Chapter 088　What Did Meta Redefine?

What did Meta redefine? The question has an easy answer and a hard one. The easy answer is the name. In October 2021, the company called Facebook became Meta. It is a rare case of an enterprise publicly declaring that it was rewriting its own reason for existing. The hard answer is what that declaration changed, and what it did not. This chapter rests on public information alone and measures the distance between declaration and reality. We neither celebrate nor condemn. One thing is in question. Does redefinition come into being by declaration?

1 Why this company is worth the question

Start with the declaration. On October 28, 2021, Mark Zuckerberg published “Founder’s Letter, 2021.” He wrote that the company would be renamed Meta and stated that it would operate “metaverse-first, not Facebook-first.” At the same time he set out two reporting segments: the family of apps, and future platforms. This is a decision to replace the identifier of the enterprise itself. A more explicit declaration of Purpose is rarely seen at a large company. Now the reality. Every figure comes from the company’s official financial disclosures. Revenue for the fiscal year ended December 2025 was $200.97 billion, up 22% year over year. Family of Apps accounted for $198.76 billion of that. Reality Labs revenue was $2.21 billion, with an operating loss of $19.19 billion (announced January 28, 2026). In the second quarter of 2026, ended June 30, 2026, advertising accounted for $59.36 billion of $60.801 billion in revenue. Reality Labs revenue was $431 million, with an operating loss of $4.619 billion (announced July 29, 2026). Place the two side by side and a plain fact appears. Advertising is about 97.6% of quarterly revenue; Reality Labs is about 0.7% (arithmetic from the official figures above). The two do not sum to 100%, because advertising is a revenue line and not a segment. The remaining 1.7%, about $1.0 billion, is Family of Apps revenue other than advertising. Nearly five years after the name change, revenue still comes almost entirely from advertising. There are three reasons this company is worth the question. First, the declaration was unusually explicit. Most corporate Purpose refreshes go no further than swapping a slogan. This company swapped its identifier. The conditions for verification exist. Second, reality barely moved. The composition of the business has run on the two segments defined in 2021. Few cases contrast the force of a declaration with the immobility of a revenue structure so sharply. Third, a second declaration is under way now. The company currently leads with AI and “personal superintelligence.” We can run the same test in something close to real time. What we want to extract is not a verdict of success or failure. It is the condition under which a declaration converts into reality.

2 The conventional answers and their limits

Three explanations of this company are in wide circulation. Each is partly right. Each mistakes the question. The first conventional answer: “The metaverse failed, so the redefinition failed” This argument rests on the scale of the losses. The numbers are facts. The Reality Labs operating loss for full-year 2025 was $19.19 billion, and $8.65 billion in the first half of 2026 (official financial disclosures). Segment revenue over the full year was only $2.21 billion. But the argument shrinks the word redefinition down to the success or failure of a new business. Enterprise Redefinition is a system with five dimensions. One losing bet in the business dimension does not mean nothing happened in the other four. The argument also misses the residue of the investment. On the second-quarter 2026 earnings call the company said it had launched new glasses developed with EssilorLuxottica. The accumulated design, manufacturing, and distribution capability put into head-mounted devices survives in a different form. The single word failure cannot handle that residue. The second conventional answer: “It is an advertising company after all, and nothing has changed” This argument rests on the revenue mix. It does not contradict the facts either. But what the word advertising denotes has changed over these five years. On the second-quarter 2026 earnings call the company gave specific figures. Improvements to its user-understanding models raised ad clicks on Facebook by 8.3% and conversions by 15.7%. Advantage+ reached an annual run rate of $75 billion. Business Agents were made available worldwide on WhatsApp and Messenger, with more than one million businesses using them weekly. Inside the same accounting line called advertising, what is being sold has moved. From inventory to the certainty of an outcome. Look only at the revenue mix and the change itself cannot be observed. The second conventional answer mistakes an accounting line for reality. The third conventional answer: “It uses an open-source strategy to lock developers in” This argument concerns the AI domain. On April 5, 2025 the company released Llama 4 Scout and Maverick, distributing the weights in a form anyone could obtain. The official blog stated that “openness drives innovation and is good for developers, good for Meta, and good for the world.” That approach certainly widened the developer base. Note that calling this form of distribution “open source” is itself contested. But public information from 2026 shows a different picture. On April 8, 2026 the company announced Muse Spark, a new model from Meta Superintelligence Labs. The official announcement said it would be provided in a limited API preview to selected partners, and added that “we hope to open-source future versions of the model.” The weights of the newest model are not public. On the second-quarter 2026 earnings call, Zuckerberg said: “We’ve always basically said that we were going to do a mix of open and closed, and that continues to be true.” He also said, in substance, that the open models are not as strong as the frontier models. The lock-in story loses its premise here. What all three miss All three ask what kind of company this is by asking what it sells. But the second dimension of Enterprise Redefinition does not ask what a firm sells. It asks what value the firm delivers. There is a more fundamental omission. Nobody has tested the first dimension. Was the name change a rewriting of Purpose? Or was it a declared bet about the means of realizing Purpose? Without that distinction, five years cannot be assessed.

