Loading page…
A $6M Indiana five-gallon water bottler that traded 30,000 delivery customers for about 15 co-packing accounts, then nearly broke itself on a robot install and an in-house trucking fleet.
~2M five-gallon bottles a year · 40 truckloads/week against capacity for 80 · needs ~$600K to become cash-flow positive
"If you're upside down on half the business, you may find you're upside down if you double it": the host's warning to a $6M water bottler planning to go from 40 to 80 truckloads a week before knowing which customers make money.
Fewer bars = easier, cheaper, or faster for an AI-assisted solo builder. Editorial judgments based on the case details.
Jay Peterson has spent 33 years in route businesses. In 1993 he and his wife Myra started as a Canteen vending distributor in Kankakee, Illinois, with one route truck and two machines; he says the vending company grew to about $10 million in sales before he sold it in 2001. Around 1996 he added five-gallon water and coffee after noticing water trucks at his own vending locations, invested as a small partner in the Wisconsin Glacier Springs plant, and sold the water and coffee routes to Van Houtte's FilterFresh around 2004. Mountain Glacier began in December 2006, when Jay bought three Nestlé routes and their customer list in Evansville, Indiana, for about $225,000 while they did about $1.1 million in revenue. Toyota's Princeton plant and an Alcoa can plant were the prize accounts. Over time the business shifted from serving about 30,000 delivery customers to co-packing for roughly 15: three very large customers and 13 regional ones, with Culligan a customer of 20 years. Jay reports about $6 million in sales and about 2 million bottles a year, running 40 truckloads a week through a plant built for 80. The last year nearly broke it. A robotic loading system installed in June took seven months instead of two weeks to work, forcing round-the-clock overtime through a hot summer and a new St. Louis contract. Jay put about $1 million of personal money in, still needs about $600,000 to clear payables and short-term debt, carries three SBA loans, and is selling off the trucking fleet he built for deliveries. The host's advice was blunt: build a 13-week cash forecast and customer-level profitability before chasing more volume.
Mountain Glacier's first customers were bought, not cold-called. Jay was already servicing vending accounts when he noticed five-gallon water trucks pulling up to the same buildings, so he added water and coffee to his refreshment routes.
The quick-reference card — acquisition channels, replication playbook, risks & traps, and our verdict — is members-only.
Join the Unicorn Club — $5/moCancel anytime, in one clickPaste into Claude Code / Codex and get a working version of this product end-to-end.
Two production-grade prompts per idea — a full build spec and an SEO growth plan — plus 10 tailored build specs a week. Members only.
Join the Unicorn Club — $5/moCancel anytime, in one clickEvery claim traces back to a listed source — open any of them and check the number yourself. The exact words are quoted below, with a timestamp that jumps to the second each figure is said.
“Yeah, our business now is doing about 6 million in sales.”
42:07 · hear it in the videoData credibility: Jay Peterson, Mountain Glacier's owner, gives these figures in an advice-style interview on David Heacock's Boring Money channel: about $6M in sales, about 2M bottles a year, 40 truckloads a week against capacity for 80, the 2006 purchase of three Nestlé routes for about $225,000 when they did about $1.1M, about $1M of personal money put in last year, a $600,000 immediate cash need, and three SBA loans. All are founder-reported; no financial statements are shown, and he says monthly cash tracking is only now being set up. The ~$10M figure belongs to his earlier vending company, not Mountain Glacier. Margins, per-customer economics and the split between bottling and freight revenue were not given.
Jay's first route dollars came in 1993 from two vending machines and a single truck, run as a Canteen distributor paying a 3% fee; landing a large Sears distribution center was big enough that he and Myra cancelled their honeymoon. Mountain Glacier's money now comes from contract bottling.
The early flywheel was buying density cheaply.
Mountain Glacier's hard stretch was a surge it could not absorb. A robotic loading system, two years in the making, was dropped into the plant on June 9, just as a new St.
Mountain Glacier is a contract bottler of five-gallon returnable water, plus a trucking arm it is trying to exit. Municipal water is piped into the Evansville plant and run through carbon filtration and softeners, then reverse osmosis into holding tanks, five-micron filtration, UV and ozone treatment, and mineral tanks that blend it to each customer's target dissolved-solids level. Customer-owned bottles arrive by the truckload, go onto a horseshoe production line with robotic loading and unloading, through a washer, fill room and capper, get Videojet date codes, and wait 24 hours for batch testing before shipping. A full truckload is 920 bottles; the plant runs about 40 loads a week and can run 80. Our read: the real product is trust and capacity. A brand like Culligan is outsourcing its label to Jay's plant, so certifications, line checks and on-time loads matter more than the water itself. Jay's own conclusion is that customers no longer care whose truck delivers; they want water at their dock, which is why he wants to be excellent at production and hand freight to specialists such as JB Hunt or a local carrier.
