Q2 Framework 2026

Question 1: What has the market stopped forgiving in Q2 2026?

It has stopped forgiving indecision dressed up as prudence. In Q2 2025, it was reasonable for management teams to say they were still evaluating tariffs, sourcing changes and customer responses because the policy environment was moving quickly. During Q1 2026, the same allowance applied to energy, logistics and geopolitical disruption. There were genuine limits to what a company could know, and investors understood that.

That period of interpretive grace has largely ended. I do not expect management to have solved every problem, but I do expect the company to have learned enough to make decisions. If the answer is still that the situation is being monitored, I start to wonder whether the organization is cautious or simply slow. Uncertainty is not a reason to avoid judgment. It is the environment in which judgment becomes valuable.

Question 2: Which Q2 earnings would you refuse to carry forward in your valuation?

I would be cautious about any earnings produced by a temporary imbalance that management presents as a permanent improvement. That could be scarcity pricing, customers ordering earlier than usual, a competitor being unable to supply the market, or an unusually favorable commodity or currency move. Those earnings are real, and shareholders are entitled to them, but that does not mean they deserve to be capitalized for the next five years.

The comparison with earlier quarters is useful. Q2 2025 contained a great deal of tariff-related purchasing and inventory movement. Q1 2026 contained energy and logistics effects that altered pricing and availability. By Q2 2026, management should know which customers stayed, which orders repeated and which margins held after conditions changed. I would underwrite evidence of a better business. I would not underwrite a favorable accident indefinitely.

Question 3: What would make you reduce a position after the company reported a clear earnings beat?

I would reduce the position if the beat revealed that the business now requires much more capital to produce the same amount of growth. I would also be concerned if cash flow was materially weaker than earnings, if working capital was doing too much of the work, or if the company raised guidance without explaining what had changed beneath the number. A beat is not automatically good news when it comes from postponing maintenance, reducing necessary investment or pulling revenue forward.

The market was willing to reward relief in Q2 2025 because expectations had fallen sharply. It rewarded resilience in Q1 2026 because companies were operating through a significant external shock. In Q2 2026, the hurdle is higher. I am not asking whether the company beat a quarterly estimate by two cents. I am asking whether the quarter increased or reduced my confidence in the next several years of cash flow.

Question 4: Are current margins evidence that companies have become better businesses, or are some of the costs simply arriving late?

Both things are happening. Some companies have genuinely improved. They have simplified portfolios, invested in automation, redesigned processes, and become more disciplined about pricing. Those changes can produce durable margin improvement. Other companies are benefiting from favorable mix, temporary shortages, delayed hiring or the fact that depreciation and operating costs from recent investments have not fully reached the income statement.

In Q2 2025, investors were worried about tariffs reducing margins. In Q1 2026, they focused on energy, transportation, and financing costs. Q2 2026 requires management to discuss the margin after all the relevant costs arrive. I want to hear about depreciation, power, labor, maintenance, customer incentives and the cost of supporting new capacity. When management calls the current margin a new baseline without addressing those items, I assume the baseline is too optimistic.

Question 5: Where do you think the market is most at risk of confusing a cycle with a permanent competitive advantage?

Parts of the AI infrastructure market are the obvious example, although the issue is broader than technology. Scarcity can create extraordinary pricing and margins for a period, but scarcity also attracts investment. Capacity expands, customers gain alternatives and negotiations become more difficult. A very good business can still produce disappointing shareholder returns if investors pay as though today’s shortage will last forever.

A year ago, the question was whether AI demand was genuine. By Q1 2026, the investment cycle had become large enough to support broader economic activity. The question in Q2 is what remains after the buildout becomes more competitive. Has the company created an installed base, proprietary data, customer dependence, a lower cost position or some other advantage that survives normalization? If not, I would treat the earnings as cyclical even when the technology itself is transformational.

Question 6: Has the market mistaken an AI capital-spending boom for an AI profit boom?

In some cases, yes. The spending is undeniable, but spending and returns are not the same thing. Suppliers can earn exceptional revenue while the system is being built, yet it may take years to determine whether the owners of the infrastructure earn their cost of capital. The companies financing the expansion, buying the equipment, and operating the assets do not necessarily capture the same economics.

This distinction matters much more in Q2 2026 than it did in Q2 2025. At that stage, proof of demand was enough to change the investment debate. Now I want to understand utilization, depreciation, energy expense, financing, customer concentration and the useful life of the assets. I would rather miss the first part of a rally than own a business whose reported growth hides a deteriorating return on invested capital.

Question 7: What is the most important AI cost that management teams are still underexplaining?

The cost of remaining competitive after the first investment. Companies often describe the initial spending on models, software, chips, or infrastructure, but the larger obligation may be the continuing need to upgrade, retrain, secure, operate, and integrate those systems. An investment that appears to be a one-time strategic step can become a permanent increase in the cost of doing business.

That was easy to overlook in Q2 2025 because the discussion centered on pilots and adoption. Q1 2026 made the scale of the investment more visible, but Q2 2026 is where investors should begin asking about the second and third rounds of spending. I want to know whether the investment creates a durable advantage or simply buys admission to the market. Those are very different shareholder propositions.

