Selling Microgreens to Restaurants: The Only Measured Account Is One Tray a Week

Every guide in this category calls restaurants the high-volume channel. Exactly one source has measured how much a restaurant actually uses, and it is about one tray a week.

What a restaurant account weighs

A 2019 study surveyed 52 gourmet restaurants in Hermosillo, Sonora, Mexico, in 2017. Among the ones already buying microgreens, average use was 214.29 grams a week, with a standard deviation of 69.01 and a mode of 200 grams.

Read the limits before the number. That is roughly seven restaurants, in one Mexican city, nine years ago. It is not a US figure and it is not a national one. It is here because it is the only restaurant-side measurement of quantity that exists anywhere in this category, and because every alternative on offer is a seller stating a minimum.

In this site's units it is 7.56 ounces a week, with the modal account at 7.05.

Now set it against what a tray produces. The yield field on all 62 entries here publishes a figure for 39 crops, and the median of those is 7.055 ounces a tray.

One restaurant account is one tray a week. Not a case, not a route, not a pallet. The spread on that measurement is wide, the standard deviation being 32 percent of the mean, so accounts vary by roughly a third either side, and even the top of that range is a tray and a half.

This is the number that should sit at the front of every restaurant plan in this category, and it appears in none of them.

What one account is worth, and what delivering to it costs

Price it at this site's own published restaurant row. The pricing guide records restaurant wholesale in New York City at $11.01 per 100 g, dated 2026-02-12. The measured account is therefore worth $23.59 a week.

A second seller's posted chef price lands in the same place. MicroFreshly lists a $15 base per 4 oz with 20 percent off for approved chef accounts, which is $3.00 an ounce, giving $22.68 for the same weight. Two independent postings, a dollar apart.

Then subtract the delivery. One Phoenix seller publishes a $10 flat delivery rate. Charge that to a single account and it is 42 percent of the week's revenue.

Route density decides whether a restaurant account pays, and price does not. Moving from one account to five changes the delivery burden by 34 points of revenue. No price negotiation available to a small grower moves anything like that far.

Put the rest of this site's published costs against it. Labor is $2.21 to $5.00 a tray and the median seed across the 50 costed crops here is $1.09.

Neither column has yet paid for electricity, packaging, the medium, rent, or one minute of the selling and invoicing time. None of the published labor figures include selling, which on a restaurant account is the part that never stops.

And that is the honest ceiling on a single account: about $1,227 a year at 52 unbroken weeks, which is a figure the next section argues you should not plan on.

What chefs named, against what growers grow

Two crop lists exist and they were compiled by people who have never read each other.

The restaurants already buying in Hermosillo named beet at 42.86 percent, pea at 28.57, and arugula and radish at 14.29 each. The national grower survey reports the most commonly produced varieties as radish at 29 percent, sunflower at 28 and pea at 27.

Three things fall out of that table and none of them is in the guides.

The crop chefs named most is the one crop of the four this site cannot cost per tray. Generic red beet is one of the 23 entries with no published yield, because the trade's one yield trial has a Bull's Blood row and no generic red beet row. So the top chef-named crop is the one where a grower quoting a weekly weight is guessing.

Sunflower, the highest yielder in this whole directory at 18.77 ounces, was named by nobody in the restaurant survey. It is the second most produced crop in the country and the easiest tray in the building to fill an order with.

The other three chef-named crops are all above this directory's median tray. Pea at 8.00, radish at 9.75 and arugula at 10.00 against a median of 7.055. Whatever chefs are selecting on, it is not a low-yielding specialty premium.

Do not turn any of that into a crop plan. Seven restaurants in one Mexican city in 2017 is not a demand signal, and the grower survey counts what is produced rather than what is sold. The observation worth keeping is that the two lists barely overlap and nobody in this category has noticed, because the guides cite the supply side and call it demand.

A distributor catalogue is not the third list, either. A dated March 2026 local produce sheet from a food service distributor names arugula, basil, broccoli, cilantro, pea shoots and tendrils, radish, sunflower and mixes in 4 to 8 oz packs. That documents what the distributor could source. It says nothing about what was ordered.

The format question has one real answer and it is small

Among the Hermosillo restaurants buying, all of them received harvested and packed product. Asked what they preferred, 71.43 percent wanted it cut and packed because they did not want to maintain it, and 28.57 percent preferred product with substrate for longer kitchen life.

That is a roughly 5 to 2 split among about seven kitchens, which is a real preference in one small market and not a rule. On the supply side, the national grower survey found 21 percent of growers sell a living tray and 56 percent hand-cut.

A culinary school's own account names the trade-off plainly: live product lasts in the kitchen and costs the kitchen bench space and cutting time. Both formats sell. Neither is universally preferred, and any page telling you chefs want live trays, or that chefs will not touch them, is stating a preference as a fact.