Revenue comes from bottling for about 15 customers, three very large and 13 regional, and Jay reports about $6 million in sales on about 2 million bottles a year. Its in-house trucking company moves about 2,000 loads a year, which Jay now wants to hand to a dedicated carrier. Separately, Jay owns buildings that customers lease long term, including a Culligan site in Indianapolis and a paid-off Texas building with a Primo tenant; he treats that rent as personal retirement income rather than business revenue. Our read: co-packing simplified the business enormously (Jay went from about 30,000 delivery customers to about 15), but it also concentrated it. When a large customer re-ran its network analysis and moved St. Louis volume to a Kansas City plant, Jay had about 10 days' notice and two dedicated drivers with nothing to haul. Without monthly closes or per-customer margins, he cannot yet say whether going from 40 to 80 loads would make money or simply double the losses. The planned upside, 4 million bottles and an exit from freight, depends on answering that first. Culligan, a 20-year customer, is also his fallback buyer, though Jay says selling is not what he wants.
The cost base is heavy. The plant needs treatment equipment, a filling line, a washer, a Kuka robot cell and compliance systems; at peak the freight arm ran 12 day cabs, one sleeper and about 30 trailers. The robot vendor gave a $300,000 concession in warranty, free service, spare parts and a discount on the final 15%, but Jay says the real cash outlay was his. He financed growth through three SBA 7(a) loans: about $1.7 million on the Indianapolis building, $700,000 on trucks, and about $1.1 million for the robots and working capital. After a bad robot year, the covenants and blanket collateral left him unable to borrow against buildings with equity, so he has raised cash by selling assets instead, including eight trucks sold back to Culligan for about $420,000, with two bottling systems and a distiller still for sale. His insurer wanted another $90,000 at renewal before he said he was exiting trucking. Books run on QuickBooks Online. Our read: none of these costs is unusual; the problem is that they sit in one blended P&L. Split plant, freight and real estate into separate ledgers, and price freight per lane with a fuel surcharge, as the host does, so each customer's true contribution is visible.
Copy the acquisition instinct. Buying route books and customer lists from big operators exiting low-density markets gave Jay industrial volume at about 22 cents on the dollar, and owning the plant buildings, then leasing space to customers, built him a second income stream. Copy the move to co-packing too: serving 15 accounts instead of 30,000 households is far simpler to run, and certification turns into a sales asset with national brands. Change three things. First, do not run trucking in-house when customers only care that water reaches their dock; the host's advice was to rip off the band-aid, auction trucks and trailers in one go, and negotiate dedicated lanes annually rather than signing three- to five-year freight contracts. Second, never let a new machine replace a working process during peak season; keep the manual line until the automation proves its uptime. Third, close the books monthly and run a 13-week cash forecast before borrowing or adding volume. As the host put it, clarity of cash flows is also what makes lenders and buyers comfortable writing a check.
Concentration is the first risk: three very large customers drive most of the volume, and one already moved a lane to another plant with about 10 days' notice. Leverage is the second. Three SBA loans with blanket collateral leave little room for another bad year, and Jay says he still needs about $600,000 just to cover payables and short-term debt at current volume. Third, the trucking exit is slow and expensive while diesel sits near $5 to $5.80 a gallon and the freight market is tight; every month of delay adds depreciation, insurance and management attention. Fourth, automation can fail again: the host, who runs his own multi-site manufacturing company, says new machines almost always take longer than vendors promise. Our read: the biggest risk is flying blind. Jay admits the business has been filling orders without month-by-month cash visibility. Until customer profitability and a 13-week forecast exist, doubling to 80 loads could double the problem. The safety valve is Culligan, which has invested heavily in five-gallon delivery and wants the plant, but a forced sale to the anchor customer is a weak negotiating position.
The four founder playbooks — first customers, first dollar, the flywheel, the silent stretch — are in the Unicorn Club, on this idea and on all 1,403 others. The story, the numbers, the sources and every later section stay open to everyone.
Join the Unicorn Club — $5/moCancel anytime, in one click