3 What was redefined — measuring the distance

between declaration and reality across five dimensions We read the case along the five dimensions of Enterprise Redefinition. The order follows the canon: Purpose, Business, Organization, Capital, Leadership. What follows is what public information supports. It does not assert anything about internal decisions. Purpose — the core did not move; only the name did The canon places an important note here. A Core Purpose can remain stable while its expression and its means of realization evolve. Apply that note and the character of the name change looks different. The company’s Form 10-K for the fiscal year ended December 2022 recorded the mission as “give people the power to build community and bring the world closer together.” As of August 1, 2026, the company information page states: “Build the future of human connection and the technology that makes it possible.” The expression changed. The core did not. Both place the connection between people at the center. The 2021 letter itself said the central mission was unchanged. What was declared to be changing was the place in which the mission would be realized. We therefore read it this way. The name change was not a redefinition of Purpose. It was a declaration about the means of realizing Purpose, backed by the most expensive collateral available: the company’s name. The object of the declaration and the collateral for it did not match. That mismatch governed the next five years. Stake the name of your existence on a bet about means, and correcting the bet looks like a denial of existence. The cost of retreat jumps. Business — what moved was not the place that was declared Two things happened at once in the business dimension. In the declared place, almost nothing moved. Reality Labs has existed as an independent reporting segment since it was set up in 2021. As of the first half of 2026, its share of revenue is still under 1%. In the undeclared place, movement ran deep. On the secondquarter 2026 earnings call, the company’s chief financial officer explained a shift away from scoring individual ads toward inferring ad content and user preference together. The unit being sold has moved from placement inventory to the accuracy of an outcome prediction. Business redefinition did occur. It occurred inside the existing business, not in the declared territory. That is the central finding of this chapter. Organization — the vessel was replaced; the function is not visible from outside Structural change in the organization dimension is confirmable from public information. On June 30, 2025, Zuckerberg announced the creation of Meta Superintelligence Labs in an internal memo. The same month, the company was reported to have invested $14.3 billion in Scale AI for a 49% stake, bringing that company’s founder in as Chief AI Officer. The core of the research organization was recomposed through external acquisition. Headcount is shrinking. Employees numbered 78,865 at the end of December 2025 and 75,472 at the end of June 2026, down 1% year over year (official financial disclosures). The total is falling while the composition is being swapped. The organization as the canon defines it is not a set of people. It is a value-creation system made of people, AI, partners, universities, and customers. But the replacement of a vessel and a change of function are different things. The second is not observable from outside. Capital — the most eloquent dimension The capital dimension says the most in this case. Capital expenditure was $72.22 billion for full-year 2025 (including principal payments on finance leases). Guidance for full-year 2026 was initially $115–135 billion, raised to $130–145 billion at the second-quarter announcement. Capital expenditure in the second quarter of 2026 alone was $31.08 billion. Free cash flow in the same quarter was $784 million, sharply down from $8.55 billion a year earlier. The operating margin fell from 43% to 31%, and net income was $15.848 billion, down 14% year over year. Capital has clearly moved. But where it moved is not the metaverse. It is the computing substrate for AI. On the January 2026 call, Zuckerberg said Reality Labs losses in 2026 would be roughly in line with the prior year and that this would probably be the peak. The shift in the center of gravity showed up in the numbers before it showed up in the words. The canon’s third equation defines capital as follows. Future Capital = Financial × Human × Learning × Trust × AI × Knowledge × Ecosystem × Purpose This is multiplication. If any single term is zero, the whole product is zero. The company has made the Financial and AI terms extremely thick. Meanwhile the Trust term is under continuing examination in a regulatory and litigation environment. Because it is multiplication, a thick term cannot compensate for a thin one. Leadership — the speed of declaring and the speed of verifying What is observable in the leadership dimension is a sequence of declarations. This company’s leadership works by placing a picture of the future first and pulling resources toward it. The canon’s fourth equation defines leadership as follows. Leadership = Purpose × Question Design × Capital Allocation × System Architecture × Trust This too is multiplication; a zero anywhere zeroes the product. What stands out here is the Capital Allocation term. The company can execute allocations in the tens of billions of dollars while its short-term financial indicators deteriorate. That property is rare among listed companies. The Question Design term is where the problem remains. The question to ask was not “when do we build the metaverse” but “who holds the layer that mediates human connection next?” The first fixes the answer to a means. Under the second, devices and AI both become candidate answers. How the question was framed decided five years of capital allocation.