The paperwork is the account, and each buyer sets it

There is no standard restaurant document set. There is a defensible baseline and a list of things to ask.

The baseline is a dated invoice showing product, quantity, price, delivery date and terms, with a signed receiving copy. Cornell's small farms guidance says orders are commonly placed days ahead and that the invoice is the communication, not just the bill.

On payment timing, the useful number is a benchmark rather than a norm. In a 2024 survey of New York food buyers, 25 reported their payment timing: 15 said 30 days, three said 14, three paid on delivery, two said 45, one said 21 and one said 60. Net 30 was the most common and roughly a third of those buyers were not on it. That is a negotiating range, and it means a weekly account can be four deliveries deep before the first payment lands.

On food safety documents, ask before you promise. NC State's buyer guide separates a third-party certified farm from a smaller farm with a food safety plan and no paid certification. In that same 2024 survey, 69 of 80 buyers, 86 percent, would accept a written attestation instead of a certification, and some still required certification outright. The federal and state guides here cover what the law requires, which is a separate question from what a given buyer asks for.

The mechanism that ends accounts is documented even though the rate is not. NC State's restaurant guidance describes it directly: miss the buyer's ordering cut-off or surprise them with short supply, and they have to find another supplier that day. Its practical instruction is to give 7 to 10 days notice of a supply change, and to agree a standing order before sowing rather than promising to grow on demand. One seller's terms make the same point from the other side, asking 2 to 3 weeks for a first order because the crop takes 7 to 28 days to grow.

What buyers say blocks local purchases is worth having in order. Across 80 New York buyers: price and budget 58 percent, timing and seasonality 30, volume and consistent supply 28, logistics 25, quality inconsistency 19, packaging and specification failures 19, poor communication 6 and traceability 3. Price leads, and price is the thing the ranking pages tell you restaurants do not care about.

The account is attached to a person, and that person probably leaves

In 2025 there were 185,989 chef and head cook jobs and 136,605 separations, a turnover rate of 73.45 percent.

That is an occupation-wide annual rate and not a measure of supplier attrition. A separation includes moving to another kitchen, and the account may survive under a new chef. Nobody has published what share of microgreen accounts survive a change of buyer.

But it is the right order of magnitude to plan against, and it is why the $1,227 annual figure above should be treated as a ceiling rather than a forecast. NC State's guidance makes the constructive version of the point: the contact who leaves turns up somewhere else and is a warm lead at the new kitchen.

What the sampling advice costs, and what it buys

Walk in with free samples, ask for the chef, follow up, is the near-universal advice. It is published by companies selling courses, consulting and grower software, and no source anywhere publishes a sample-to-order conversion rate, or even a denominator.

The cost side this site can price. Using this site's own labor and seed figures and the 105 gram clamshell that a published grower note treats as a commercial unit, roughly three to a tray:

• A clamshell costs $1.08 to $2.00 in labor and median seed alone, before the packaging it goes in.
• Twenty prospects at two clamshells each is $43 to $80 of unrecovered cost.

That is a modest number and it is not the point. The point is that it is the only half of the equation anyone has published. You can price the sample and you cannot price the conversion, so a sampling campaign is a cost with a made-up return attached to it.

The one piece of that advice with better support is when to arrive. NC State advises calling first and avoiding meal service, suggesting 9 to 11 a.m. and 2 to 5 p.m. That is extension guidance rather than a tested tactic, and it is still better sourced than the sample tray.

What the ranking pages get right

They are right about the operational core, and it is not supplier rhetoric. The buyer survey names volume consistency, logistics, quality, specifications and communication as real obstacles, and the extension guidance says the same. A kitchen needs product that matches its use, arriving when it said it would.

They are right that live product is a genuine option rather than a gimmick, which the format split above supports.

They are wrong about the premium. The claim is some version of "restaurants pay premium prices and provide reliable high-volume accounts". This site's own price table does show restaurant wholesale above grocery wholesale, at $11.01 against $4.16 to $6.17 per 100 g. What is not established is that the channel is more profitable, because no source pairs a restaurant price with the delivery, sampling, invoicing and unsold stock that channel costs. The one measured account, the delivery arithmetic above and a 58 percent price objection all point the other way.

And the exposure is real. The national grower survey measured average refrigerated time from cut to sale at 14.6 hours with a standard deviation of 14.1, which is a practice so unconverged it barely has a center. A cancelled standing order lands on product that is already cut.

