---
title: "Understanding Hema: The Key to Retail's Next 20 Years"
description: "Hema is planning a funding round, while various new retail models have fought from first-tier to fifth-tier cities over the past two to three years, severely weakening traditional retail giants but also leaving themselves struggling. This outcome was foreseeable before the pandemic, which only created an illusion of prosperity, ultimately benefiting consumers but losing investors money. Today, amid the most intense transformation in China's retail industry, we should consider what retail will look like in 20 years. We believe Hema best represents the future. This analysis is based on interviews with over 20 experts and research on more than 30 Hema stores."
author: "蒋文伟"
publisher: "New Distribution"
email: "zhaobo258@gmail.com"
telephone: "+8615854817671"
published: "2022-03-18"
language: "en"
canonical: "https://xinjignxiao.com/en/articles/understanding-hema-the-key-to-retail-s-next-20-years-4d81adb5/"
markdown: "https://xinjignxiao.com/en/articles/understanding-hema-the-key-to-retail-s-next-20-years-4d81adb5.md"
original_source: "https://mp.weixin.qq.com/s/xKftnmHNX_lqUnHuYSr7Wg"
translation: "https://xinjignxiao.com/zh/articles/%E7%9C%8B%E6%87%82%E7%9B%92%E9%A9%AC-%E6%89%8D%E7%AE%97%E7%9C%8B%E6%87%82%E9%9B%B6%E5%94%AE%E6%9C%AA%E6%9D%A520%E5%B9%B4-4d81adb5.md"
attribution: "New Distribution — https://xinjignxiao.com/en/articles/understanding-hema-the-key-to-retail-s-next-20-years-4d81adb5/"
usage_policy: "https://xinjignxiao.com/ai-policy.txt"
---

# Understanding Hema: The Key to Retail's Next 20 Years

> Hema is planning a funding round, while various new retail models have fought from first-tier to fifth-tier cities over the past two to three years, severely weakening traditional retail giants but also leaving themselves struggling. This outcome was foreseeable before the pandemic, which only created an illusion of prosperity, ultimately benefiting consumers but losing investors money. Today, amid the most intense transformation in China's retail industry, we should consider what retail will look like in 20 years. We believe Hema best represents the future. This analysis is based on interviews with over 20 experts and research on more than 30 Hema stores.

Recently, Hema planned a funding round, while over the past two to three years, several new retail models have fought from first-tier to fifth-tier cities, severely damaging traditional retail giants but also leaving themselves on the brink. This outcome was foreseeable before the pandemic, which only created an illusion of prosperity, ultimately benefiting consumers but losing investors money.
****Today, amid the most intense transformation in China's retail industry, we should consider what retail will look like in 20 years. We believe Hema best represents the future.**
First, let's discuss the research basis: this analysis is primarily based on interviews with over 20 relevant experts, surveys of more than 30 Hema large stores and mini stores, and public information.
Additionally, over the past two years, I have benefited from insights from former colleagues and friends, including core members of top-3 community group buying companies (reporting to CEOs), executives from top-2 front-warehouse companies, former China executives of the world's largest retailer, and my mentor, former president of JD University.
I also thank my partner, a former Principal at Bain & Company, a leading consulting firm in consumer retail, who provided professional perspectives honed through consulting projects worth hundreds of millions.
****Rich first-hand research, combined with a top expert team, aims to deliver insights that are both profound and professional.**
**What core customer value should future hypermarkets provide?**
Before diving in, let's look at the aftermath of the same-city retail war over the past two years. The former "Four Little Dragons" of community group buying are now reduced to one, desperately defending its home base in Hunan, awaiting attacks from giants. Nine other dragons have burned through billions of dollars from investors and vanished. The two front-warehouse companies that have "landed" have less than a year of cash flow—who will save them? This battle has not only lost money but also made it hard for street vendors to survive, and traditional supermarkets that were once stably profitable are closing stores en masse but still bleeding.
Yonghui just released its earnings forecast, losing 3.9 billion yuan in 2021, though that's 4.5 billion less than the previous year. In terms of risk-taking, local supermarkets can't match Yonghui, so they barely maintain 0.5%-1% net profit, making them industry leaders (Figure 1). Even Walmart closed 35 stores in 2021, with only 400+ stores nationwide. So to be fair, Hema has over 300 stores and closed only a few—that's pretty good!
