Birch wants $10 million a year online within three years. This plan gets there without asking Birch to carry a loss: on our planning numbers the program has paid for itself by month 12, and Birch never has more than about $110,000 in at any one time. It starts with the people who already buy Birch every morning, and three numbers decide every step up in spend.
Every number from outside our agency sits on a catalogue card, like the ones in a library drawer: the real page it came from with the figure highlighted, what kind of source it is, what it covers, and a link. Tap a card's picture to read the page up close. Planning ranges are marked as planning ranges.
Well over $100 million in Facebook ad spend has passed through our hands. We've run campaigns for Warner Brothers, Tony Robbins, Robert Kiyosaki and Gary Vaynerchuk, and for supplement, consumable and skincare brands that live or die on repeat orders, just like coffee.
We're a boutique team of specialists, and we stay small on purpose. We pay to keep the best people and keep every client list short, because scaling a brand fast takes a kind of focus that a big team of mediocre hires, each juggling too many clients, can never give. We would rather have seven clients who stay for years than fifty who leave within months.
We're a Facebook Managed Partner Agency. Facebook published a case study of our client work: a life coaching institute spending more than $250,000 a month on ads, earning a 5x return on ad spend and a 33% lift in sales. Facebook also invited us to a closed-door event for its top agencies. Our specialists have run campaigns at $75,000 a day and more.
Mastering return on investment Facebook for Business, Success Story
5X return on ad spend; 33% increase in sales; 1.5 million potential customers reached a month; 2X email subscribers
A life coaching institute's Facebook ads, run by our team. Monthly spend of $250,000+ is from our own records.
Infinite Age, a supplement brand, couldn't get past $3,500 a day in ad spend: every attempt to spend more killed the margin. We rebuilt its offer around what a customer is worth over a year, gave each kind of buyer its own message and page, and restructured the account so winning ads kept running. Four months later it was spending $45,000 a day, and its Shopify store sold $3.46 million in those four months, 77% more than in the period before. Across every channel it sold, revenue reached $7,254,984 over the same four months, a blended return of $2.17 for every dollar of ad spend.
“You guys have been the game changer.”
Russ Symes, founder and CEO of Infinite Age Message to Robert Grant, 3 August
“Yes. You can quote me. You guys have been the game changer.”
Shown with his permission. The spend and Shopify figures above are from our Infinite Age case study: Shopify store sales, 4 April to 4 August, against the four months before; blended revenue and return across all sales channels for the same four months.
James Hagerty founded
Presto Coffee in 2018 and took it from 41 supermarket stores to 1,500, while it sold on Amazon at the same time. He knows what a retailer asks for before it says yes. He has worked with us for years, and he is part of the team on this project.
Presto Coffee investor deck: roadmap Presto Coffee, shared by Robert Grant, 13 Nov 2026
From 41 supermarket stores to 1,500; Waitrose Oct 2025, Tesco Nov 2025, Co-op and Sprouts Mar 2026, Morrisons Aug 2026
A private investor deck; store count on page 2, the roadmap shown here is page 3.
Two in three American adults drink coffee every day, about three cups each. Of the people who had coffee yesterday, 71% made it only at home, up from 63% in 2020. And the share of them who bought coffee online doubled in five years.
Yesterday's coffee drinkers who bought coffee online.
National Coffee Association.
National Coffee Data Trends, Spring 2025 National Coffee Association, 15 Apr 2025
66% of adults drink coffee daily; 71% of yesterday's drinkers made it only at home; online buying doubled from 7% to 14%
US adults, survey fielded January 2025. Shares of people, not of sales.
Read the release
Black Rifle Coffee is the nearest thing to a public scorecard for coffee sold online. Its own-site sales fell from $143.2 million in 2023 to $117.6 million in 2025, and its subscribers from 225,800 to 159,900. Over the same two years its sales through stores and distributors grew from $225.1 million to $258.0 million.
Computed from the two filings: own-site sales down 17.9%, subscribers down 29.2%, sales through stores up 14.6%, in two years.
Online sales are earned one subscriber at a time, even for a famous brand with years of ads behind it. So this plan doesn't open with a revenue promise. It opens with what one Birch subscriber is worth.
Swipe sideways to read the whole table.
BRC Inc. fourth quarter and fiscal 2024 results (8-K, exhibit 99.1) Black Rifle Coffee, 3 Mar 2025
Direct to consumer $143,232k in 2023, $123,779k in 2024; 225,800 subscribers at end of 2023; wholesale $225,059k in 2023
Full years 2023 and 2024, dollars in thousands as filed.
Read the filingSwipe sideways to read the whole table.
