A2 Corporation's first US entry was built on licensing: another company's brand on the carton, a supplements group as its American partner, and a disease-risk science story as the pitch. It sold for less than a year in the Midwest before the company ended its US licences and went home. When a2 came back in 2015, it reversed almost every one of those choices. A teardown of what the first attempt got wrong - and why the fix was a redesign, not more patience.
"In 2008 there was a change in our strategic direction to shift from a licensing model to a branded product model - where we would have more direct control over its marketing, selling and product quality. Hence, we ended our US license agreements."
- Jim Smith, US marketing director, The a2 Milk Company, 2015
A2 Corporation was founded in New Zealand around a single scientific idea. Cow's milk contains two main variants of a protein called beta-casein, A1 and A2. The founders, scientist Corran McLachlan and investor Howard Paterson, believed A1 was linked to health problems, and that milk from cows carrying only the A2 variant avoided them. The company patented a DNA test to identify A2-only cows and set out to license the idea to the world.
That last part matters. In its early years A2 Corporation described itself as "committed to licensing A2 milk in the major world markets." It owned the intellectual property. Others would produce, brand and sell the milk.
Today The a2 Milk Company is one of the most successful consumer brands ever to come out of New Zealand, built mostly on Australia and China. But its first attempt at the US was a retreat, and the reasons for it are one of the clearest lessons available for any NZ company with a science-led product.
The US looked like the obvious prize. It is a vast dairy market, and a premium milk with a health story looked like an easy fit. A2 Corporation had a patent, a test and a body of research it believed in. Licensing looked like the capital-light way to get there: find a well-connected US partner, let them do the heavy lifting, collect royalties.
In September 2003, after both founders had died, the company announced a strategic partnership with IdeaSphere Inc., a US natural health company, to bring A2 milk to America. IdeaSphere's chairman was David Van Andel, a board member of Amway's parent company. Its management included a former Amway chief operating officer and the self-help author Tony Robbins. That same month IdeaSphere bought the supplements maker Twinlab and continued acquiring in the supplements market.
On paper, it was a partner with money, marketing talent and a mission around "science-based products." The press release described A2 milk as a way for consumers "to avoid a known cause of disease."
The partnership moved slowly. In June 2005, A2 Corporation and IdeaSphere formed a 50/50 joint venture, a2 Milk Company LLC, to develop North America.
In April 2007, the JV finally got product on shelves - but not under its own name. It licensed the rights to The Original Foods Company, whose branding the milk would carry, and the product launched through Hy-Vee, a Midwest supermarket chain with around 200 stores and US$4.6 billion in turnover. A2 Corporation described the arrangement as a test market.
The feedback was framed positively. The licensee's product won "Best New Product of 2007" in the special needs category from Dairy Foods magazine. In June 2008, A2 Corporation said the JV was "making good progress understanding the US market" and had learned about consumer response to publicity, price points and a new value-added milk.
Then it stopped. In 2008 the company shifted strategy from licensing to owning its brand, ended its US licence agreements and focused on Australia, which was starting to work. Its 2009 annual report said the JV had regained all US rights through a settlement with Original Foods. In 2010 A2 Corporation bought out IdeaSphere's stake. The product had been on US shelves under licence for less than a year.
a2 did not return to the US until April 2015 - this time through a 100%-owned subsidiary, under its own brand, in California, with $20 million committed over three years.
The first US attempt did not fail because Americans rejected the milk. The company itself said the early consumer response was positive. It failed because the entry model handed away the only things that could make a new milk category work.
a2 milk looks and tastes like milk. What the customer is buying is a reason to pay more for it. That reason has to be built through brand, consumer education and sustained marketing - exactly what A2 Corporation later did in Australia, where its own 2008 results credited "a large investment in brand building advertising."
In the US, the licensing model put that job in someone else's hands. The milk carried Original Foods' branding. The marketing budget, the message and the retail relationships belonged to a licensee with its own priorities. Any brand equity built in the Midwest accrued to someone else's label.
From New Zealand, licensing looks efficient: no capital, no US staff, royalties in the post. In a category that has to be created from nothing, it means paying someone else to build a brand you will never own - and trusting them to spend enough to build it at all.
IdeaSphere shared A2 Corporation's view of the science and brought money and marketing credentials. What it did not bring was a dairy business. Fresh milk is a cold-chain, low-margin, high-volume product sold through supermarket dairy buyers. IdeaSphere's growth came from supplements.
The consequence was time. It took two years to form the JV and nearly four years from the partnership announcement to get product on shelves - and even then, only through a further sub-licence to a separate brand owner, in one regional chain.
This is a pattern that recurs in NZ-to-US entries: the partner is chosen because they are enthusiastic about the product and well-connected, rather than because they already move the product category through the channel that matters. Enthusiasm does not get milk into a dairy case.