Figure 1: Performance of typical listed retail companies in China, 2021H1
Back to the topic. A common question in recent years is whether the hypermarket format will be eliminated.
Before concluding, let's review: 30 years ago, having goods was king; 20 years ago, when Walmart and Carrefour entered China, novel products and cheaper standard goods were king.
Today? Which standard product can beat Pinduoduo's clearance prices? What discount can be more affordable than HotMaxx? Some might say Costco, but is it really successful because of extreme low prices?
**Price should never be the sole advantage of a hypermarket; it wasn't in the past, and it shouldn't be in the future.**
****If there are no good products and no price advantage, traditional hypermarkets end up like today: becoming vegetable markets for retirees and housewives, since they are close to home, vegetables are cheap, and the environment is clean.
**This is a huge trap.** Fresh produce is traditionally a traffic driver for hypermarkets, with high loss rates, intended to drive sales of high-ticket, high-margin items like daily necessities and alcohol. But today, these categories have many alternative channels that offer genuine products at lower prices, leading to continuously declining turnover for non-fresh FMCG goods.
Faced with this situation of being "fleeced" after attracting traffic, traditional hypermarkets are helpless, waiting for losses and falling into a vicious cycle.
Those retail giants that can barely survive either rely on the first-mover advantage of early site selection (**offline is always about location; with good location, even 60-70 point products and operations can be profitable**) or delay payments to suppliers.
So, looking ahead 10 years, what is the value of hypermarkets?
Let's go back 20 years. I remember when Walmart first came to Shanghai in the early 2000s, my family would drive 40 minutes to Tangqiao Walmart every two weekends. It was packed, with nearly 30 checkout lines, each with at least a 15-minute wait, and families with children everywhere. Wasn't that the offline Taobao of 20 years ago?
Now, on weekend evenings, whether in first-tier or third-tier cities, is it possible to see such scenes in traditional hypermarkets?
Figure 2: Hypermarket foot traffic comparison: 20 years ago vs. today vs. Hema X Membership Store
We once believed that according to the four elements of retail—more, faster, better, cheaper—achieving excellence in any one could create a successful retail model. But today, that doesn't hold.
Pinduoduo lost over 25 billion yuan in three years; JD Daojia has never been profitable (and continued to lose after IPO); hypermarkets with tens of thousands of SKUs are losing money year after year; and even Hema, as a good example, is still at the breakeven line.
**20 years ago, a hypermarket carried almost all daily consumption; today, fragmented, no model can claim to have truly succeeded except Alibaba.**
Try without subsidies? Try with double delivery time? So what is the endgame for hypermarkets? If only price advantage remains, that's a discount store. How to bring people back to hypermarkets? Taobao is the best case: its slogan is not about the lowest price, but "So fun to browse!"
The future of hypermarkets should be similar: they should become the offline Taobao. The biggest challenge is bridging the gap from "operating products" to "operating users," as "goods finding people" will completely overturn traditional hypermarket operations.
Ask young people today: do you visit traditional hypermarkets for fun? If so, you might be depressed.
Unless for work, I wouldn't visit a traditional hypermarket once a year. I've asked peers with families, and most say the same. But we are all Hema members and Sam's Club members. Even my parents, with a combined retirement income of 8,000 yuan (less than second-tier retired civil servants), spend nearly 10,000 yuan annually at Sam's Club.**
****The old advantages (geographic locations enjoyed for decades) will eventually run out. Traditional retail leaders, under such intense consumption changes, if you don't change, you'll be the next Zhongbai, and Hema will replace you.
So our view is that hypermarkets will definitely exist, and in 20 years, they will gradually return to their state of 20 years ago. Hypermarkets can not only meet daily procurement needs but also become new places for people to kill time, even enhancing family bonds. All of Hema's efforts are moving closer to this direction.
**What can we learn from Hema for the next 20 years of retail?**
**This part could take a day. From its internal project in 2015 to today (Figure 3), Hema has never been about renovation or imitation. The innovations Hema has made in two years rival the cumulative innovations of many local retail giants since their inception.