BRC Inc. fourth quarter and fiscal 2025 results (8-K, exhibit 99.1) Black Rifle Coffee, 2 Mar 2026
Direct to consumer $117,638k; 159,900 subscribers; wholesale $258,005k
Full year 2025, dollars in thousands as filed.
Read the filingBirch's own guide pairs one cup a day with one 12oz bag every two weeks. On the subscription, the first two bags cost $14.25 and every bag after that $16.15. That's about $416 a year for someone who stays twelve months, before shipping.
So $10 million a year means about 24,000 subscribers who stay. That's 15% of Black Rifle's base, and almost ten times what Equator Coffees built in six years (it hoped for 100 subscribers in 2016 and had 2,500 by 2022). It can be done, but not in one jump. So we split it into three years, and each year has to earn the next.
By the end of year 1: about 7,700 subscribers, a $3 million a year pace. Year 2: about 19,000, $8 million. Year 3: 24,000 or more, $10 million. Larger orders, gift subscriptions and one-off shop sales count toward the same total, so subscribers are the floor, not the ceiling.
Subscribe & Save: how often birchcoffee.com
One cup a day: one 12oz bag every 2 weeks (a bag makes about 24 cups)
Birch's own frequency guide, opened from the subscription page.
See the guideSwipe sideways to read the whole table.
Subscribe & Save birchcoffee.com
25% off the first two orders, then 15% off from the third
Subscription terms shown on the shop today. Skip or cancel any time.
See the offerSubscriptions drive digital growth for Equator Coffees Digital Commerce 360, 28 Apr 2022
Hoped for 100 subscribers in 2016; 2,500 about six years later
From the article's public opening; the rest is paywalled.
Read the articleOur own store database lists 20,855 Shopify stores. Below are three well-known coffee brands from it, with each one's position in the list, what its own site says it sells, and how many ads it had live on Facebook and Instagram on 15 November. Birch isn't in the list, and it has no ads live.
All three run ads on Facebook and Instagram right now. Birch runs none.
Swipe sideways to read the whole table.
DTC Brand Database: 20,855 Shopify stores by niche Reach Global Media (Bogdan Radu), compiled around February 2026
Coffee brands and their positions in the list; birchcoffee.com is not in it (a search for 'birch' finds only Birchbox and The Birch Cottage)
Position is the store's rank in this list. The list does not say what the rank measures and holds no ad data.
Open the listTap the screenshot for more of this source.Meta Ad Library: Atlas Coffee Club Facebook and Instagram, US
About 1,100 ads live
Ad versions running in the US on 15 Nov 2026. A count of ads, not of spend.
Open the libraryMeta Ad Library: Cometeer Facebook and Instagram, US
About 320 ads live; this one offers up to 50% off the first order
Running since 2 Nov 2026. Counts move daily.
Open the libraryBirch already has what most online coffee brands spend years building: 15 cafes full of regulars, shelves in 19 Whole Foods stores and select Costcos, and a subscription built into its shop. What it doesn't have yet is traffic. Today Birch runs no ads on Facebook or Instagram.
Meta Ad Library: Birch Coffee Facebook and Instagram, US
No ads live
Active ads in the US on 15 Nov 2026. Meta keeps no history of ended US ads, so past spend is unknown.
Open the libraryEverything a subscription business needs is built. What's missing is the traffic.
Before we write a single ad, we read what coffee drinkers already say about Birch and the brands around it: reviews, Reddit threads, comments. Consilium, our research tool, sorts it into themes and weighs each one by how often it comes up and how strongly people feel. One angry review is one bad morning. A complaint that shows up in hundreds of reviews is the market talking.
We read four-star reviews hardest. A four-star review says exactly what would have made it five, and that gap is usually the offer.
Almost every coffee brand loses money on a new customer's first order. That part is normal. Brands fail when they never earn it back.
Coffee Brand Unit Economics: A 2026 Operator's Guide Eightx, 18 Jun 2026
Ad cost to win a food or drink customer: $45 to $53. On a $60 order the first order is minus $17.80; about 28% of cancellations come in the first three months
The page calls these 'planning bands, not category averages'; the worked example is illustrative.
Read the guideTap the screenshot for more of this source.The ad cost to win a food or drink customer runs $45 to $53 on Eightx's planning band. At Birch's prices a new subscriber's first three bags sell for $44.65 together (computed).
Eightx's worked example assumes a $60 order. One Birch bag is $19 to $23, so a single-bag order carries the full cost of winning the customer on a third of the money.
