“Investing may be an art, but it’s one painted with the brushstrokes of science” - Jenga Investment Partners Ltd
As of today, May 22, 2025, there are 29,029 globally listed companies with a market capitalisation above $50 million. Like many concentrated global equity funds, my mandate at Jenga Investment Partners is to select a portfolio of around 15 companies. That’s roughly one company for every 2,000 listed globally, meaning being selective is not just important, it is necessary. To manage this, one needs a combination of sharp quantitative filters, mental models, and qualitative judgment. For context, if you spend just one hour researching each company, it would take you nearly a whole year to get through only a third of the global listed universe.
Admittedly, until recently, my approach to narrowing the investment universe was structureless and randomised. While it led me to some good investments, it wasn’t sustainable, particularly for a fund manager with long-term aspirations. Recognising this gap, I spent time reflecting and developing a more robust and repeatable framework and this article is a documentation of that process and I hope gives a clearer picture of my investment process to global investing. Here's what I will cover in this article:
The benefits and drawbacks of an industry-led approach to investment filtering
Categorising the world into 60 industry buckets
The seven industry indicators: presence of outperformers, knowledge and understanding, degree of resilience, replacement cycle, profitability and growth, innovation capacity and barriers to disruption
The industry indicator formula and how it guides my research prioritisation
Investment Process Stages 1 - 4: With the aid of one of 60 industries - Industrials trading distributors and conglomerates, I will walk you through how I breakdown the investment process over a four-step investment process
An industry approach
Global investors typically categorise investments using three main lenses: industry, country or region, and thematically. Each has its merits. After exploring all three, I found that an industry-led approach provides the most clarity and utility for my investment process.
Country or region
It's common for investment teams to be structured by their geographical focus, with each analyst focusing on a specific region or country. This has benefits, particularly when the strategy tilts towards an emerging market focus. Countries and regions could have high barriers to becoming experts, and analysts could gain more unique and deeper insights when assessing opportunities from a country perspective.
I experienced this country-led benefit first hand during my recent visit to Greece. A top-down dive into Greek equities led me to discover Optima Bank—a company that would have likely been filtered out under a pure industry or thematic screen. Its investment case required one to understand the deeper insights on Greece’s economy.
However, the drawback of this approach lies in the sheer diversity of businesses within a single country. In Greece alone, the five companies I visited had little in common in terms of customers or value drivers. For a solo manager, the business model range makes time allocation more challenging and less efficient.
Thematic
Over the past decade, several themes and trends, such as artificial intelligence, cloud computing and data centres, and electrification, have driven markets to new highs. These opportunities are concentrated in a handful of companies, and investors who have solely focused on these big trends have spectacularly made a decent return on investment and time. A thematic lens brings additional focus on the key factors that matter in an investment case, and if time is a key limiting factor, a thematic approach presents the best result among the top three methods.
The challenge, of course, is that not every investment case is driven by a theme or trend per se, and as we learned from the Global Outperformers study of ten-baggers, even some of the best-performing companies are driven by purely company-specific factors. Another challenge is that scanning for companies from a thematic lens could be tedious. In many cases, a key theme could cut across several industries and countries, which leaves analysts with even more legwork to list the genuine opportunity set.
Why Industry?
Despite its own limitations, an industry-led approach gives me the consistency and focus I need. It allows for comparative analysis among similar businesses and makes it easier to identify patterns, threats, and opportunities. One challenge, though, is that companies today are increasingly complex and diversified. Even within a single industry, business models may vary substantially across regions. To address this, I use the Global Industry Classification Standard (GICS) as a foundation, which organises companies into 11 broad industries. I then segment these further into 60 clearly defined sub-industries.
Segmenting global industries
My first step is segmenting the industries into smaller but defined lists, and here, I have split all 11 industries into 60 industries, each with a clear definition and roughly 500 per industry. In the table below, I highlight all the industries that come from financials (6 industries) and consumer staples (5 industries), and you can download the Excel spreadsheet for a list of all 60 industries.
