Zhuoliwei: Put Time and Energy Where It Matters Most | Oasis Signals

绿洲资本·April 1, 2020·0·1

*The Oasis Signal* series

The past three weeks haven't been easy: one week of concentrated circuit breakers, one week of sharp rebounds, one week of tug-of-war between bulls and bears. I'm recommending an older piece by Zhuo Liwei of Gao Yi Asset Management, hoping it helps you filter out the "noise" and capture the "signal."

Enjoy!

1 Reflections on Research and Investing

Let me start by sharing some lessons from my years of doing research and investment.

I. Time is the only truly scarce asset for everyone; learn to study what matters most

The most important thing in research is time management. For each of us, time is our scarcest resource. Research confronts us with enormously complex, unstructured, vast amounts of information. The scarcity of time and the infinity of information give rise to two problems:

1. Study what matters most

I've always said we should research big questions, grasp the major directions of industries and companies, and aim to be directionally correct in our judgments and decisions. In short: big questions, big directions, high probability.

Whether you're a senior analyst or a fresh graduate, you must learn to stand at the highest vantage point and leading edge of an industry, to think like a business owner about big questions. Even if you achieve perfection on minor details, the value is limited. Fuzzy accuracy on big questions matters far more than precise perfection on small ones.

2. Accumulate steadily on the right path

The first step to solving the efficiency problem is establishing a sound, rational research methodology, then continuously optimizing this intellectual framework and method through learning. Accumulate steadily on the right path; be conventionally sound yet occasionally surprising. Don't sprint down narrow trails in the jungle — find the broad highway that cuts through it.

Even a turtle crawling on the right road will reach the finish line, while someone running directionless through the jungle may stay lost forever. The power of steady accumulation on the correct path is immense.

2 Four Dimensions for Thinking About What Matters

How do we think about what matters most? My personal framework: when researching a topic, examine it through four logical lenses — macro logic, industry logic, business logic, and financial logic.

Macro logic includes macroeconomic factors, but also social trends, collective psychology, even political-economic relationships at the grandest scale. For example, under globalization and the internet, the logic of traditional commerce has shifted dramatically; boundaries between firms and organizations have blurred, and competition must be understood through the lens of whole-industry ecosystems.

Another example: in this era, the influence of business — and of great entrepreneurs — far exceeds that of previous times. Influence is power, and this changes the logic of social, economic, and industrial development in fundamentally different ways.

Industry logic: Different industries at different developmental stages have different core drivers and competitive factors, and in the current era of the internet and globalization, the principles of change differ even more from the past.

For instance, because creative destruction is more prevalent, the relationship between manufacturers and users has changed fundamentally; technological advantages and business model lifecycles have become shorter.

Business logic: When examining a company's operations, first look at whether its strategic layout and business direction align with the macro and industry logic described above, whether they follow the great social and industrial currents. No matter how stellar a company's team, going against these two major directions makes operations enormously difficult and success highly improbable.

Financial logic: Just as mathematics is the best expression of other natural sciences, finance is the digitized, structured summary and record of a company's past operating behavior. Through these data we can effectively analyze and verify its business characteristics and problems.

These four dimensions form a mutually verifying, mutually cross-referencing logical闭环 for examining a company and an industry. Studying a micro-level enterprise still requires strong top-down thinking. Companies with serious flaws in their big-picture logic face both greater difficulty in succeeding and greater difficulty in being researched — the return on research investment is poor.

3 Grasping Industrial Essence, Core Elements, and Critical Inflection Points

Different industries have markedly distinct characteristics in their economic and business model essence.

For example, the capacity utilization and downstream demand of the hydropower industry remain relatively stable over the long term; cost structure, output, and pricing are fairly transparent. Its essence resembles a leveraged (high-debt) interest rate product. Traditional retail approximates commercial real estate leasing. Animation resembles an IT industry where content creativity and computer software mutually reinforce each other.

When examining an industry, one must also identify what drives industrial development and corporate growth. For instance, advances in chip technology drove computing power, which enabled rapid development across hardware, software, and applications in the TMT sector; technology path selection and whole-industry-chain ecosystem competition became the most critical strategies for relevant companies.

