Dismantling the "Serenity" investment methodology

The Serenity Bottleneck Investment Act, in short, identifies a major trend of certainty, dismantles the industrial chain, finds the most difficult upstream link to replace, and finally places a bet before the market is sufficiently priced。

FOR EXAMPLE, WHEN THE MARKET IS NOT AWARE THAT THE LIGHT UPGRADE OF THE AI DATA CENTRE WILL MAKE AN UPSTREAM MATERIAL, LASERS, TEST EQUIPMENT A SCARCE ASSET, THIS SMALL SEGMENT COULD BE RE-PRICING FAR BEYOND THE BASIC CURRENT INCOME。

Like a restaurant, the most expensive is the main course, but the real thing that's stuck is a little bit of a tributary; if this is out of supply, all the main course can't be made。

It's like a five-factor model when it's essentially a bottleneck method:

The demand side must be sufficiently defined, the supply side must be sufficiently narrow, market perceptions must lag, potential values need to be sufficiently clear, and verifiable events must catalyse in the future。

Five conditions are met at the same time for small companies to be able to generate excess earnings。

First tier: identification of needs。

AI DATA CENTRE EXPANSION, CLOUD MANUFACTURERS ASIC, SELF-STUDY CHIPS, REASONING NEEDS, BANDWIDTH NEEDS, WHICH FORM THE CONTEXT OF DEMAND。

The repeated references to AMZN Trainium, MSFT Maia, Google TPU, NVDA push 800V DC, show that he did not look at a small company in isolation, but instead put it into capital expenditure and structural migration of giants。

For example, he wrote logic in the AAOI/LITE tweet that markets reward Google TPU supply chains, but may underestimate AMZN Trainium and$MSFTMaia's light interconnection needs。

Second tier: restricted supply。

The so-called bottlenecks are not “this thing also benefits” but “not without it” and are not easily replicated in the short term。

For example, the InP liner, the CPO external light source, CW DFB laser, SOI crystal circle, light transceiver test equipment, etc., all sound cold, but once the AI data centre migrates from electrical to light connection, these links become bottlenecks to capacity, good rates, certification cycles, and customer import。

For example, in the InP liner, the InP has a key position in high-speed light communication lasers, detectors and some photons, especially in the direct gap, light efficiency and high-speed layout scenarios。

At the same time, it is difficult to quantify reproduction in the short term due to, inter alia, the certification cycle, long periods of time spent on equipment, high barriers to production processes, the rate of expansion of capacity to keep pace with surges in demand and structural shortages。

Third tier: low focus。

Low focus = real low price。

Many of the targets of Serena are not in the mainstream narrative centres, but there is a greater risk of misprice where “institutional coverage is low, dispersed, and the media is not clear”。

Level 4: Value capture。

Whether there are pricing rights, Maori space, customer targeting, supply shares。

The real bottleneck translates into excess earnings, with several conditions in the middle: whether the company has access to capacity, can pricing, can be pressurized by its customers, needs for financing diluted, whether the Maori are cashier, and whether the demand has been overdrafted at an earlier stock price。

Level 5: Catalyst。

While long-term space is important, short-term catalysts are also price engines。

Short- and medium-term triggers: financial reporting, customer mass production, Jabil Freside Chat, CHIPS Act, index inclusion, Nasdaq double listing, M& A, crowded, capital flows from local markets to American investors are good clues and catalysts。

$AXTI: THE CLASSIC CASE OF BOTTLENECKS。

Since the early days in Reddit, when AXTI was analyzed, why was it sealed

At that time, the AXTI market value was small, the business was cold, and the main battalion, the InP line, was considered a “push ticket”; however, Serena understood that the IAI data centre light communications required IP-based materials, and that the entire photosupply chain would be affected if supplies were limited。

THE SUBSEQUENT $AXTI INCREASE OF NEARLY 10 TIMES FROM APPROXIMATELY $14 FURTHER DEMONSTRATED CORE COMPETENCIES: RATHER THAN LOOKING AT STOCK PRICE INCREASES, IT IS A FIRST STEP IN DETERMINING WHETHER THE CHAIN WILL CHANGE FROM A “COLD-DOOR MATERIAL” TO A “STRATEGIC BOTTLENECK”。

2. $RPI: SMALL MARKET VALUE COMPANIES ARE EXTREMELY SENSITIVE TO MARGINAL DEMAND。

The same change in demand may be a 1 per cent income disturbance for large companies and a revaluation of small firms。

FOR EXAMPLE, THE INCREASED DEMAND FOR AI HARDWARE, DEVELOPMENT BOARDS, PERIPHERAL EQUIPMENT HAS LIMITED IMPACT ON GIANTS SUCH AS APPLES, BUT$RPIThis smaller hardware company may change the growth curve directly。

Serenity's judgment is that if AI anent requires a large amount of low-cost local nodes or edged hardware, this “small computer” may suddenly become an infrastructure for the proliferation of AI applications。

3. AAOI /LITE: from single-point bottlenecks to supply chain maps。

Serenity places LITE in the TPU / OCS profit chain, places AAOI in the MSFT Maia and AMZN Trainium camp chains, and suggests that IP may become a 2026 bottleneck like HBM。

The bottlenecks are not just looking at the points, but thinking about them on the wire and in the face: when the Google TPU chain is rewarded by the market, the next step may be to excavate AMZN, MSFT self-study ASIC-related light connections。

copyticker is easy, learning how to think and how to implement it. if you really can hold the marker, you have to form your own knowledge system。