About 28% of cancellations come in the first three months on Eightx's planning band, and across 15,000+ subscription stores only 45% of subscribers were still there at six months, 33% at twelve. Birch's deepest discount, 25%, sits on the first two orders.
Our read, not a statistic: ad platforms credit themselves generously. We count real orders in Birch's own store and match them to each ad.
State of Subscription Commerce 2023 Recharge
45% of subscribers still active at 6 months, 33% at 12 months
2022 data from 15,000+ stores on Recharge, all product types, not coffee alone.
Read the reportThe brands that make it are patient. Bean Box, an online coffee club, took more than a decade to get past half its revenue from subscriptions. Its average subscriber stays almost two years, and its CEO says new customers are “incredibly hard to come by.” Black Rifle, with all its fame, had 65,900 fewer subscribers at the end of 2025 than at the end of 2023 (computed from its filings).
With high acquisition costs, Bean Box turns to its members for growth Chief Marketer, 5 Sep 2024
More than 50% of revenue from subscriptions, after more than a decade; the average subscriber stays almost two years; new customers 'incredibly hard to come by'
Bean Box CEO Matthew Berk. Bean Box is an online coffee club.
Read the articleTap the screenshot for more of this source.Most agencies take a share of ad spend: often 20 to 30% when spend is under $5,000 a month, and 10 to 15% above $50,000, plus a monthly minimum of $1,000 to $5,000. In a 2019 survey, half of search-ad agencies set that minimum at $1,000 to $3,000. Our fees sit inside that range.
So the difference isn't the price. It's what Birch gets for it, and how long the work keeps paying. A 2026 study found 25.9% of agency clients leave every year, after about 40 months on average. We plan to give Birch no reason to look elsewhere.
Agency Pricing in 2026 Swydo, updated 20 May 2026
Paid media fees: 20-30% of spend under $5,000 a month, 10-15% above $50,000; minimums $1,000 to $5,000 a month
A vendor's guide. The page names no source for its fee table.
Read the guidePPC pricing: how much do PPC services cost? Credo, 2019 survey, page updated 28 Jun 2022
50.97% of PPC firms set a monthly minimum of $1,000 to $3,000
Survey of US search-ad agencies; 2019 data.
Read the surveyAverage Marketing Agency Churn: 2027 Report Focus Digital, 11 Nov 2026
Retainer clients: 25.9% leave each year; 40.4 months on average
Agencies of all kinds; sample size not given. Lifespan is computed from churn.
Read the study
You work with the people you meet. Many agencies send their best people to win the account, then hand the work to a junior on a small salary with little experience. Not here. The people you meet are the people who do the work:
Robert Grant
Bogdan Radu
Yan Gozalvo
James HagertyBehind these four stand our video editors, designers and a customer success manager who keeps every deadline on track. But these four make every decision on your account, and these four answer to you for the results. Meet them before you sign.
Three years, three stages, each one earned by the last. Every step up in budget is decided by the same three numbers, agreed with Birch before the first dollar.
Turn cafe regulars into subscribers, then a paid test mostly in New York. Fix the offer so subscribers reach the third box. Scale only the arms that pass all three numbers.
Take the winning messages and pages nationwide. Add bigger orders (two bags, the 2lb bag), gift subscriptions for the holidays, and new ad angles every month from what drinkers say.
Keep subscribers longer, win back the ones who leave, and push harder wherever Birch's shelves and cafes already are. By now most growth comes from customers ads won in years one and two.
Our planning path, not a promise: revenue pace is subscribers times about $416 a year at Birch's subscription prices. Every number resets with Birch's real data after the first 90 days.
Ad spend starts at $15,000 a month and climbs $5,000 a month, to a ceiling of $150,000. Climbing slowly keeps the cost of winning a first sale near $43; past $150,000 a month it rises, because bigger budgets reach colder buyers. That pace means Birch never carries a big hole.
About $110,000, at month 6. That includes ads, creative, landing pages and our fees.
Month 7. From here each month's subscribers pay for the next month's growth.
By month 12. Year 1 ends about even; year 2 returns about $1.1 million and year 3 about $2.7 million.
Search Amazon for “birch coffee” today and you won't find a Birch bag you can buy. Birch's two old listings, a French Press blend and a Colombian Decaf, both show “Currently unavailable”, and a search filtered to the Birch Coffee brand returns nothing. Meanwhile other roasters, such as Lifeboost, Counter Culture and Brooklyn Roasting Company, pay for the top ads on the search “birch coffee”. People already look for Birch on Amazon, and today other brands sell to them.