For consumer goods, the core driver is product strength. The stronger a company's product strength, the higher its bargaining position relative to channels; the accumulation of product strength over time and its spread across space form the brand in consumers' minds.

When studying an industry's long-term direction, grasping critical changes at inflection moments is also crucial. For example, at this point in the TMT industry, smartphone penetration and traffic dividends have likely ended; mobile internet user time is approaching a ceiling; incremental innovation is becoming harder, with competition increasingly zero-sum substitution among existing players.

These relatively minor innovations are also more likely to be acquired early by large companies. Against this backdrop, the logic and methods for understanding relevant sub-industries and specific companies must differ fundamentally from those of a few years ago.

Similarly for consumer goods: with the gradual disappearance of demographic dividends, and the completion of product penetration and channel distribution, the era of blockbuster single-product growth has essentially ended.

Current opportunities likely lie more in structural upgrading of existing markets and small-to-medium scale category innovation — not good news for companies with enormous bases, since the marginal contribution of product innovation will be limited and earnings growth difficult. But for smaller companies with genuine innovation capabilities and deep consumer understanding, this may instead represent a favorable opportunity.

4 Core Corporate Capabilities Through the Lens of Time, Space, and People

Time, space, and people are three dimensions through which to view industries and enterprises, and important angles for illuminating many fundamental principles.

Good business models should align with the direction of social development, with the benign development of human nature, and with contributing more total welfare to society. Such business is sustainable — a friend of time.

There's also the concept of space. A good enterprise, besides being a friend of time, must not be an enemy of scale.

Most commercial activities, as they expand in scale, generally experience diminishing efficiency. There may be an optimal or near-optimal solution between space (scale, geography, management span, multi-division structure, etc.) and profit. Expansion beyond this optimal point can lead to diseconomies of scale.

For example, restaurant companies dependent on chefs are enemies of scale. Virtually all restaurant enterprises that have achieved effective chain expansion are chef-independent. Process standardization is the prerequisite for scaling services. Good business models should possess systematic capabilities for continuous replication and optimization across time and space.

The most important dimension is people, and at its core is entrepreneurial spirit.

A company's core competitiveness is fundamentally its governance structure. Good governance structure is the intellectual product of the entrepreneur and a small team, depending to a considerable degree on the entrepreneur's vision, commitment, strategy, and the team's execution capability — on the entrepreneur's continuous learning, sharing, and innovation.

Founders, entrepreneurs, and governance structure constitute a company's ultimate core competitiveness. Great entrepreneurs and excellent governance structures unleash every individual's positive energy and creativity. Good organizations and good business models should fully激发 the good side of human nature. Only people are the ultimate source of value creation.

A company's core capability is its ability to build sustainable barriers relative to competitors — whether through powerful systematic capabilities that create sustainable cost and scale advantages, strong R&D that enables continuous technological or product innovation leadership, or excellent products and services that build brand and user stickiness. These resultant core competencies all fundamentally derive from the entrepreneur's vision and excellent corporate governance.

5 Brief Analysis of Several Major Sectors Through the Four Logics

1. Consumer Goods

From macro and industry logic perspectives, several important factors merit attention in consumer goods:

The demographic dividend in consumption has largely disappeared. China's unique demographic structure (especially the legacy of the one-child policy) combined with rapid production and channel expansion over the past two decades means product penetration and channel distribution are largely complete. Going forward, there is little to no growth in customer numbers or per capita consumption — declines may even occur. The era of blockbuster single-product growth has essentially ended.
The proportion of middle-class and internet-generation populations will rise rapidly in coming years. Opportunities exist in structural upgrading of existing markets and category innovation, with experience, service, health, and aesthetics becoming important purchase decision factors. However, the marginal contribution of innovation will diminish, offering limited marginal benefit for large companies.
Manufacturer-consumer relationships. From manufacturer-led in the past to consumer-led now, consumers are ultimately the most influential guides. Yet good manufacturers must deeply understand consumers while also transcending them. The most important thing for manufacturers is establishing efficient, interactive communication models with consumers.

