Charlotte’s Web does not publish a marketing line at all; the $8,127 is an advertising figure disclosed in the notes, and it sits inside a much larger selling, general and administrative expense. Comparing these three as though they measured one thing is the mistake most published benchmarks make. We are showing them together to make the difference visible, not to average them.
The high margin brand is the one that can afford to advertise, and it spends the most.
The intuition runs the other way. A brand with thin margins is usually told it has to work harder on marketing efficiency, and a brand with fat margins is told it has room to be patient. The filings show the opposite behaviour, and the arithmetic explains why.
Subtract the marketing line from the gross profit and you have what is left to run the rest of the company: people, premises, product development, everything.
| Company | Gross profit | Less the marketing line | What is left |
|---|---|---|---|
| Niagen Bioscience | $83,189 | $35,506 | $47,683 |
| Charlotte’s Web | $21,700 | $8,127 | $13,573 |
| Jones Soda | $6,764 | $5,254 | $1,510 |
Jones Soda has one and a half million dollars of gross profit remaining after it markets, on twenty five million dollars of sales. Niagen has forty seven million, having spent proportionally more. That is not a difference in discipline. It is the difference between selling a powder at 64 percent and a canned drink at 27 percent, and it decides in advance how much advertising a business can carry.
The practical read for a founder: your gross margin sets your marketing budget before you write a media plan. If you are in a canned format at high twenties, a marketing programme that looks like a supplement brand’s will consume everything you make. The plan has to be built from the margin outward, not from the tactics inward.
Getting on the shelf cost about as much as all its marketing.
Jones Soda discloses something in its notes that almost nothing else in this category makes public. Retail revenue is recorded net of what the brand pays retailers to be stocked, and the filing states the amount plainly:
Set that beside the selling and marketing line of $5,254 thousand and the proportion is the point. Shelf fees and promotional allowances came to roughly half of everything the company spent on selling and marketing that year. Because they are netted against revenue rather than booked as an expense, they never appear in the marketing budget at all. A brand looking at its own profit and loss would not see them next to its media spend.
The year before was worse, at $4.3 million, so the direction is good. But the structural fact stands, and it is the same fact the retail transition study on this site found from the other end: the money that goes into being on a shelf and the money that goes into telling anyone you are on it are managed by different people, in different lines, and only one of them is called marketing.
What this means for a budget conversation. If you are moving into retail, the trade spend is not a line item beside advertising. It is a reduction in the revenue you will report, and it can be the same order of magnitude as the entire marketing programme. Model it before you sign the distribution, not after.
Most of what a filing calls marketing is not advertising.
Charlotte’s Web is the useful case here, because it publishes both numbers. Its selling, general and administrative expense for 2025 was $41,968 thousand, against revenue of $49,897 thousand. That is 84 percent of revenue in overheads. Inside it, the advertising figure disclosed in the notes is $8,127 thousand.
Advertising is 19 percent of what that company reports as selling, general and administrative expense. The other 81 percent is salaries, premises, professional fees, share based compensation, bank charges and depreciation. None of it buys an impression.
This is why a percentage of revenue rule of thumb is close to useless in this category. Told that a wellness brand spends thirty percent on marketing, a founder will imagine a media budget. In a filing, that same thirty percent may be almost entirely people and rent. The two brands in this table with a genuine advertising or media figure spent 16.3 percent and 20.8 percent of revenue. The one reporting a full function spent 27.4 percent. Those are three different questions wearing the same units.
The one comparison in this table that is close to like for like is Niagen’s consumer products segment. The company states that segment’s sales and marketing expense fell to 36 percent of that segment’s net sales in 2025, from 37 percent in 2024. That is a marketing function measured against the sales it supports, which is the number a founder actually wants, and it is much higher than the company wide figure.
What this page is not.
It is three companies. It is not a benchmark, an average, or a target, and using it as one would repeat the error it was built to avoid.
- Public companies are not typical. All three carry the cost of being listed, and two of the three ran an operating loss in 2025. A private brand of the same size has a different cost structure and different incentives.
- The lines are not the same lines. Sales and marketing, selling and marketing, and advertising are three different measurements, as the section above sets out. We have not averaged them and neither should anyone quoting this page.
- One year is one year. Jones Soda’s slotting figure moved from $4.3 million to $2.7 million in a single year. Any of these ratios could look different in the next filing.
- Segment figures and company figures are not interchangeable. Niagen’s 36 percent is a segment number and its 27.4 percent is company wide. Both are true and they answer different questions.
- We have not adjusted for anything. No inflation, no seasonality, no like for like restatement. The numbers are as filed.
Everything here can be checked. Each company’s annual report is public on the Securities and Exchange Commission’s EDGAR system, the figures are on the consolidated statement of operations, and the slotting sentence is in the revenue recognition note. If our arithmetic is wrong, the filing will say so.
The three filings, so you can open them.
- Niagen Bioscience, Inc. Annual report on Form 10-K for the year ended 31 December 2025. Revenue $129,423 thousand, cost of sales $46,234 thousand, gross profit $83,189 thousand, sales and marketing $35,506 thousand. The company states its overall gross margin as 64.3 percent and its consumer products segment sales and marketing as 36 percent of that segment’s net sales. Read the filing
- Charlotte’s Web Holdings, Inc. Annual report on Form 10-K for the year ended 31 December 2025. Revenue $49,897 thousand, cost of goods sold $28,197 thousand, gross profit $21,700 thousand, selling, general and administrative $41,968 thousand. Advertising expense of $8,127 thousand is disclosed in the notes to the consolidated financial statements. The company states gross margin of 43.5 percent. Read the filing
- Jones Soda Co. Amended annual report on Form 10-K/A for the year ended 31 December 2025. Net revenue $25,303 thousand, cost of goods sold $18,539 thousand, gross profit $6,764 thousand, selling and marketing $5,254 thousand. The slotting and promotion allowance sentence is in the revenue recognition note. Read the filing
Filings opened and read on 5 September 2026. Every percentage on this page was computed from the figures above and, where the company states the same ratio itself, checked against it before publication.