Before expanding, let's look at some data on the demographics of fresh retail and Hema's formats (Figures 4-6).
Figure 4 compares all fresh e-commerce models by core age group, delivery time, price positioning, and online order volume. **Hema leads in capturing the youth segment (18-39 years old), while community group buying, with its low-price, average-quality fresh produce, targets the high-frequency grocery shoppers (30-60 years old).**
The high-frequency grocery shoppers are a group not to be ignored. Figure 5 shows Beijing's situation: middle-aged and elderly people account for nearly 80%, and this proportion increases in lower-tier cities. They are both the core decision-makers for daily groceries and daily necessities and the high-repeat customers of traditional hypermarkets.
But the downside is that this group's demand for daily necessities, alcohol, mother and baby products, snacks, and consumer electronics is increasingly captured by Pinduoduo, moving them further away from the optimal category contribution structure for hypermarkets.
Figure 3: Hema development milestones, 2015-2021
Figure 4: Comparative analysis of major fresh e-commerce models, 2021
Figure 5: Fresh retail market share by consumer group, Beijing, 2021
Conversely, Figure 6 shows that Hema's core user base is young people, with some middle-aged and elderly users.**
****Imagine 10 years from now: today's young Hema users will still be with Hema, and Taobao's younger user base (~800 million MAU) will continue to feed Hema with incremental users, while traditional hypermarkets are still consuming the residual spending power of these middle-aged and elderly groups.**
In terms of price, Hema is slightly higher but acceptable. **The high quality, delivery efficiency, and affordable prices accumulated over more than a decade are almost irreversible in shaping consumption habits.** To reinforce this "irreversible" consumer mindset, Hema has made numerous innovations over the past six years. Let's analyze them one by one.
Figure 6: Hema user profile, 2019
**First, from day one, Hema has operated a retail format using a consumer brand approach.**
**What are the basic elements of a brand? Sufficient user trust and differentiated competition.
Consumers who have used Hema, Dingdong, and Youxian generally agree on Hema's quality. We crawled many third-party platforms; although not rigorously statistical, this universality is confirmable. Second is the differentiated competition approach.
Six years ago, using king crab as a hero product for cold start directly elevated the store's档次. To this day, Hema's 2.0-2.5kg king crab remains the lowest price nationwide. Even on several trips to Sanya, I haven't found such low prices.
Here's a tip: Hema's king crab is best bought during promotions at 1299-1399 yuan; go to the store before 9 am to pick one, and you can almost always get one over 2.4kg. Other front-warehouse platforms charge 1599-1699 yuan but deliver crabs barely 2kg.
**This is the consumer brand approach: use an exceptional hero product to make consumers remember you forever.**
Hema has more than one such SKU. It also pioneered as a launchpad for internet-famous brands, like Beyond Meat's China debut at Hema, and co-branded products with trendy tea brands like Heytea are endless.
Hema's own-brand hits like Wuchang rice, Daily Fresh milk, Max chicken feet, and Niulandi account for an estimated ~15% of GMV, while the best in China, Yonghui's private brands (like Caishixian), only account for 3-4% of GMV.
Of course, globally, European retail giants' private brand revenue typically accounts for over 30%, showing the gap.**
****Second, Hema's store expansion over the years has not been random; the first third of the time was spent refining methodology and testing the UE model.** According to our statistics (Figure 7), Hema opened only about 20 stores in the first two years (2016-17), then rapidly opened over 300 in the following four years (2018-21), with an expected 50-100 more this year.
During this period, Hema paused expansion for a long time and restarted from the second half of last year. Lao Cai once admitted: "Previously, large store operations focused too much on localization, delegating to city companies. Now we've adjusted to a 'One Hema' strategy."
Honestly, this strategy was initially dismissed and ridiculed in the industry, but Lao Cai surely understands that "the mighty dragon cannot crush the local snake." **Precisely because of his deep understanding of traditional retail, he dared not rashly use entirely new product and operation strategies in the 2016-18 phase.**
Only after being beaten by various local players, and with the success of a series of new models and mature methodologies in recent years, did he reflect and have enough determination to adjust strategy.