Current IT technology may help achieve this — using big data to record the full consumption behavior process, then mining this data deeply to optimize processes and supply chain efficiency. On this basis, continuously启发 effective innovation to bring more products that match supply with demand.

In the past, manufacturers produced too many mediocre products that consumers purchased mediocrely. Going forward, there should be more precisely targeted offerings.

Brand and product strength. Internet penetration has largely eliminated information asymmetry; consumer cognitive ability has risen substantially, improving consumers' bargaining position relative to manufacturers. Brand loyalty faces greater challenges, yet premium brands gain market share more efficiently.

More certain is that channel value has been massively compressed and weak brands are rapidly eliminated. In this sense, product strength (broadly defined, including service) will more easily demonstrate its value. Sustained product innovation capability, effective and precise communication (replacing traditional image endorsement and advertising), and positive consumer interaction become increasingly important.

Under such macro and industry logic, the micro-level business and financial data to observe differ from the past. For instance, we should examine whether gross margins on mature categories are improving or stable, whether new category share is rising, whether marginal profits from new category innovation are improving, and whether overall market share is increasing.

Similarly, consumer observation requires examining customer count, average transaction value, repurchase rates, and other business and financial dimensions to verify whether a company possesses sustained learning and innovation capabilities.

2. Services

From a macro logic perspective, against the backdrop of largely completed goods consumption penetration, the share of experience consumption and service consumption will continue rising — consistent with the characteristic consumption tendencies when per capita GDP reaches certain levels in macroeconomic terms. Meanwhile, the integration of goods and services is increasingly common; for consumers, "service is product, product is advertising."

Against the backdrop of current macroeconomic conditions and the rapid rise of the middle class, services still have enormous room for growth.

From an industry logic perspective, Maslow's hierarchy of needs tells us that once physiological and safety needs are met, the demand for belonging — social connection, respect, love — and self-actualization increases dramatically.

Therefore, what matters most in services is standing in the user's shoes to build better experiences, word-of-mouth, and the resulting user stickiness. It's about maximizing total utility across functional needs, experiential satisfaction, emotional resonance, and cultural identity.

Following these two logics, there are several important dimensions for evaluating services businesses:

User base and its changes — this is the foundation, especially paying attention to users who can actually transact or pay;
User renewal rate (and corresponding churn rate) and ARPU (or ASP) — this is the best data validation of word-of-mouth and experience;
Marginal cost and marginal profit of customer acquisition: a company's ability to acquire new customers through brand, communication, and advertising is critical, and whether new customer acquisition generates marginal profit.

For example, customer referral rate is an excellent metric (NPS, Net Promoter Score; NPS = (% Promoters) - (% Detractors). This thinking applies to consumer goods too). Good products or services make existing customers the best evangelists.

Whether customer learning costs are low enough and switching costs are high enough — meaning customers find it easy to enter but hard to leave.

Of course, "services" is an extremely broad concept, encompassing both traditional services and various information services built on internet infrastructure. But the core logic behind them is similar, and the framework for understanding and researching them can be fully connected.

However, traditional services and emerging internet-enabled services differ significantly in the development paths of their operational and financial metrics — users, traffic or sales, cash flow, net profit. Traditional services tend toward linear growth across time and space, while emerging internet-enabled services more often exhibit exponential growth characteristics, leading to winner-take-all industry structures where a tiny number of companies capture most market share.

3. Manufacturing

From a macro logic perspective, manufacturing (including hardware broadly defined) has several key points:

China's overall manufacturing capabilities remain difficult for other economies to replicate. After more than 20 years of accumulation, China's comprehensive industrial配套 capabilities still maintain strong global competitiveness — and this competitiveness may actually strengthen further;
Although China's overall demographic dividend has disappeared, the human resource dividend of engineers and high-quality workers may persist for a considerable time;
A more distinctive factor is China's enormous domestic market serving as an excellent testing ground. For many companies, trial-and-error costs are far lower than in other economies;
Given generally sluggish economies in developed countries and weaker financial conditions among some manufacturing firms, this creates rare globalization opportunities for China's leading enterprises.