Question 8: How do you distinguish a real AI advantage from basic modernization?

I ask what would remain differentiated if every competitor had access to the same models and computing resources. If the answer is nothing, then the company may be becoming more efficient, but it is not necessarily strengthening its competitive position. Efficiency matters, although it may ultimately be passed to customers or competed away.

The more interesting businesses combine the technology with something that is difficult to reproduce. That may be proprietary data, customer relationships, distribution, regulatory expertise, workflow ownership or a culture that can implement change more effectively than peers. In Q2 2026, I do not need another list of use cases. I need to understand why the economic benefit stays with this company rather than migrating to its suppliers, customers or competitors.

Question 9: What does a genuine compounder do differently after a period of disruption?

A compounder uses disruption to improve the business rather than simply returning to where it started. It may deepen a customer relationship when competitors cannot deliver, improve its supplier position, enter a market at an attractive moment or take share without sacrificing long-term economics. The key is that the benefit remains after the external condition fades.

Q2 2025 tested trade and sourcing. Q1 2026 tested energy, liquidity, and operational continuity. In Q2 2026, I want to see whether the company converted those experiences into a better organization. Did it shorten decision times, improve pricing discipline or create a more flexible operating model? Survival is valuable, but compounding begins when the company emerges from the disruption with a stronger competitive position than it had before.

Question 10: When does resilience become expensive theater rather than a real investment?

It becomes theater when management can describe everything it has added but cannot explain what those additions protected or enabled. More suppliers, more inventory, more regional capacity, and more contingency planning can all be sensible. They can also create a slower, more complex and less profitable company if the decisions are not connected to an important economic outcome.

Investors were understandably tolerant of resilience spending in Q2 2025 and Q1 2026 because the risks were highly visible. The standard should be higher now. Did the extra inventory protect revenue? Did the second supplier keep a major customer? Did regional capacity allow the company to gain share? Resilience is not automatically valuable because it reduces one category of risk. It is valuable when the protection it provides is worth the capital and complexity it consumes.

Question 11: Which demand indicators do you trust least this quarter?

I am skeptical of order growth without an explanation of customer behavior. Customers may be reserving capacity, ordering ahead of tariffs, building safety stock or placing orders with several suppliers because they are worried about availability. A strong backlog can look reassuring while containing much less economic commitment than investors assume.

That problem was obvious during Q2 2025, when purchasing patterns changed around trade policy. It reappeared in Q1 2026 as energy and transportation risk affected availability. In Q2 2026, management should be able to discuss cancellations, conversion, delivery timing, and customer concentration. I trust orders when they become revenue, cash and repeat business. I trust them less when management asks me to accept the headline number without understanding what produced it.

Question 12: What are companies still failing to explain about the consumer?

They are still speaking about the consumer as though there were one consumer. Higher-income households, middle-income families and financially constrained customers are experiencing the environment differently. A company can report resilient total demand while transaction frequency declines, promotions increase or the weakest customers leave the category entirely.

The changes since Q2 2025 are important. Tariffs affected prices and purchasing decisions, while Q1 2026 added pressure through energy and food costs. By Q2 2026, I expect companies to understand which customer groups are holding up and which are becoming more selective. I want to hear about trade-down, basket size, frequency, credit use and customer-acquisition costs. The average is often the least useful number when the customer base is becoming more divided.

Question 13: How do you know whether a company has real pricing power?

Real pricing power is not the ability to increase prices during a shortage. It is the ability to preserve attractive economics while maintaining the customer relationship. I look at volume, retention, market share, and competitive response after the price change. A company that raises price and loses its best customers may have demonstrated leverage, but it has not demonstrated durable pricing power.

In Q2 2025, pricing was largely about passing through tariffs. In Q1 2026, it was about energy, freight and other input costs. Q2 2026 is the point at which investors can assess what stuck. If prices remain higher because the product is more valuable or deeply embedded, that matters. If they remain higher only because customers have not yet found an alternative, I would be careful about assuming the benefit lasts.

Question 14: What capital-allocation decision tells you the most about management quality?

The decision to stop funding something. Starting a project is easy to present as optimism and ambition. Stopping one requires management to acknowledge that an earlier assumption was wrong or that the opportunity is no longer attractive. That is often a more revealing test of discipline than approving another investment.

Many commitments were made during Q2 2025 to respond to tariffs and during Q1 2026 to address resilience, energy or capacity concerns. Those decisions may still be correct, but they should be re-underwritten. I want to know whether management would make the same commitment today using current information. A company that treats every past decision as untouchable will eventually confuse consistency with capital destruction.

Question 15: Are boards challenging management hard enough on AI and strategic investment?

In many cases, I doubt it. Boards may accept that an investment is strategically necessary without asking whether the scale, timing and financing are appropriate. “We cannot afford to fall behind” can be a legitimate concern, but it can also end the discussion before the return analysis begins.