A2 Corporation's early language was about avoiding disease. Its 2003 US announcement called A1 "a known cause of disease." Its 2008 results talked about consumer awareness of a2 milk's benefits "for medical conditions such as heart disease, autism and the reduction in the incidence of childhood diabetes." The US launch product was recognised in a "special needs" category.
In the US, that framing is a trap. Food labelling is regulated by the FDA, and a claim that a food reduces the risk of a disease is a defined "health claim" governed by statute and FDA regulation. A claim the science community disputes is a liability, not a marketing asset. It also narrows the audience to people with a medical worry - a niche, not a mass market.
The 2015 relaunch shows what changed. a2 came back selling comfort, not disease prevention: milk for the roughly one in four US consumers the company said experience digestive discomfort after drinking milk, under the line "The milk that might change everything." Same cows, same protein, very different reason to buy.
THE FIRST ATTEMPT
How long a2 milk was sold in the US under licence before A2 Corporation ended its US licence agreements in 2008
Four years of partnership-building produced less than a year on shelf, in one regional chain, under another company's brand. It took seven more years - and a completely different entry model - to come back.
FAILURE DIMENSION ANALYSIS - A2 CORPORATION (FIRST US ENTRY)
The decision to pull out of the US was not triggered by a single bad number. It was triggered by a better one somewhere else. In 2007-08, A2 Corporation's Australian volumes nearly tripled on the back of its own brand-building advertising. The company reviewed its strategy, prepared a capital raising, and made the call that its own brand in Australia was worth more than a licence in the US.
That was the right call. But it revealed the flaw in the US design: the company learned in Australia, with its own brand and its own marketing, the lesson the US licensing model could never have taught it.
a2's first US attempt is one of the rare cases where the company's own later behaviour is the verdict. When it came back in 2015, it used a wholly owned subsidiary instead of a joint venture, its own brand instead of a licensee's, a dedicated budget instead of royalty economics, and a digestive-comfort message instead of a disease-risk one. By 2018 it said a2 products were in around 9,000 US stores.
The science did not change between 2008 and 2015. The entry model did. That is the whole lesson: the first attempt was not too early, and the US was not the wrong market. It was the wrong architecture for creating a new category in a market as large and competitive as the US.
If you are creating a category, own the brand. Licensing works when the category already exists and the licensee has a reason to push your product. In a new category, the brand and the consumer education are the business. Do not hand them to a partner and hope.
Choose partners for their channel, not their conviction. The right US partner already sells your category through the channel you need. A partner who believes in your science but has never sold your kind of product will spend years learning what an operator already knows.
Rewrite the claim for the US before you enter. What you can say on a label, and what the market will believe, is different in the US. Test whether your core claim is permitted, defensible and broad enough to reach a mass market - and if it is not, find the version that is.
Treat the home market as the lab. a2 figured out its winning formula in Australia and then took it to the US. Many NZ companies do it the other way round, and the US is an expensive place to experiment.
A2 Corporation's first US entry is a warning about capital-light thinking. Licensing felt like the low-risk option because it needed little money. In reality it was high-risk, because it gave up control of the brand, the message and the pace - the three things that decide whether a new category takes off.
A pre-entry architecture would have asked three questions in 2003. Who owns the brand the US customer will see, and who pays to build it? Does our partner already sell milk through US supermarkets, or will they be learning on our time? And is our core claim one we can legally make and the mass market will act on?
Answered honestly, those questions point to the model a2 used in 2015. The company got there eventually. It just took seven years and a retreat to find out.
"In a new category, the brand and the reason to buy are the business. License them away and you have not entered the US. You have paid someone else to try."
- PIVOTAL CATALYST VERDICT
FREQUENTLY ASKED
Did a2 Milk fail in the US?
Its first attempt did. A2 Corporation's US joint venture sold a2 milk under licence, under another company's brand, in a Midwest test market for less than a year from 2007. In 2008 the company ended its US licences to focus on Australia. It re-entered the US in 2015 with a wholly owned subsidiary and its own brand.
Why did licensing not work for a2 in the US?
Because a2 milk was a new category that depended on brand building and consumer education. Under licence, the product carried the licensee's branding, and the marketing and retail relationships sat with the licensee. The company later said it switched to a branded model to get direct control over marketing, selling and product quality.
What changed when a2 re-entered the US in 2015?
Almost everything about the entry model: a 100%-owned US subsidiary, its own a2 brand, a committed $20 million budget over three years, launch in California through major grocers including Kroger, Safeway and Whole Foods, and a message built around digestive comfort rather than disease risk.
Go In Knowing
The US Market Entry Diagnostic identifies the structural issues most likely to determine whether your move works or stalls — before you commit capital to the wrong plan.
Start the conversationNZD $10,000. Fee credited in full toward the Architecture Engagement if you proceed.