Evaluating industries: The seven indicators
After listing all 60 industries, the next step is establishing a process that supports selecting which industries one might prioritise. Here, I created seven indicators that are crucial for selecting long-term investment opportunities. These seven indicators include:
Presence of outperformers
Knowledge and understanding
Degree of resilience
Replacement cycle
Profitability and earnings growth
Innovation capacity
Barriers to disruption
Let’s briefly review what each indicator means.
1. Presence of outperformers
This is a purely backwards-looking track record-focused indicator, and I examine which industries have historically produced the most outperformers. Outperformance is defined by companies that have either returned 15% per year over the past 20, 30 or 40 years or 5% over the past 40 years (latter rewards survival). Among the top five industries include:
Consumer staples - Tobacco and beverages: 26 of 235 companies or 11%
Industrial - Aerospace and defence: 20 of 219 companies or 9.1%
Consumer staples - Food and staples retailing: 24 of 332 or 7.23%
Industrials - Containers and packaging: 14 of 218 companies or 6.4%
Information technology - Semiconductor equipment: 18 of 218 companies or 5.9%
Of course, past performance doesn't necessarily predict future performance, so I combine this indicator with other more forward-looking indicators.
2. Knowledge and understanding
This examines my circle of competence and rewards industries where one can better grasp the value drivers and economics. It's highly subjective and dependent on your circle of competence, so adjusting the figures and scoring here with your reflections on the industry circle of competence is crucial if you choose to utilise this industry breakdown.
3. Degree of resilience to shocks
This ranks industries relative to their ability to withstand economic shocks and downturns. There isn't any industry that isn't somewhat impacted by the economy, so it's important to recognise to what extent each is and quantify accordingly. In my view, the top five industries with the highest degree of resilience to economic shocks include:
Consumer staples - Tobacco and beverages (9/10)
Consumer staples - Food and staples retailing (8/10)
Consumer staples - Packaged foods and meat (8/10)
Healthcare - Providers and services (8/10)
Utilities - Water, gas and multi utilities (8/10)
4. Replacement cycle
Next, another vital industry quality indicator I've identified is the product replacement cycle, which examines how frequently customers purchase products or services. The more frequent and recurring-ness, the easier it is to forecast future growth and clarity we have to gauge how things might change in the future. The shorter the replacement cycle, the better and here, I have the following industries as my top five industries with replacement cycle:
Information technology - Software (9/10)
Industrials - Transportation ground and infrastructure (9/10)
Consumer staples - Tobacco and beverages (8/10)
Consumer staples - Food and staples retailing (8/10)
Utilities - water, gas and multi-utilities (8/10)
5. Profitability and earnings growth
Profitability and earnings growth is mainly a quantitative lens into the average profitability (operating profit margin and return on invested capital) and growth operating profit growth and free cash flow growth industries have historically shown and are likely to show over the long term. The tricky issue here is that we must also account for who makes these profits and earnings growth. If future profit growth flows to only startups and new entrants, then this situation doesn't benefit the listed companies we target. Here, my top five industries include:
Information technology - Software (8/10)
Communication services - Interactive media services (8/10)
Information technology - Semiconductors (analog, digital and manufacturing services) (8/10)
Financials - Diversified capital markets & consumer finance (7/10)
Financials - Financial exchanges & research, investment banking and brokerage (7/10)
6. Innovation capacity
A crucial factor in future outperformance is an industry's capacity for innovation. That is creating new and better products and services for customers and diversifying into new business lines that solve new problems for existing customers. There's a close correlation between innovation capacity and growth potential, and here, I have the following five industries among the top ranking:
Healthcare - Life sciences and tools and biotechnology (10/10)
Information technology - Software (9/10)
Information technology - Semiconductors (analog, digital and manufacturing services) (9/10)
Information technology - Semiconductor equipment (9/10)
Communication services - Interactive media services (8/10)
7. Barriers to disruption
At the other end of the innovation capacity spectrum are barriers to disruption. I analyse how difficult it is for new entrants to enter the industry and outcompete existing players. Several factors make industries hard to disrupt, ranging from quantitative factors such as startup costs and capital expenditure requirements to more qualitative factors such as regulatory hurdles, switching costs and existing brand loyalty. My highest-ranking industries include the following:
Industrials - Transportation: Ground and infrastructure (9/10)
Financials - Insurance (8/10)
Utilities - water, gas and multi-utilities (8/10)
Consumer staples - Food and staples retailing (8/10)
Financials - Insurance (8/10)
Building the industry quality formula
To rank all 60 industries on a consistent broad-based indicator, I synthesise these seven metrics, with metrics such as barriers to disruption (20%) and degree of resilience (20%) representing a larger share of the overall formula while metrics such as knowledge and understanding (5%) and innovation capacity (10%) representing a smaller weight.