So how do we use Serenity's thinking path better? There are six steps。

First step: Looking for broad trends: has demand been validated

Let's see what we can do first。

FOR EXAMPLE, AI ' S ALGORITHM EXPANSION, CPO LIGHT INTERCONNECTION, 800V DCS, HUMAN ROBOTS, STABLE CURRENCY PAYMENTS, AND RWA MONETIZATION ARE TRENDS。

If the trends themselves are not certain, the latter supply chain analysis is also a calf。

Step two: Drawing maps: what are the links from terminal to upstream

Draw the chain。

FOR EXAMPLE, WE CAN'T JUST KNOW$NVDAAnd know about ASIC, switches, light modules, external light sources, lasers, InP/SOI materials, seals, tests, fibre-optic arrays, microscopes, etc。

Serenity itself mentioned that it would not be able to read enough if the chain of light communications industries were not covered from the upstream IPP to the downstream light module。

Step three: Looking for bottlenecks: which is the hardest to expand/replace

Whether it is a real bottleneck or a false bottleneck。

The real bottlenecks usually have several characteristics: concentration of supply, long certification cycles, high customer switching costs, poor technology rates, slow expansion and reliance on mega-road maps。

Hypocrisy bottlenecks are usually “in the chain of industries”, but without scarcity, anyone can do it and the ability to raise prices is weak。

Step four: Is there a customer, authentication, capacity, order trail

It's the evidence, not the emotions。

Evidence could include customer trails in annual newspapers, management minutes, vendor profiling, CHIPS Act/Government funding, index inclusion, patents, recruitment, expansion of capacity, cooperation announcements, customer product road maps, and industry capex。

THE HIGHEST LEVELS ARE CORPORATE ANNOUNCEMENTS, REGULATORY DOCUMENTS, FINANCIAL REPORTS/TELECONFERENCES; THE INTERMEDIATE LEVELS ARE CUSTOMER WEBSITES, RECRUITMENT, PATENTS, SUPPLIER LISTS, GOVERNMENT PROJECTS; AND THE LOWEST LEVELS ARE PEER MAPPING, AI PROMOTIONS, SOCIAL MEDIA. THREE TYPES OF EVIDENCE HAD TO BE SEPARATED, OTHERWISE IT WOULD BE EASIER TO DEDUCE。

Step five. Wind control: If wrong, where is it

The "reverse watch" must be made。

Let's be bold and careful. Not once and for all。

if the customer does not limit, when does the income become false? if the competition is replaced, do the bottlenecks disappear? if valuations were overspent earlier, could stock prices afford a window of performance? if overtransmission leads to overcrowding, who will take the last one? if corporate finance, dilution, financial restatement, full case changes

Step 6: Match the position and depth of the study。

If you just look at other people's summaries, the space should be small; if you can draw the chain of business, read the annual newspaper, disassembly the customers, do the scenario valuation, it can be even bigger。

Learning methodology is accompanied by a bowl of cold water. Because there are limits to the best。

It is easy to extrapolate。

Serenity is very good at combining regulatory documents, cooperative announcements, customer websites, financial language, but this approach is naturally at risk of miscalculation. The removal of a supplier from the customer website, the presence of a company in the Blueprint and the association of a partner with hyperscaler may be strong leads or just noise. There is a need to distinguish between inferences and facts。

Let's be bold. Take care。

2. When early finance was poor, valuations were not anchored。

For targets such as SIVE, XFAB, AAOI, Serenity often looks at future revenues, ramp, structural migration and potential M& A, not current profits。

This is a high rate of success in the right direction, but is prone to miscalculation in the wrong direction。

3. Liquid reverse risk: Serenity has become a market variable。

Serenity is not an ordinary researcher, but a market participant with hundreds of thousands of fans, high subscriptions and quoted by the media. Once he has publicly looked after a small capitalization, following him may push the price directly and have a direct impact on the rate of compensation。

There has also been some bias in the identification of survivors。

THE RATE OF RETURN OF UP TO 4500%, IN ADDITION TO THE LOGIC THAT IS WORTH LEARNING, IS LARGELY DUE TO CATCHING UP WITH AI'S CALCULATOR, THE BIG UNILATERAL COW MARKET。

Serenity is great, but we have to be careful。

Past experience may not be tested in the future, and will the giants try to bypass it after the current neckline

In addition, the success of Serenity requires, in addition to its strong analytical power, a steady deposit of first-hand information and a strong heart that resists retreat。

That's the same sentence, bold assumptions, careful testimony. Responsible for their own positions。

 

In other words, the Bottleneck Investment Act is effective because the market tends to price major narratives before second-tier suppliers and finally realizes the real shortage of materials, devices, tests and production chains。

But the most dangerous place for this approach is here: it relies heavily on professional judgement, information puzzles, non-consensus tolerance and silo discipline。

What we really need to do is not use Serena's holdout, but his research sequence: find trends, find bottlenecks, find evidence, look at valuations, wait for catalyzes, and then bet on affordable slots。

And finally, looking at Serenity's methodology, there's only three words left in the head: walk the narrow door。

IN LARGE TRENDS SUCH AS AI, RATHER THAN BUYING THE MOST VISIBLE POPULAR SHARES, IT IS GOING DOWN THE INDUSTRIAL CHAIN TO FIND THE MOST DIFFICULT BOTTLENECKS IN FUTURE STRUCTURAL MIGRATION AND TO BET IN ADVANCE WHEN OLD FINANCIAL ACCOUNTS, OLD VALUATIONS, OLD GEOGRAPHICAL BIASES STILL SUPPRESS PRICES。

It is a narrow door to investment or to life。