The same brand, the same product, the same store format. 172 stores in Australia. 53 in New Zealand. 83 in Canada. Zero in the US — not because Michael Hill ran out of ideas, but because it kept applying the same idea to a market where the underlying conditions were categorically different. A forensic teardown of the product-market fit transfer failure that cost NZ$24.85 million to exit.
“The US business had been struggling with a reported $12 million loss over the previous 12 months.”
— CEO Phil Taylor, announcing the US exit, January 2018
Founded in 1979 in Hamilton, New Zealand by Sir Michael Hill, Michael Hill International grew from a single jewellery store in the Waikato into one of the Southern Hemisphere’s most successful specialty retail chains. The business model was straightforward and consistently executed: accessible mid-market jewellery — engagement rings, wedding bands, gold chains, diamond pendants — sold from high-street and mall locations at price points designed for the aspirational middle-income customer. No pretension, no excessive luxury positioning, no need to be Tiffany.
By the time Michael Hill entered the US market, the brand had proved itself in three distinct markets. Australia was the largest operation — 172 stores — where the company was a genuine category leader in accessible jewellery. New Zealand was the home market, 53 stores. Canada, which Michael Hill entered in 2002, had grown to 83 stores and was profitable. The Canada success was particularly significant: North America was not foreign territory. The brand had demonstrated it could operate in a shopping mall environment with North American consumers. The thesis for US entry seemed well-supported.
The US exit in January 2018 — all nine stores closed by April 30, no buyer found despite a sale process, US$4.5 million in lease terminations and employee severance, and a total exit cost of NZ$24.85 million — was not the result of one bad decision. It was the result of a systematic misread of how different the US competitive environment was from every market the company had succeeded in before.
The US entry logic was defensible on the surface. Michael Hill had a proven format, disciplined operational execution, and a demonstrated ability to run profitable mall-based jewellery retail in English-speaking markets. Canada shared North American mall culture, North American consumer psychology, and North American real estate economics. If the model worked in Calgary and Toronto, the reasoning went, it should work in the United States.
The error in that reasoning is not obvious until you examine the US mid-market jewellery category specifically. What looks like a single category — accessible jewellery in shopping malls — is actually a structurally different competitive landscape in the US than in any market Michael Hill had operated in. Canada is not a proxy for the US. It shares a language and a mall format. The competitive density, the brand recognition requirements, and the customer acquisition economics are different in ways that only become visible when you are already inside the market with leases signed and staff hired.
The exact date of Michael Hill’s US market entry is not publicly documented, but by the time the exit was announced in January 2018, the company had been operating in the US for approximately a decade. Nine stores across US mall locations. CEO Phil Taylor disclosed a NZ$12 million loss in the 12 months prior to the exit announcement, with same-store sales declining a further 10% in the most recent trading update — the deterioration accelerating rather than stabilising. No buyer was found in the sale process. The company wound down the US business entirely, incurring US$4.5 million in lease termination and severance costs as part of a total NZ$24.85 million exit charge. The global business — 308 stores across Australia, NZ, and Canada — continued operating without the US. The exit was described by management as allowing the company to “redeploy capital to markets where we have scale.”
Michael Hill did not fail in the US because the brand was wrong or the execution was poor. It failed because the US mid-market jewellery category is structurally different from every other market the company operated in — and those differences were not adequately mapped before stores were opened, leases were signed, and a decade of capital was committed.
The Canada success created a dangerous inference: that Michael Hill’s model was portable to North America. Canada does share surface-level features with the US — English language, shopping mall culture, broadly similar consumer demographics. But the Canadian jewellery market is a fraction of the US market in scale, and scale changes competitive dynamics in ways that small-market success consistently obscures.
In Canada, Michael Hill competes in a market where Peoples Jewellers, Birks, and independent stores define the mid-market. The competitive set is manageable, the brand recognition gap is closable through consistent mall presence, and the category leaders do not have the marketing budgets or store density that US chains carry. In the US, the mid-market jewellery category is dominated by Signet Jewelers — the parent company of Kay Jewelers, Zales, and Jared — which operates over 2,800 locations across the country and spends hundreds of millions of dollars annually on brand advertising. Helzberg Diamonds and Sterling Jewelers add further density. The brand recognition gap Michael Hill faced in the US was not the same gap it had closed in Canada. It was structurally larger, more expensive to close, and defended by incumbents with scale advantages that cannot be overcome by nine stores.