These factors provide relatively favorable conditions for China's manufacturing upgrade, but likely only a very small number of excellent companies will seize these opportunities.

From an industry logic perspective, the core logic of traditional manufacturing is product and process standardization and scaling, thereby achieving self-replicating capabilities across time and space. The lean production represented by Japanese companies in the late 20th century fully embodied this logic, which will remain applicable in China for some time — China's white goods industry amply demonstrates this.

On the other hand, with the rapid development of computing technology, big data, and cloud computing in the present and future, non-standard products that were previously difficult to scale can now, to a considerable degree, undergo full-process digital transformation and potentially achieve personalized mass production.

This is already beginning to appear in certain niche industries (such as custom furniture and apparel, sectors where increasing industry concentration was previously extremely difficult), potentially breaking through the "scale is the enemy" constraint for non-standard products.

Of course, this exploration process will be lengthy, with high upfront sunk costs in R&D, and the difficulty far exceeds the scaling of standardized products through capacity replication. But successful execution builds much higher barriers. This area deserves long-term strategic attention, while requiring rigor and patience in investment timing.

From a business and financial logic perspective, whether through capacity replication or personalized mass production, two factors are critical for manufacturing:

First, whether expansion of generalized capacity scale (financially including fixed assets, intangible assets, goodwill, and operationally including manufacturing, R&D, marketing, and frontline personnel) generates positive marginal profit. Good expansion should bring rising marginal profit rates, thereby improving overall ROE.

Second, whether this scale expansion significantly enhances long-term competitiveness — such as mastery of core technologies with barriers, continuous brand accumulation from terminal scale, and the resulting product premium and share gains. In summary: achieving scale and cost advantages in space, while accumulating brand premiums over time.

4. TMT Industry

From a macro logic perspective, technological progress is one of the most important factors in long-term economic growth potential. Over the past decades, the TMT industry has concentrated the largest scale of technological advancement and business model innovation in human history, and has been the most important global growth driver.

The technological revolution, especially the internet revolution of the past two decades, has completely restructured the global economic landscape, industrial evolution, and human lifestyles.

But from another angle, perhaps the pace of technological progress has exceeded the growth rate of aggregate macro demand. Combined with the characteristic of creative destruction, this has shown effects of low economic growth and low employment over the past decade.

The technological revolution, amplified by modern financial technology and capital, has also made initial distribution more concentrated toward technology elites and capital (shareholders), without significantly raising ordinary workers' compensation. Wealth polarization continues to intensify globally, including in developed countries — perhaps a deep underlying cause of the financial crisis since 2008.

As discussed in Sapiens, ordinary people, whether as workers (increasingly replaced by machines and data, with machines' productivity improvement potential likely exceeding humans') or as consumers (with little compensation growth and reduced consumption elasticity), may see their economic value continue to depreciate.

From an industry logic perspective, the path of TMT technology evolution, the commercial普及 of technological achievements, and the resulting innovation paths for business models are particularly important. Due to the "law of increasing returns," technology and business model leaders will see their advantages and market shares continue expanding over considerable periods, with industry concentration rapidly increasing.

This holds true for hardware, software, and services alike — whether chips, displays, storage, various applications, traffic platforms, or social networks. Thus, within the same sub-sector, different companies' fates can be worlds apart, making mid-level industry research especially critical.

From today's vantage point, smartphone普及 and traffic dividends have largely ended. New technological progress needs to cross over to the next S-curve, and major technological innovation may require extended exploration. Big data, cloud computing, and next-generation smart terminal forms (AI, AR/VR — unlikely to produce standardized blockbuster products at the hundreds-of-millions-unit scale, more likely to be integration of hardware, software, content, and services) will follow development logics different from the past two decades.

From a business logic perspective, wrong technology path choices in TMT carry extremely high costs — a mistaken choice can be catastrophic. Some "excellent companies" featured in the classic business book In Search of Excellence are now either struggling or no longer exist. Research on the leading mainstream technology and its dominant full-industry-chain implications is especially important.