Formula
Presence of outperformers (15%) + Knowledge and understanding (5%) + degree of resilience (20%) + replacement cycle (15%) + profitability and earnings growth (15%) + innovation capacity (10%) + barriers to disruption (20%)
There are certainly ways this formula can be further refined and improved to better capture the opportunity set I'm aiming for, but I feel its current format balances simplicity and clarity.
After adding all seven indicators together, each industry is given a score out of 10, and I then rank all 60 industries and split them into five tiers, with tier 1 being industries that score best, and thus industries I intend to prioritise. Here's a list of all the current tier 1 industries:
From the table above, you might notice the broad mix of companies ranging from consumer staples like food and staples retailing and packaged foods and meat to industrial industries like aerospace and defence and ground transportation and infrastructure. These sectors particularly tilt towards industries that produce high moat and high-quality companies that have tested time, and I believe we are more likely to find companies that will be around and thriving in these top twelve industries.
This doesn't mean that being in a good industry guarantees good results. Every industry experiences external shocks and cycles, which is a crucial point to remember. For example, the consumer staples tobacco and beverage industry (the sin industry) is currently experiencing its shocks with the downturn and potential structural decline in alcohol consumption.
On the other hand, being in a bad industry doesn't mean one can't find exceptional companies. The table below highlights the five worst-performing industries and some great investments these industries produced in past years.
As you see from the industries above, even the worst industries can produce outperformers. Here, we want to be more selective and focus on unique setups. For example, Federal Agricultural Mortgage (Farmer Mac), the government-sponsored enterprise (GSE), is a unique American financial company that is high quality and within our core shortlist of companies despite being listed in a poor-performing industry.
The research stages
While useful, ranking industries by areas of focus is just five per cent of the work required, and the real work is going from a list of 300 or 400 companies to one or two companies worth investing in. Admittedly, I've made several mistakes over the years, from relying too much on screeners as filters to poorly allocating time focus. After reflecting and testing different methods over the years, I concluded that developing a four-stage research process works best for me, and I will next explore what these four steps look like. To support my explanation of how this works, I will use an ongoing industry I'm currently assessing: the industrial trading, distributors and conglomerates industry.
Industrials - capital goods (trading companies, distributors and conglomerates)
Industrial trading, distribution and conglomerates have a bad reputation of being shareholder unfriendly given many of these are family businesses, low return and sometimes unfocused companies. While this is true for some, some of the best performing companies globally can be found in the industry.
Among the 497 listed industrial trading, distributors and conglomerate companies, 24 of them have been long term outperformers (they returned 15% over either a 15, 20 or 30 year period), a 4.83% conversion, ranking 15th of 60 industries. More impressively, what truly really set them among the tier 1 industries was their consistency - many of these companies have features such as a high resilience to economic cycles, high product and service replacement cycle given how diversified they are, consistent and growing profit margins and a high barrier to disruption given how critical these companies are for their respective markets. They only scored poorly on innovation capacity, resulting in an average score of 6.9, the 7th highest of 60 industries.
Some good examples of past outperformers from this industry include American industrial distributors like W.W. Grainger, Watsco, United Rentals, Swedish serial acquirers such as Investment AB Latour, OEM International and conglomerates like Metlen in Greece. With dividends included, these six companies have returned 271% (30% per year) over the past five years, 467% (19% per year) over the past 10 years and 2,699% (18% per year) over the past 20 years.