Nine stores in the US is not a beachhead. It is a pilot that can never generate the density, the word-of-mouth, or the advertising efficiency to compete against a category leader operating 2,800 locations. Michael Hill entered the US at a store count that guaranteed it would always be an unknown brand competing against household names — and in a category where trust and brand recognition are primary purchase drivers, unknown brands do not win on product quality alone.
The economics of jewellery retail are marketing-intensive in a way that most retail categories are not. Jewellery is a high-consideration, low-frequency purchase. Customers shop for engagement rings once. Wedding bands once. Anniversary gifts occasionally. The purchase decision is driven heavily by brand trust — which in the US mid-market is built through decades of television advertising, in-mall visibility across hundreds of locations, and the cultural familiarity that comes from being the brand your parents bought from.
Kay Jewelers’ “Every Kiss Begins with Kay” campaign has been running since 1985. Zales has operated in the US since 1924. These are not brands that a new entrant dislodges with a good product and nine mall locations. The customer acquisition economics for a challenger jewellery brand in the US require either a massive advertising budget — to build the brand recognition that drives consideration — or a distinctive product or experience differentiation strong enough to generate organic word-of-mouth in a low-frequency category. Michael Hill had neither.
The company’s mid-market positioning — which was genuinely differentiated in NZ and Australia where the category was less saturated — read as generic in the US. Without a specific differentiation story that US consumers would seek out, and without the advertising scale to build brand awareness, Michael Hill stores were competing on location and product quality against brands that US consumers already trusted by default.
The most revealing detail in the Michael Hill US story is not the exit — it is the duration. The company operated in the US for approximately ten years. Nine stores. No material expansion. Same-store sales declining in the final year. The question that should have been asked at year three — or year five — is: what is the evidence that this model can scale in this market, and what would it take to reach the store count where the economics work?
A mid-market jewellery retail brand in the US requires geographic concentration to build brand awareness efficiently. Fifty stores in a single metro area generates the mall presence, word-of-mouth density, and advertising ROI that nine stores spread across the country cannot. Michael Hill was in the US for a decade and never reached the store count in any single geography where the model could prove itself. The capital was deployed, but not at the scale or concentration required to test the thesis properly.
This is a specific version of the broader NZ-to-US failure pattern: confusing sustained presence with a working model. Being in a market for ten years does not mean the model is working. It may mean the model is slowly not working — which is more expensive than a faster failure because it ties up capital, management attention, and strategic options for years while the core business waits.
The correct decision structure would have been to define, at year two or three, the specific conditions under which the US model would be judged viable: a target store count in a target geography, a target same-store sales benchmark, a target CAC for new mall locations. If those conditions were not achievable given available capital, the decision to exit would have come earlier, at lower cost, with more strategic optionality intact.
THE COMPETITIVE REALITY
Signet Jewelers US store count (Kay Jewelers, Zales, Jared combined) at time of Michael Hill’s US exit
Michael Hill operated nine US stores against a category leader with over 300 times its store count. Nine stores cannot generate the brand awareness, advertising efficiency, or customer trust required to compete in a category where the incumbents have been on television for forty years.
FAILURE DIMENSION ANALYSIS — MICHAEL HILL
The decision to exit crystallised when the most recent trading update showed same-store sales falling a further 10% — not stabilising, not recovering, but accelerating downward. A 10% same-store sales decline in a mall jewellery business means existing customers are not returning, new customers are not discovering the brand, and the competitive position is eroding rather than holding.
By January 2018, the US business had recorded a NZ$12 million loss in the preceding twelve months. No buyer emerged from the sale process — a revealing signal about how the market assessed the US operation’s standalone value. A buyer would have seen the same competitive landscape, the same store count, the same brand recognition gap, and concluded that the value of the business was negative after accounting for the lease obligations and exit costs. The company wound down entirely rather than transfer the liabilities to a new owner.
The timing of the exit — after a decade, not after a year or two — is itself the structural lesson. Michael Hill stayed in the US long enough to establish that the model was not working, but not long enough to reach the scale at which it might have. The result was neither a successful market entry nor a fast, cheap exit. It was a slow, expensive discovery of what should have been identified as a design problem before the first US lease was signed.
Michael Hill’s US failure is the clearest available case study in NZ-to-US expansion of what happens when a company treats geographic adjacency as market similarity. Canada was not a proof point for the US. It was a proof point for Canada. The competitive dynamics, the brand awareness requirements, the advertising economics, and the scale thresholds for viability in US mid-market jewellery are structurally different from every market Michael Hill had operated in — including the one that shares a continent and a language.