From a financial logic perspective, technology-oriented companies do not see revenue, profit, and cash flow change linearly as in traditional manufacturing.

For technology R&D companies, the first increases may be in technical staff and R&D expenses, with revenue following later, and net profit and cash flow lagging further behind. For business model innovation service companies, traffic and users may appear first, then revenue growth, and finally net profit and positive cash flow.

The broad industry categories mentioned above are not strict distinctions. These rough frameworks simply illustrate some important commonalities and patterns.

Many industries may combine multiple characteristics, though with clear primary-secondary relationships. Specific sub-sectors require deeper analysis with rigorous logical frameworks. Moreover, in today's economic and industrial context, interdisciplinary, integrated research is increasingly important.

Three Stages of Research: Induction, Deduction, and Empirical Verification

Researching or understanding a problem involves three processes: induction, deduction/reasoning, and empirical verification. These three mutually reinforce and validate each other.

Induction: We need to format fragmented information — first deconstruct, then reconstruct. According to the primary-secondary relationship of contradictions, we must identify what is most important, what ranks first and fifth, and set aside the secondary. In induction, we are not merely information transmitters but information integrators, clarifying logical relationships and primary contradictions.

Deduction: On the basis of induction, we propose the most probable hypotheses. Studying the history of science reveals that all of scientific history is a process of hypothesis-deduction. After inducting past information and theories, we propose several most likely hypotheses, then verify through empirical research.

Deduction requires logical, creatively nonlinear thinking that breaks from linear frameworks. The hypothesis process requires a certain rational imagination. Without imagination, none of the past internet companies could have been invested in.

Empirical Verification: Scientists studying natural phenomena need experiments to verify hypotheses and reasoning. 100 years ago, Albert Einstein proposed gravitational waves, which he himself doubted; then the world's top scientists spent 100 years ultimately proving their existence. Empirical research results may sometimes completely negate previous hypotheses while yielding new answers.

The empirical process can also inspire new thinking — an unexpected bonus of empirical research. You may find apples while looking for peaches. Sometimes researching problem A yields insights for problem B.

Of course, for investors playing the game of speculation, they may only do induction and reasoning, or be slightly earlier and stronger in logic and reasoning than others, and still make money from intellectual advantage. But this is hard to sustain, and the process will be anxiety-ridden.

For fundamental research, we must strive to do all three steps well. We aim to elevate understanding from typically 60% certainty to above 90%; in fact, a 60% certainty judgment has almost no decision-making value.

The principle may work like this: through more effective empirical research, perhaps you only master 10% more effective information than the market, but this may elevate your understanding and foundational information processing by a whole level; thereby raising research certainty significantly above most people's. This may be the process of creating alpha value.

Individual Traits of Excellent Researchers

To do good research and investing, I believe practitioners need traits significantly different from other industries. A certain company's campus recruiting mentioned three words: intellectual curiosity, honesty, and independence. I consider this an exceptionally brilliant summary.

First, intellectual curiosity/curiosity. This is the deepest self-driving force within a person, the underlying code of one's inner self.

I call this "self-driving force without pressure or assigned tasks" — when there's no exam pressure, no performance pressure, no one assigning you tasks, and you're still obsessed with researching a meaningful problem, then by nature you're very well-suited for research work.

Second, honesty. Most people, including myself, fall far short in this regard. Honesty seems like a simple requirement, but it's actually extraordinarily difficult to achieve.

People are always more inclined to accept — and even reinforce — things that favor them, arguments that support their own views, and ideas that earn them validation. When someone challenges you, most people will avoid or resist the confrontation, which is detrimental to investing or research.

We need exceptionally strong self-reflection and error-correction capabilities. Most importantly, we must know where we are wrong, make each mistake less costly, and continuously reduce the frequency of errors. Investing shouldn't involve pessimistic or optimistic views — only objective ones. We must be objective about our research subjects, and even more objective and honest about our own self-assessment.

We need crystal-clear awareness of our own boundaries of competence. The important issues you've been too afraid to confront, the ones you've been evading — they will come knocking one day. "If you are not completely honest, you will eventually become a victim of yourself."