Figure 7: Expansion of Hema's core formats, large stores & mini stores, 2016.01-2022.01
Speaking of losses, rapid expansion requires money, especially the investment in supply chain, software and hardware technology, and fresh produce origins from 2018-2020, which was Hema's most cash-burning phase. But based on public data from investment days and UE calculations (analyzed later), front-end losses have narrowed significantly, and the omnichannel order mix is adjusting to a reasonable ratio. This burning of cash is worthwhile.
As a core piece of Alibaba's next 20 years, **Hema's operations are actually improving. If not for the emergence of other models, Hema would likely be fully profitable now.**
Another point: to be fair to Hema, it only entered the hypermarket business in 2016. As mentioned, offline retail is all about location! Location! Location! The best spots were already taken by established retailers, and many regional retailers also have local support, both overt and covert.
Hema achieved this in six years. To use a slightly exaggerated analogy: someone who started working and bought a house in 2000 now has a house worth 10 million yuan. Can that be compared to someone who started working in 2015 and bought a 10 million yuan house from zero in five years?**
****Competing head-on with those who have enjoyed dividends while holding a poor hand offline—this difficulty is well understood by traditional retail people.**
But this is the battlefield; if you want to win, you have to accept it. In this regard, Lao Cai is a model. In many public speeches, he has been candid about what has been done well, what hasn't, and what needs improvement, unlike some competitors who made bold claims and then were proven wrong.
**Now let's talk about Hema's product and display strategy. Lao Cai left himself no room for retreat from the start.** Figures 8-9 show the display of mini stores and typical SKU structure of large stores, based on field measurements and third-party online data.
Such a high proportion of fresh SKUs and display area is rare among hypermarkets nationwide.**
****Fresh produce is a double-edged sword: high loss rates come with high repurchase and consumption frequency.** If operations succeed, it benefits brand momentum, store traffic, customer fission, origin construction, customer acquisition costs, and user stickiness.
If operations fail, especially with Hema's high quality requirements, product loss rates can be 4-6% higher than ordinary hypermarkets. Additionally, Hema's high-margin SKU proportion is relatively low, posing a major challenge to profitability.
Furthermore, to be honest, as a heavy user of Sam's Club, I find Hema X Membership Store's value for money mediocre. If 2 out of 10 products at Sam's are disappointing, Hema X has 6-7, and many SKUs are even worse than HotMaxx (though the models aren't comparable).
Overall, product selection seems rushed. **Piling up thousands of SKUs in 2-3 years of hype still lags behind Sam's decades of global supply chain and selection methodology.**
Figure 8: Layout and display proportion of a Hema mini store in Shanghai, 2021
Figure 9: Hema large store vs. traditional hypermarket, online SKU proportion, 2020H2
Figure 10 shows the fresh brand penetration rate of Hema mini, which refers to the proportion of PB and branded products in fresh SKUs, including packaged vegetables and prepared dishes. **These product operation strategies also increase loss rates but play a key role in building the "good products" consumer mindset.**
We believe that for local fresh brands, it might be wise to appropriately reduce this proportion, adopt stronger buyer strategies to absorb regional quality brands' short-shelf-life SKUs, and even test PB products, rather than adopting homogeneous product strategies.
Retail innovation is often in details, with few secrets. Anyone can visit hypermarkets, but how many competitors are willing to invest effort in reflection and learning? **The most terrifying competition is when all your cards are on the table (playing open cards). At first, they disdain to learn, then they are unwilling to learn, then they claim it's impossible to learn, and finally, they truly cannot learn.**
Figure 10: Fresh brand penetration rate of a Hema mini store in Shanghai, 2021
**Next, let's discuss Hema's online operations.** We tried the apps/mini-programs of the top 30 domestic supermarkets. Fewer than 10 have apps. Except for Duodian and Sam's Club apps, which have reached the top level of mobile internet operations, **the self-operated apps of other traditional retail giants are about 10 years behind Hema's app.**
Yes, that's the level of apps from around 2011. The barely usable Yonghui app is about the ceiling. You can download and try it for details. I won't post product test results here; I'll just say this much.**
****Leveraging Alibaba's e-commerce foundation, Hema currently has no rivals in membership systems, campaign operations, and private domain operations in fresh e-commerce.**
From online operation data, Hema's app MAU is around 23 million, compared to over 60 million registered users. This stickiness is comparable to Xiaohongshu and Bilibili.