The commitments are larger and less reversible than they were a year ago. That makes board oversight more important, not less important. I want the board to ask what must be true for the investment to work, which indicators would show that the thesis is weakening, and whether management has preserved the ability to change course. The board does not need to operate the business, but it should make sure strategic urgency has not become a substitute for financial discipline.

Question 16: Where is balance-sheet risk most likely to be underestimated?

It is often underestimated in commitments that do not look like conventional debt. Long-term capacity agreements, leases, supplier financing, purchase commitments, project structures and guarantees can consume future cash flow even when the reported leverage ratio looks comfortable. The accounting treatment may be correct, but the strategic flexibility may still be narrower than investors realize.

The concern is greater in Q2 2026 because the investment cycle extends beyond the cash-rich companies at its center. More businesses are relying on debt, partners or long-term contracts to fund growth. I want to know how much capital the strategy still requires and what happens if utilization or demand develops more slowly. A strong balance sheet is not simply a low leverage number. It is the ability to keep making good decisions when the original plan does not unfold perfectly.

Question 17: How has the global competition for investor capital changed over the past year?

It has become much broader. Companies are not competing only against the closest domestic peer. They are competing against global equities, credit and businesses benefiting from different parts of the investment cycle. A company with a premium valuation needs to explain why its growth, returns and competitive durability justify that premium relative to all those alternatives.

That makes the quality of the company-specific case more important. A US business cannot rely on the strength of its index. A European or Japanese company cannot rely only on being cheaper. An emerging-market company cannot ask investors to overlook governance or currency risk because the growth rate is attractive. Q2 2026 offers investors real choices. Management teams have to explain not simply why the company is good, but why the expected return is better than the return available elsewhere.

Question 18: When does a diversified corporate portfolio deserve to remain together?

It deserves to remain together when common ownership produces an advantage that would be difficult to achieve separately. That may involve customers, technology, procurement, distribution, data or capital. The advantage should be visible somewhere in growth, margins, returns or reduced volatility. If management can describe only broad strategic benefits, I become skeptical.

The past year created a useful test. Q2 2025 showed whether a diversified footprint helped companies respond to trade barriers. Q1 2026 showed whether geographic and operational diversity provided protection from energy and logistics disruption. Q2 2026 should reveal whether those benefits were real. If the businesses did not support one another when conditions became difficult, investors are entitled to ask whether another owner could create more value.

Question 19: What management behavior raises your confidence most quickly?

A clear explanation of what management previously believed, what changed and what it did next. I do not lose confidence because a company changes its mind. I lose confidence when it changes the story without acknowledging the change or continues defending an old position after the evidence has weakened.

The strongest teams are comfortable distinguishing fact from judgment. They can say what they know, what they are assuming, and what would cause them to reconsider. That was important in Q2 2025 and Q1 2026, but it is even more important now because investors have a longer history against which to test management’s claims. Credibility comes from an honest decision process, not from pretending every decision was obvious in advance.

Question 20: What behavior causes you to lose trust even when the financial results remain strong?

I lose trust when the explanation becomes more complicated as the evidence becomes weaker. A company may begin with a clear strategic argument, but over time more exceptions, adjusted measures, and shifting milestones are required to defend it. Strong current earnings do not resolve that concern if management is avoiding the underlying issue.

I also become cautious when management claims a favorable outcome as evidence of superior execution but describes an unfavorable outcome as entirely external. That asymmetry tells me more than the quarterly number. Investors understand that luck exists in business. We become uncomfortable when management appears unable to recognize when luck helped or when its own judgment fell short.

Question 21: What question should every CEO answer before investors have to ask it?

The CEO should answer this: which assumption changed during the quarter, and what did you do because it changed? That question reveals whether management is actually learning or simply producing another explanation after the result. It also forces the company to connect information with action.

The answer should not be limited to a tactical adjustment. It may involve the pace of investment, the attractiveness of a market, the behavior of a customer segment or the durability of an advantage. Q2 2025 and Q1 2026 generated an enormous amount of new evidence. The best management teams will show that this evidence changed the way they allocate capital and run the business.

Question 22: What would world-class investor communication sound like in Q2 2026?

It would sound thoughtful, specific and economically grounded. Management would identify the few changes that mattered to the company, explain the decisions made in response, and show how those decisions affect revenue, margins, cash flow, risk and capital requirements. It would separate what the company created from what the environment provided.

The conversation would also leave room for uncertainty. I do not expect management to know exactly how technology, inflation, geopolitics or customer behavior will evolve. I do expect it to understand which variables matter and what evidence would trigger another decision. The best communication does not make the future appear simpler than it is. It makes the company’s way of navigating that future easier to understand.

Question 23: What is the one conclusion you would want management teams to take from this reporting season?

Q2 2025 was largely a test of exposure, and Q1 2026 was a test of resilience. Q2 2026 is a test of whether the company can convert what it learned into better long-term returns. That is a more difficult standard because it requires management to show not only that the business survived, but that its decisions improved the business.

The market is not asking management to predict every outcome. It is asking whether the company can make a sound decision while the evidence is still incomplete, allocate capital without being captured by the prevailing theme and change course without losing strategic direction. That is the kind of judgment I am willing to underwrite.