Our Meta ads will send more of those people to Amazon. So Birch goes live on Amazon from day one, with branded search ads that protect its own name. Amazon takes time to build: it needs a solid base of reviews, all 4.5 stars or higher, before it sells at full strength, and those reviews have to be earned within Amazon's rules. We plan for it to grow over the first nine months, not to arrive on day one.
The right partners can make this take off quickly from an early start. One partnership brings traffic, content for new ads and Birch's first Amazon reviews at the same time, which shortens the slow early months.
For buyers aged 50 and over this can be especially strong. Well-known authors and actors from their younger years still feel like heroes to this audience, and many will make a short video for a modest fee. We would test this as its own audience, with a clear paid-partnership label on every video and written rights for how the video is used in ads.
Before a big retailer gives a brand shelf space, it asks one question: does it sell? Six months of steady online sales answers it with numbers, not hope. That is how James took Presto from 41 stores to 1,500.
The prize is large even at a slow pace. A bag that sells just twice a day in each store sells 730 bags a store a year.
Planning numbers: at $10 a bag wholesale. Store counts: Target, Q2 2026; Costco, end of 2025.
Birch is already in select Costcos in New York, New Jersey and Connecticut. Going national is the next step, not a leap.
Birch's app, its email list, a card at every register, gift subscriptions. This sets the baseline: how many cafe regulars subscribe, and how many take a third box.
About three quarters of the budget around New York, where Birch's cafes and its Whole Foods and Costco shelves already are; about a quarter nationwide, to learn what strangers do. Six weeks, so the first subscribers from ads reach their third delivery inside the test.
Each arm grows only while it passes all three numbers. The national arm overtakes New York only by beating it on cost.
Why New York first: when a retailer opens a store, online sales in the area around it rise, by 6.9% on average across 2,103 stores in one industry study, and by 13.9% for newer direct brands. People who see Birch on the corner are the cheapest people to reach online. And a cafe roaster that saw a “dramatic increase” in online subscriptions found it came mainly from its own cafes' customers.
The Halo Effect III: Where the Halo Shines ICSC, 12 Dec 2023
A new store lifted online sales nearby by 6.9% on average, 13.9% for newer direct brands
69 retailers and 2,103 stores. Store openings, not ad tests.
Read the studyWholesale coffee brands are testing out DTC Modern Retail, 23 Jun 2020
A cafe roaster's online jump came mainly from its own cafes' customers
Methodical Coffee, South Carolina. No figures given.
Read the articleNo more than Birch's margin on that subscriber's first three bags. At today's prices those bags sell for $44.65; Birch's margin per bag sets the exact line at kickoff.
Subscribers from ads take a third box at least four times in five as often as subscribers from the cafes do in stage 1.
More than half of each arm's subscribers aren't already on Birch's email or app lists. Otherwise the ads are paying for customers the cafes would have brought for free.
Two offers, really. One for Birch's customers, built so they reach the third box without being trained on discounts. One for Birch, built so the risk sits with us.
Discounts bring in bargain hunters who leave when the discount ends. So instead of cutting the price, we add things only Birch can give. These are the pieces we'd test, each agreed with Birch before launch:
Four landing pages at $1,500 each, built before the first dollar of ad spend, so every ad has a page that tells the same story. Three is the minimum; we recommend four.
A $5,000 monthly retainer plus 10% of ad spend. Creative is its own line: we recommend the $10,000 monthly package, about 120 video-first ads a month (the bare minimum is $5,000, which we don't advise). As spend scales, Birch moves to a larger creative package, because more spend needs more ads. Shoots, creator videos and new landing pages are quoted up front, and more landing pages come as spend grows. No share of revenue.
The first term runs until we hit a benchmark agreed with Birch up front. Only then do we renew for six months.
If the three numbers aren't hit by week 9 of the test, Birch pays no $5,000 retainer and no ad spend fee for the following month. We keep working. Creative and landing pages are still billed as normal.
New ads running within 10 days of kickoff. Video ads ready by week 3.
Monthly strategy plan for sign-off, a call every two weeks, one shared scoreboard of ads and email side by side.
One hour to see whether this fits both ways.
We'll bring the first messages we'd test and a draft of the subscription page.
P.S. Every number in this plan sits in a live model. Put in Birch's real costs before our call, and we'll walk through what you find.
Robert Grant, Bogdan Radu, Yan Gozalvo and James Hagerty, Reach Global Media
Everything was checked on 15 November 2026. Figures are quoted with their source and window. Planning ranges are marked as planning ranges. “Computed” means we did the arithmetic from the cards shown. Ad counts are what Meta's public Ad Library showed for active US ads that day; they move daily and count ads, not spend. Photos on the cover and in part 5 are Birch's own, from its website.