DAU/MAU is also ~30%, higher on weekends, summer, and during promotions. Whether in stickiness or online operations, it can even beat most apps.
Some might ask: why build your own app? Isn't it fine to use Meituan or JD Daojia?
First, if you don't want to serve young people well, that's fine. Second, JD Daojia (Dada) lost nearly 10 billion yuan from 2017-2021; how many more years can it subsidize you? Third, look at the restaurant industry today: being strangled by the neck is not pleasant.
But **building a fresh e-commerce app with an 80-point experience today requires a 300-person R&D team and a year of work, and it still might fail during promotions. This is a huge barrier in itself.**
Hema has another key barrier: the Hema Operating System (originally called RexOS). We often talk about retail digitalization. **Globally, the true leaders in retail digital systems are Hema OS and Bianlifeng OS, and possibly no others.** CC is also an entrepreneur we admire.
Many people ask me about dozens of retail SaaS or digital middle-office companies in China. The real experts are at Hema, Duodian, and Bianlifeng. Bianlifeng is China's best convenience store chain and will definitely overtake 7-Eleven and FamilyMart in the future, but it needs time, as the latter two have deep roots in Shanghai.
Back to the topic: how strong is Hema's system?
Hema large stores are classified into A-D categories based on daily order volume. A-class stores have at least 8,000 orders per day. In extreme cases, like Yanggao South Road store during summer promotions, daily orders exceed 15,000, with 80% being online orders. That means about 10,000 online orders need delivery from one large store per day.
Due to order peaks and troughs, **peak hourly orders are ~2,500, meaning 40 online orders are dispatched to the store per minute.** They need to complete picking (mostly in the back) within 10-15 minutes, and theoretically deliver within 30 minutes. But in such surge conditions, **delivery averages 1.5 hours without errors.**
Behind this are real-time inventory calculations, replenishment system upgrades, route design for all residential areas in the city, and dispatch between stores and third-party logistics, all requiring extensive modeling and complex algorithms, no less than Didi's algorithm capabilities.
Additionally, a minute-level SOP is needed (Figure 11). I don't know if RexOS still belongs to Taoxianda, but this capability, built on Alibaba's technology platform, if productized, could be worth at least $2-3 billion. So is Hema's rumored $10 billion valuation expensive?
Figure 11: Hema store fulfillment process SOP and efficiency analysis, 2020H2
Of course, **an excessively high online order ratio is not good. If the average order value is below 90+ yuan, it's almost impossible to cover fulfillment costs, and the order basically doesn't make money.** Moreover, rider costs are rising irreversibly.
The most suitable for online are actually Metro, Sam's Club, and X Membership Store, where offline average order values are never below 300 yuan, and online is at least 150 yuan, fully covering fulfillment costs. Going online won't affect UE profit levels.
This is also the challenge for front-warehouse and delivery services today: even a bottle of soy sauce must be delivered within 30 minutes. Even if users pay a small delivery fee, can platforms charge supermarkets a 25% commission?
If soy sauce costs only 4 yuan, what's the point of a 25% commission? So Hema's latest strategy is coordinated online and offline development, with ensuring offline orders account for over 40% as a key line.
Finally, let's discuss **topics investors care about: profitability and going down-market. When can Hema become profitable?**
We reviewed all brokerage reports' UE analyses, and some are ridiculous—some models are wrong! Hema has 330 stores with vastly different operating conditions. Large stores and mini stores have different SKU structures, and rents vary by city. Even within Hema, probably fewer than 20 people truly understand it.
But that doesn't mean it can't be analyzed; it's about how. The most critical four metrics in the entire UE table are: daily order volume and composition, pre-loss product gross margin, fresh produce loss rate, and marketing expenses.
Trunk transportation and warehouse costs, average rider cost per order, local rents, in-store shrinkage/theft, depreciation, and utilities have similar industry averages.