The company had the operational capability, the product quality, and the retail experience to succeed in the US. What it did not have was a US-specific market entry architecture: a realistic assessment of the store count required to achieve brand visibility, a capital plan sized for the advertising investment needed to compete against Signet, and an honest answer to whether that level of capital deployment was justified given the competitive position of the incumbents.
Succeeding in three markets is strong evidence that a retail model works. It is not evidence that it works everywhere. The US is the market that most consistently destroys the assumption that success elsewhere transfers automatically — and mid-market retail with high brand trust requirements is one of the categories where that assumption is most dangerous.
Canada is not the US. Proximity and shared language are not market evidence. Before treating success in one English-speaking market as validation for another, map the specific competitive dynamics of the target category in the target market. The jewellery category in the US is not the jewellery category in Canada — in scale, in competitive density, in brand awareness requirements, or in the capital needed to build consideration against incumbents who have been operating there for decades.
In brand-trust categories — jewellery, financial services, healthcare, professional services — scale is not just a growth metric. It is a prerequisite for credibility. Nine stores in the US does not generate the brand familiarity that drives purchase decisions in a low-frequency, high-consideration category. Before entering a category where incumbents have deep trust advantages, define the minimum viable scale that makes the economics work, and ensure the capital plan can fund that scale before the first lease is signed.
Define the exit conditions before you enter. Staying in a market for ten years without reaching the scale at which the model proves out is not persistence — it is a capital allocation problem. A structured market entry plan should specify, at the outset, the benchmarks at which the US investment is judged viable or not, and the timeline at which the exit decision is made if those benchmarks are not met. That decision structure costs nothing to design before entry. It costs NZ$24.85 million to avoid designing.
Michael Hill’s US failure is not a story about a bad product or a weak management team. It is a story about what happens when a company enters a market without mapping the specific competitive conditions of that market’s category — and then stays long enough to make the discovery very expensive.
A pre-entry US architecture for Michael Hill would have surfaced three questions before the first US lease was signed. First: what is the minimum store count in a single US geography at which the brand becomes visible enough to drive purchase consideration, and can we fund that? Second: what is the brand differentiation story that gives a US consumer a specific reason to choose Michael Hill over Kay or Zales — not a better version of the same offer, but a meaningfully different one? Third: is Canada’s competitive landscape actually comparable to the US’s, or are we using one market as a proxy for another because they share a surface-level feature?
Those questions would not have guaranteed a different outcome. But they would have produced an honest answer to whether the US was the right market, the right time, and the right capital allocation — before a decade and NZ$24.85 million established the answer the hard way.
“Succeeding in three markets is strong evidence that a retail model works. It is not evidence that it works everywhere. The US is the market that most consistently destroys that assumption.”
— PIVOTAL CATALYST VERDICT
FREQUENTLY ASKED
Why did Michael Hill succeed in Canada but fail in the US?
Canada and the US share a language, mall culture, and surface-level consumer demographics — but the mid-market jewellery category in the US is structurally different. Signet Jewelers operates over 2,800 US locations with decades of brand recognition built through mass television advertising. In Canada, Michael Hill competes against a smaller set of incumbents with less entrenched brand positions. The brand awareness gap to close in the US is categorically larger than the one closed in Canada, and requires a capital investment in advertising and store density that nine locations cannot generate.
Could Michael Hill have succeeded in the US with more stores?
Possibly — but not at nine. The economics of brand-trust categories in the US require scale before they produce returns. A concentrated strategy of 40–50 stores in a single US metro area — with a supporting advertising budget built around that geography — would have tested the model under conditions where it had a real chance. Nine stores across multiple locations produced neither the brand density nor the economic returns to prove the thesis. The question is whether the capital required for that concentrated strategy was available and justified given the competitive landscape.
What does “product-market fit transfer failure” actually mean for a retailer?
It means the conditions that made the product the right choice for customers in the home market do not exist in the target market in the same form. For Michael Hill, those conditions included: a competitive set where Michael Hill’s brand and price point were genuinely differentiated, a customer base for whom Michael Hill was a known and trusted option, and a store density that generated awareness. None of those conditions existed in the US, and replicating them would have required a different entry strategy and a substantially larger capital commitment than was deployed.
Go In Knowing
The US Entry Diagnostic delivers a structurally sound plan to enter the US before you invest $500,000 in painful mistakes.
Book the Paid Diagnostic$10,000 NZD. Credited in full toward any engagement.