Stage 1 (The initial 5-minute filter)
With all industries, the first step is pulling out a list of all listed companies within the industry, and for industrial capital goods - trading, distributors and conglomerates, that's around 500 companies. We could not deep dive into each, so stage 1 is focused on the initial filter. My goal here is to quickly decipher what sort of company (quality and growth prospects) the company could be. Here, there's no consideration for valuations. In 5 minutes, I scan their business description, figure out what they do, and quickly examine their balance sheet, income and cash flow statements to study recent trends.
If the company looks attractive, it goes into the stage 1 list with a short note on key financial highlights and the business case, e.g. acquisitive growth cyclical with high margins, stalwart, etc.
A screenshot of Stage 1 Industrial: trading, distributors and conglomerate. I always try to include a brief note description to guide stage 2 research and classify the company’s potential investment style e.g. serial acquirer, broken growth story, stalwarts etc.
The image above shows some highlighted companies from the industrial trading, distribution, and conglomerates industry. One company is highlighted in green: Metlen Energy. I flag companies that are particularly interesting from an investment perspective so I can prioritise them in stage 2.
Finally, once I complete the first stage, a 5-minute scan of all companies, I screen companies to ensure I don't miss out on potentially attractive companies. These metrics include:
Return on Capital
EBIT margins
7-year earnings CAGR
Forward 2-year EPS growers
Lowest P/E and EV/EBIT companies
Worst performing share prices on 3-year and 5-year lens
New listings (last 5 years)
Scanning by these metrics can highlight companies worth investigating. For example, OEM International in Sweden was a company I missed in the initial full list scan but was flagged due to its high return on capital. Some flagged companies in these metrics don't make the final cut due to potential issues, and one example of one of these was Idun Industrier AB, which is listed in Sweden (see image above).
As a reminder, we still know nothing about the individual companies at this stage but have some understanding of the opportunity set, which sets us up for the second stage.
Stage 2 (Learning key drivers)
With 70 companies in total from our stage 1 initial screen, there are still too many to investigate deeply, so we still need to spend some time learning the basics about each company, albeit quickly.
In stage 2, I aim to spend 45 minutes studying each company, examining the website, the latest earnings presentation and a brief scan of their latest annual report. At the end of the 45-minute research, I assign a potential earnings growth and exit multiple to calculate a potential IRR over 4 years and then assign a score on quality (50%, growth (25%) and value (25%). Of course, it's impossible to have an accurate guesstimate of four years of earnings growth on an hour of research, so these numbers are typically guided by long-term management goals, a mix of past track records, and what seems achievable given market conditions and overall industry dynamics.
Above, I share a screenshot of how this looks like, and as you see on the right-hand side, there's a potential IRR column and the initial rating on the left-hand side (a sum of quality, value and growth out of 10). Highlighted in green are companies that meet the 15% hurdle rate in stage 2 and advance to stage 3 for deeper research, while in amber are companies that appear to be of high quality (quality score above 7.5 out of 10) but don't meet our IRR hurdle due to high valuations. Take Lifco, for instance which was valued at 48x earnings during the initial research, a price that seemed too high given its growth outlook.
It's important not to discard these "expensive" but high-quality companies. In stage 3, I maintain a list of all amber companies and set notifications if their share prices ever experience sharp declines. Reece Limited, an Australian-listed plumbing and bathroom products distributor, appeared too expensive at the time of the research but has since fallen by 30% in share price and while still at 26x earnings, a price above what I'd want to pay, I've taken note of its current valuation and could potentially move into the core industry shortlist on further price declines.
Overall, 15-20% of all stage 1 companies might typically transition to stage 3, so of 70 companies, we have 14 in stage 3.
Documenting the process
While it seems unprecedented to calculate IRR targets this early in an investment process, it is helpful, especially later on when actual results come out and comparing estimates with actual performance has been a great way for me to reflect on how I assess opportunities. For example, it's possible Lifco could potentially grow much faster by 19% and command an exit multiple even higher than my estimate of 28x (see image above), leading to a higher IRR and by comparing their actual results, it's much easier for me to reflect on what I'd missed when examining Lifco. This is why I've become a big believer in documenting the process for even companies I don't initially find attractive.
Next is stage 3, where the real work starts.