Third, independence. If everything we say consists of second-hand information and second-hand opinions from others, it holds little value. Independent thinking and judgment matter most because they enable clear attribution analysis and gradually build a self-sustaining, continuously optimizable mental framework of one's own.

Moreover, no matter how authoritative an opinion may be, it must undergo rigorous independent judgment before adoption. Direct citation without this process makes innovative thinking impossible.

The Right Mindset: Necessary Training Time and Letting Go of Experience

In Outliers, the author writes that the reason we perceive certain people as geniuses is not superior talent, but sustained, unremitting effort.

Ten thousand hours of deliberate practice is the necessary condition for transforming from ordinary to professional. For younger researchers or recent graduates, regardless of what work you do, you need roughly ten thousand hours of professional training to achieve a qualitative leap from quantitative accumulation — about five years to become a professional-caliber practitioner.

If you have strong innate aptitude and your effective daily work hours exceed others' by 20%, your growth timeline will compress. For those who were academic high-achievers, newcomers to the industry often feel somewhat lost and impatient — they need to establish the right attitude. NBA superstar Kobe Bryant once asked: "Have you ever seen the 4 a.m. darkness of Los Angeles?"

For senior professionals, we instead need to emphasize letting go of all prior experience. In this era where internet technology and technological progress are fundamentally reshaping society, many traditional modes of thinking have become obstacles to researching numerous questions. We need to shed past preconceptions and face the world with a cleared-slate mentality.

The longer you've been in the industry, the more susceptible you become to certain types of errors. From this perspective, for views you hold with great conviction, you should deliberately seek out and value counterarguments. Be grateful to those who repeatedly challenge you.

Additionally, for a researcher to perform well, they must accumulate independent-judgment success cases and failure cases that belong to them — those unforgettable successes and unforgettable failures are what truly catalyze growth.

Recommendations for Daily Work

1. Build Your Own Circle of Competence

Whether senior or junior, everyone should gradually learn to build their own circle of competence. For any important research topic, cultivate at least three true experts as friends. This means standing on the shoulders of those three exceptional people when researching that topic.

Why three? Because even a brilliant person — even an entrepreneur themselves — may produce flawed judgments due to biases stemming from self-interest (financial stakes, emotional attachments, etc.).

Munger said, "Incentives cause bias." Assume five important topics in your mind, three friends per topic — that's fifteen people; each of whom has their own network behind them. Such a circle of competence becomes extraordinarily powerful. This resembles internet thinking; it's a form of cognitive surplus.

Our individual brains are boundless, and so is our circle of competence. None of us can become expert in many fields, but we can continuously learn from many experts and exceptional people, engage in deep intellectual exchange, and mutually illuminate each other's thinking.

2. Broad Learning and Reading

Our research work is fundamentally about learning. Learn from peers, colleagues, and books. Beyond immediate, present tasks, engage in broad reading. Deep reading and forward-thinking on important long-term topics matter enormously. Broad reading, independent thinking, deep discussion and interaction.

3. Build Your Own Research System

Establish your own logical framework and methodological system for industry and company research, organized in your own way. For younger people especially, foundational work must be done solidly — only while doing basic data work will you find yourself immersed in genuine thinking. Continuously optimize and refine this system through persistent learning and accumulation.

4. Structured Research Documentation

How to improve daily research efficiency? I recommend maintaining structured records of your working papers. For a company under intensive research, for example, consolidate key information in an Excel spreadsheet, expressing various elements through logical, quantitative frameworks.

Past research, periodic reflections — store everything in these files, continuously updating them. When you next need to discuss, draft, or output documents, everything becomes readily accessible. Over several years, these accumulated research outputs become tremendously valuable, and facilitate retrospective review of your past research.

Q&A Session

Q: On time management — researchers typically cover numerous industries and companies. How to use time efficiently? How to quickly decide where to spend time? Some short-term opportunities, not necessarily major multi-baggers, involve shifts in operational rhythm. How should researchers allocate their time? How to预判 whether an industry or company merits attention? For many short-term market hotspots, how can researchers rapidly judge whether deep investigation is warranted?