What to actually do

• Plan on about one tray a week per account. It is the only measured figure and it is far below what the guides imply.
• Build the route before the price. Delivery on a route of one takes 42 percent of an account and on a route of five takes 8.5. Nothing you can negotiate on price moves that far.
• Sell the second account in the same neighborhood as the first, for the same reason.
• Agree a standing order before you sow, and give a truthful last-order date and availability window rather than growing on request.
• Ask each kitchen what documents it needs before promising any of them. Most buyers accept a written attestation and some do not.
• Ask the payment term explicitly and plan four unpaid deliveries. Net 30 was the most common answer and a third of buyers gave a different one.
• Weigh your own trays for the crops you plan to offer. For 23 of the 62 crops here, including generic red beet, no published yield exists to check an order against.
• Record every sample, order, change, payment and unsold unit, so the account is judged on what it actually returned rather than on its list price.
• Treat every figure on this page as a snapshot with a place and a date attached. Prices, wages and delivery rates move faster than anything else on this site.

What nobody has measured

• A representative US survey of what chefs buy, by variety and quantity. The one restaurant-side measurement is about seven kitchens in Sonora in 2017.
• Single crops against mixes, as a share of restaurant orders, from any source.
• A sample-to-order conversion rate, or a sample cost per account won.
• A matched price comparison, same crop, same place, same day, across restaurant, farmers market and retail.
• A realized restaurant price after samples, spoilage, unsold stock, delivery and fuel.
• Why microgreen accounts end, and how many survive a change of chef.
• Seasonal restaurant demand for microgreens. Food away from home is seasonal overall, easing after December to a June peak, and that cannot be turned into an order calendar for one crop.
• How many accounts one grower can service, which is a route and order-size question rather than a headcount.

Sources

Opened 2026-08-12. Every conversion and every share on this page is arithmetic performed for this page on the published figures below, at 28.3495 grams to the ounce. The median tray yield of 7.055 oz was computed from the yield field on all 62 variety entries here, 39 of which publish one, and checked against the count on this site's yield article. The labor and price figures are this site's own published guides, cited in the text.

• Yanes-Molina et al., microgreens as an alternative of horticultural production and market (PDF) - the 2017 survey of 52 gourmet restaurants in Hermosillo, Sonora, and the only restaurant-side measurement of quantity found anywhere: 13.46 percent then using microgreens, average weekly use of 214.29 g with a standard deviation of 69.01 and a mode of 200 g, the named crop preferences, and the 71.43 to 28.57 split between cut and living product. About seven current users, one city, nine years ago.
• Misra and Gibson, characterization of microgreen growing operations, Food Protection Trends, January 2021 - the national survey of 176 US-serving growers: 21 percent selling living trays, 56 percent hand-cutting, and average refrigerated time from cut to sale of 14.6 hours with a standard deviation of 14.1 among 92 respondents.
• 2024 New York Food Buyer Trends report (PDF) - the 80-buyer barrier ranking led by price and budget at 58 percent, the payment timing distribution among the 25 buyers who reported it, and the finding that 69 of 80 would accept a written attestation instead of a certification. Food buyers broadly rather than a restaurant-only microgreen panel.
• NC State Extension, marketing to restaurants - the ordering cut-off mechanism, the 7 to 10 day notice of a supply change, the warning about restaurant staff turnover, and the guidance to call ahead and avoid meal service. Last updated 2010-01-16.
• NC State Extension, buying local, a guide for retail and wholesale buyers - what buyers ask for on invoicing, insurance, food safety, delivery and packaging specifications, and the distinction between a certified farm and one with a plan and no paid certification.
• Cornell Small Farms, using your bookkeeping system with wholesale customers - the invoice baseline, the definition of Net 30 used here, and the note that established buyers may not change their billing cycle.
• National Restaurant Association, 2026 hiring and staffing research (PDF) - 185,989 chef and head cook jobs against 136,605 separations in 2025, a 73.45 percent occupation-wide turnover rate. Not a measure of supplier account attrition.
• USDA Economic Research Service, food away from home seasonality, 2025-02-10 - the seasonal pattern of US food-away-from-home sales, cited here only to explain why it cannot be converted into a microgreen order calendar.
• Institute of Culinary Education, on microgreens - the kitchen-side trade-off between living product's usable life and the bench space and cutting it costs.
• Seller pages and grower accounts, used as practice only and as the source of every posted price, minimum and delivery rate above: Phoenix Microgreens order terms, the $10 flat delivery rate, the half-tray weekly minimum and the 2 to 3 week first-order lead time; MicroFreshly wholesale pricing, the $15 per 4 oz base with 20 percent off for approved chef accounts and a $150 order minimum; Foodomarket rainbow microgreens, the New York wholesale quote of $24.95 per 8 oz on 2026-02-12; Hubb Microgreens pricing; The Leaf Origin wholesale page, its order cut-off model; What Chefs Want local produce list, March 22 2026 (PDF), a distributor catalogue and not demand data; Fresh Origins chef FAQ; Native Roots Farm chef testimonials, attributed chef statements hosted by the supplier receiving them; Lucky Leaf Gardens oral history, UNC Charlotte, the fixed harvest and delivery days; and the premium and sampling claims from On The Grow and its selling guide, Farmzz and Microgreens Consulting, each of which sells a course, consulting or software alongside the advice.