Stage 3 (Fundamental profiling)
If the first two stages were about scanning, stages 3 and 4 are about diving into the fundamentals. Here, I'm focused on learning the key drivers of each shortlisted company. I do this by reading the annual report, earnings calls, and presentations and studying the web for any information about its operations.
For example, Trinidad & Tobago is a country I never guessed I'd find something interesting in, but Agostini's Limited, a national leader in food & healthcare product distribution and industrial services, turned out to be a business I admire. First, it is misclassified (it should be consumer staples distribution, not industrial distribution) - nearly all of its revenue comes from pharmaceutical and consumer distribution, not industrial. It's also built an enviable track record, compounding earnings by 18% over the past 19 years, a great culture and a sharp eye for growth across the Caribbeans and at 16x earnings, I think we can double our money in its stock over the next five years.
Another example of a high-quality company we found even more attractive after digging deeper was Metlen Energy. Here, its founder clearly stated an ambition to double its revenue and profits in five years, and they've made well-timed investments in power and renewables over the past few years after surviving the worst of the Greek debt crisis. At the time of our shortlist, Metlen was valued at 7x earnings, which seemed too low for the well—managed conglomerate.
On the other hand, some companies don't turn out to be as attractive as I initially thought. Here, Teqnion is a good example. While I still see its potential, it seems the Swedish serial acquirer may have overestimated the durability of moats in previous acquisitions and will likely spend 2025 and 2026 fixing past mistakes rather than doubling on new acquisitions, making my initial 16% earnings CAGR projection between 2024 and 2028 potentially unattainable. Ashtead Technology, on the other hand, hasn't necessarily made bad decisions, but upon learning more about its end users in oil and gas, it seems its market is more cyclical than I initially judged, and a downturn in oil prices below $60 could impact its revenues and bottom-line which dampens the initial quality score.
Overall, I plan to spend 5-6 hours studying each of the stage 3 companies and should have covered their latest financials, updated earnings growth, and exit multiples projections. Once completed, companies below the 15% IRR are left on stage 3, while those like Investment AB Latour, Metlen Energy, and Agostini's Limited advance to stage 4.
Stage 4 (Investment sleuthing)
Admittedly, stage 4 has recently become my favourite part of the investment process, sleuthing. I focus on information beyond the company accounts, ranging from management meetings to customer and supplier reviews and interviews, broader industry deeper dives, and trade shows, among other sources. I spend around 20 hours on each company, and on average, only 1-2% of the companies within the initial industry dive make it here. In the industrial trading, distribution and conglomerates, 6 of 500 companies progressed passed stage 4 - Metlen (Greece), Latour (Sweden), Agostini's (Trinidad and Tobago), LT Group (Philippines) and two other companies I will reveal later on.
Sleuthing: I visited the Metlen Energy headquarters in Athens and got to learn more about the culture and people at the conglomerate.
The research in this phase is what truly matters, and improving my process at this stage is my number one priority going forward. In fact, a primary reason for launching this newsletter is to chronicle companies that make it to stage 4 of the research process. So far, I've written about three: Alphabet, the parent of Google, and two Mexican airports, OMA and ASUR.
I also employ other tools, such as the Jenga IP 60-point investment checklist, which you can read here. Once assessed on my investment checkpoint, I quantify all our research on the Jenga IP company score. In the image below, you can see how some recently studied companies have performed. You can also read the latter half of the Alphabet deep dive I published earlier to see this stage 4 research in action.
Final thoughts
"Investing may be an art, but it's one painted with the brushstrokes of science" is a motto I currently gravitate towards, and while the investment process isn't flawless with room for further refinement, building and documenting your process is highly impactful as it helps highlight strengths and weaknesses, improves our ability to find new opportunities repeatedly and lastly, improves our chances of staying disciplined to what we believe matters. By repeating this process with six industries each year, thus 30 industries covered every 5 years and 60 industries covered every 10 years, I hope this allows us to go through A-Z of all listed companies in a systematic way.
Feel free to play with the industry indicator spreadsheet, let me know your thoughts, and also make your own adjustments to the industry scores.
Dear Dede, thank you for sharing your process it most enlightening. It is like you have given me a torch as I was finding my way through darkness!
Excellent work!