A: In terms of research time and energy allocation, I personally believe 70-80% of time should be spent on important questions, with less time devoted to companies experiencing short-term changes. Devote most time to the most important directional trends and most critical changes in an industry, and under that logic, the most important handful of companies.

Throughout one's long career, you must cultivate the mindset of doing important things to develop this capability — the earlier you understand this, the better.

If you're constantly occupied with short-term matters, your ability to think about important long-term questions cannot accumulate. People typically neglect things that aren't urgently pressing in the short term but matter enormously in the long term, while busying themselves with short-term urgencies of limited long-term value — this is putting the cart before the horse.

The starting point for researching any topic should follow this principle, though of course specific cases require specific analysis. Actually, many industries and companies can be filtered out through elimination.

For example: industries with clearly downward long-term trends, where the bottom is difficult to call, and overall valuations aren't low — basically don't bother; companies that haven't achieved much in their familiar domain, with unchanged management teams, claiming they'll accomplish something grand in an entirely new industry — this is basically bluster; companies claiming extraordinary technology and high-tech prowess, yet with very low product gross margins and low revenue per employee — basically fraudulent.

These simple common-sense judgments can eliminate many companies. Whatever their price performance, these types aren't our research targets. As I mentioned earlier, the purpose of research is to raise the certainty of high-value, important matters to very high levels — only then does it have genuine decision-reference value.

Q: For instance, in a given industry, year-end review reveals many fundamentally-driven multi-baggers we endorse. How do we expect researchers to capture these companies?

A: Take an example: China manufacturing-upgrade companies dispersed across various industries — say twenty companies requiring research. These twenty companies will certainly share certain significant commonalities, with macro-, meso-industry-, and micro-enterprise-level characteristics that are mutually resonant.

Based on these logical hypotheses, if we deeply research five of these companies, ultimately take concentrated positions in three, and achieve excellent returns, I consider this highly successful investing — or rather, if the returns on the three purchased companies exceed the aggregate returns of all twenty, that's successful investing.

The fifteen companies not invested in are logically irrelevant. Asking a researcher to conduct very high-quality research on twenty companies within one year is actually extremely difficult. Getting five companies to 90 points is far easier than getting twenty companies to 70 points — and 70-point research has no alpha value.

Q: To research five companies well, don't you need to research all twenty well?

A: This isn't logically contradictory. To research five companies well, you certainly examine the other fifteen — you compare across all twenty, then select five for truly deep research according to importance principles. Research on these five must be demonstrably superior to industry peers.

Moreover, selecting these five from twenty through certain methods isn't difficult. The four research logics discussed earlier, judgment of industry core elements and critical changes, understanding of entrepreneurial spirit and corporate governance — these are all methods for selecting a small number of key companies in a given domain, and can at least largely achieve elimination-based results.

Q: Which opportunities must be captured? Which opportunities are unacceptable to miss?

A: As mentioned, things representing important directions of social and economic development and critical industry changes must be captured. But I still believe that among ten companies representing these industries, if a researcher can seize three and translate them into actual investment contribution, that's excellent.

Demanding capture of all ten would likely reduce practical investment certainty — you might become患得患失, afraid to take concentrated positions or hold for extended periods. Instead, dramatically raise research certainty on those three.

If among ten companies representing important directions and critical changes, you capture zero, there must be fundamental problems with your research methodology.

Looking back, fundamentally-driven multi-baggers are actually abundant in the market. Most investors haven't failed to research many companies — they've failed to research them deeply enough, leading to患得患失, missed opportunities, or premature exits during the investment process, ultimately yielding suboptimal returns.

Additionally, once portfolio holdings exceed a certain number, risk diversification improves significantly and performance volatility actually becomes modest — this logic is quite simple.

For researchers, regardless of industry, you must personally learn to build your own mental framework and research system. Through this system, logically identify opportunities and manage time.

Q: You've researched a good company, but valuation seems expensive — what then?