We made over 30 UE assumptions. Given sensitivity, I'll mention a few points. **When pre-loss product gross margin is 35%, with daily orders <4,000, it's basically hard to profit regardless of online ratio; with daily orders >8,000 and online ratio >80%, it's a slight loss or breakeven at best,** so the online ratio must be reduced.
Fresh loss is calculated at 11-13%, as Hema's high fresh ratio and quality requirements mean losses are higher than traditional hypermarkets. Marketing expenses are normally 5-6%, and 8-10% during promotions. If interested, we can discuss further.
**Overall, we firmly believe Hema will achieve overall front-end profitability soon.**
Normally, if pre-loss product gross margin is below 28%, it's hard to profit unless the location is excellent.
Strategies like not charging entry fees typically result in gross margins above 30%. Hema's PB products account for 15-20% of GMV, and their contribution is quickly exceeding 20%. High-quality PB products have gross margins of at least 50%, often 60-70%.**
****In the short term, Hema needs to overcome fresh produce losses, but the recent opening of outlet stores may offset some losses.**
Stepping back, Alibaba is always Hema's backer. As long as the direction is right, Hema won't worry about money (of course, Lao Cai faces internal pressure, and when the time is right, it can spin off).
So to this day, Hema continues to try new models, from Hema Wine Cellar, Hema Garden, Hema Butcher Shop, etc. Truly, for the best value and most stable quality flowers in Shanghai, Hema is either first or second.**
**Regarding going down-market, in the short term (3-5 years), it's challenging for large stores to penetrate third-tier coastal city centers.**
Let's look at population: coastal third-tier main urban areas average ~3 million residents, central ~2 million, coastal fourth-tier ~800,000, and central/western ~500,000.
Convert to households by dividing by 2.5, and consider those registered but not living there, then discount by 20%. The active consumer base for large stores is at least light middle-class (no authoritative definition; we consider non-first-tier households with annual after-tax income above 150,000 yuan). In third-tier cities, this might be 8%, so 3 million / 2.5 * 8% * 80% ≈ 80,000 households.
Based on Hema's user stickiness, opening one large store requires at least 20,000 active household users, meaning a 25% penetration rate among households with purchasing power.
Another key challenge not yet considered: **a hypermarket's base is typically determined by high-frequency grocery shoppers, and in third-tier cities, this burden often falls on retired elderly, which is difficult.** If they don't buy groceries, shopping once a week is not easy.
Alternative calculation: assume daily orders of 8,000, i.e., 240,000 monthly orders.
If each household consumes once a month (and they can afford it; from purchasing power, third-tier urban households have monthly after-tax income of 6,000-8,000 yuan, annual after-tax ~100,000), a coastal third-tier main urban area has ~1 million households, so a 25% penetration rate is needed, which is not easy even with other positive factors.
Not to mention X Membership Store, where a few items cost 700-800 yuan. In our view, second-tier cities are far from saturated, with average main urban populations above 5 million and significantly higher household after-tax monthly income.
**Let's share one-third first. Feel free to add the customer service at the bottom for private discussion. Last year, we also surveyed hundreds of Hema members, and my partner has deeply studied various membership store models.**
****Back to the beginning: why talk about same-city retail and Hema? We believe that in another 20 years, China will definitely produce a "Walmart." What kind of company is Walmart?**
Over the past 20 years, it has rarely fallen out of the global top 3 by revenue. In fiscal 2021, it had profits of ~$13.5 billion and a market cap of ~$380 billion.
In contrast, China's largest supermarket chain has revenue close to 100 billion yuan, and the top 100 combined have revenue of 1 trillion yuan, but in supply chain capability and profit levels, they are far behind Walmart.
In 20 years, China's "Walmart" will have a market cap of no less than $100 billion. Looking around, the most likely candidate is Hema.
Source: Ruizhe Insights (ID: gh_f10a7c242141) Author: Jiang Wenwei
_-END-_


---

## Copyright and AI use

This article is sourced from New Distribution. Search, quotation, summarization, and model training are permitted, but every use must credit New Distribution and retain the canonical source URL.

Contact: zhaobo258@gmail.com · +86 158 5481 7671