A: This is primarily a portfolio manager consideration, not mainly a researcher concern. This involves overall valuation assessment, including cross-asset valuation comparisons, inter-industry valuation comparisons, and especially valuation comparisons among companies with different industry life cycles, growth trajectories, and earnings growth logics.

Whether different industries' or companies' valuations are expensive — the ultimate thinking follows long-term discounted free cash flow logic, judging future value rather than simplistic current-period PE, PB, or similar metrics. Deep fundamental research is likely the most reliable method for judging future value.

If all stocks in a portfolio are significantly more expensive than other assets (such as fixed-income assets), that's naturally a time to sell equities overall. If Company A is significantly more expensive than Company B, that's a case for swapping A into B.

Of course, good companies rarely trade at low valuations for long, but the A-share market's considerable volatility frequently creates investment opportunities born of market mispricing — and this year has offered plenty of such chances. With forward-looking, in-depth research, one naturally won't miss these opportunities.

For researchers, the key is to sustain study and tracking of the most important companies mentioned earlier. The market will always present us with chances to buy.

Q: Of your working hours, how much energy goes into tracking what matters, and how much into finding new companies?

A: As a fund manager, I personally devote more time to ongoing research on major industry directions, key inflection points, and priority companies — through exchanges with industry contacts, discussions with researchers, and my own broad reading and reflection on relevant issues.

As for new targets, most ideas also emerge from the above research and reading process. In truth, industry and company changes aren't that dramatic in short periods, so the pace of genuinely important new company research isn't that fast either. Of course, I also spend some time on reading that seems to have no direct connection to my current work — I enjoy all kinds of books.

Q: Many opportunities fall outside current coverage. How do you capture and allocate time for them?

A: This is similar to the previous question. Read broadly, and gain inspiration through exchanges with your own circle of wisdom. In reality, as long as you maintain continuous research on important questions, sustained communication with industry, and forward-looking reading, new opportunities and inspiration will certainly emerge. Many ideas are generated in the process, not through working in isolation — "those who act often succeed, those who journey often arrive" captures this meaning.

Q: How do you structure a typical workday? What do you usually read? For breadth reading, what do you read, and what would you recommend?

A: For fund managers and researchers, this certainly differs.

For myself, my time allocation looks roughly like this: the bulk still goes to research, focused mainly on verifying important mid-level industry logic and tracking relevant priority companies — including field research, discussions and learning with researchers, industry experts, or investment circle friends, reading company disclosures, and personally doing regular formatting and accumulation of important data. Particularly valued among these are deep discussions and reflection on key questions.

Reading also takes up considerable time. I may concentrate for a period on materials related to topics I consider important for the future, including statements from the most influential entrepreneurs; I also spend significant time on things with little relation to near-term investing — such as books or articles on philosophy, psychology, history of science, and pieces by excellent writers across various fields on new media platforms, which actually offer considerable inspiration.

Additionally, I particularly value interaction and learning with industry friends, including some remarkably interesting founders — conversations with them always yield tremendous gains.

For researchers, I suggest devoting more time to building and continuously optimizing your own research framework, cultivating the ability to judge major industry trends and key inflection points, and achieving top-tier research capability on the most important handful of companies in your coverage, rather than spending time on fragmented minutiae.

For those new to the industry, start by going deep in one focus area, learning the method for doing so, then gradually extend that method to other areas.

Overall, whether fund manager, researcher, or young person just starting out, one must cultivate the habit of thinking about big questions. Research big questions, big directions, do things with high probability of success. Being right on small questions has limited impact.

What life lacks most is time — time is every person's only truly scarce asset. Put time and energy where it matters most, and learn to do the most important things to the extreme.

Oasis Capital is a new-generation venture capital firm in China, dedicated to discovering the most vital entrepreneurs of the next decade and growing alongside them to create long-term value. "Nurturing Vitality" is Oasis's vision and mission. This vitality is both the direction of structural transformation in the era and the resilience and evolutionary force of entrepreneurs. Oasis Capital focuses on early and growth-stage investments, with individual checks ranging from $3 million to $30 million, concentrating on technology-enabled services in education, healthcare, enterprise services, and related fields, supporting China's technology-driven new services upgrade.