4 Structure — maturity and the value chain

4.1 Where it sits on the Enterprise Redefinition Maturity Model

Within what public information supports, maturity varies by dimension. It cannot be placed at a single level. In the business dimension, continuous improvement is running inside the existing business. But the declared new territory has barely moved the revenue mix. In the source paper’s words, the behavior still resembles the description of an Improvement Enterprise: increasingly efficient while remaining fundamentally unchanged. In the capital dimension, higher-level behavior is visible. Dynamic capital allocation is a characteristic of a Continuous Redefinition Enterprise. The company has moved the center of gravity of its allocation twice in five years. The source paper’s diagnostic question asks whether resources are allocated toward Future Value rather than historical success. That question can be answered yes, at least in part. The purpose dimension splits. The source paper’s diagnostic question is this: does the organization periodically re-examine its Core Purpose and adapt its expression without unnecessarily weakening organizational identity? The company did re-examine. Whether replacing the identifier itself left identity unweakened cannot be asserted. In the organization and leadership dimensions the change of vessel is visible; the quality of the decisions is not. Recall the canon’s notes here. The Enterprise Redefinition Maturity Model (ERMM) evaluates organizational coherence rather than isolated excellence. An organization may hold high AI capability while other dimensions fail to keep up. Maturity is also assessed across all five dimensions in balance: exceptional technological capability with weak leadership redesign does not produce higher maturity, and strong purpose without adaptive organizational systems remains insufficient. And Level 5 must not be treated as a target to reach as fast as possible. The appropriate level differs by industry and by environment.

4.2 Where in the Future Value Chain the value arose

The canon fixes the causal order as follows. Purpose → Learning → Redefinition → Creation → Enterprise Value Lay five years over that order and the distance between declaration and reality appears as a structure. On the declared path, the traces read as an attempt to jump from Purpose straight to Creation. The name was changed, capital was committed, products shipped. But the Learning and Redefinition stages are thin. The process of learning what the market wanted and rewriting the definition of the business from there cannot be adequately confirmed in public information. A path that skipped stages did not reach Enterprise Value. On the actual path, the order was kept. The company learned from user behavior and advertiser outcomes, rewrote the content of the advertising business, and built new offerings from there. Learning came first. Two paths ran in parallel inside the same company. The one that showed up in the financials is the one that was not declared. The second equation shows the same structure. FVCC = Purpose × Learning × Redefinition × AI Integration × Ecosystem × Capital Allocation × Trust Seven terms, multiplied. On the declared path, Capital Allocation was large. But if Learning and Redefinition are small, the product is small. The size of the capital does not make up for the smallness of the other terms. That is what multiplication means.

4.3 A declaration does not buy time

The fifth equation places time as an independent term. Future Value = Future Time × Future Capability This is multiplication as well. A name change does not raise Future Capability overnight. Nor can it buy Future Time. What a declaration changes is only external expectation and the legitimacy of a resource allocation. The Value Equation shows the same structure. Value = Purpose × Trust × Capability × Time Trust and Time enter as terms in the product. Repeat declarations and the Trust term can wear down. That a second large declaration is received more warily than the first follows from this structure. The eighth of the First Principles states, Trust Compounds Faster Than Capital. Trust compounds faster than capital. What compounds fast also falls fast.

5 What it looks like in practice — the turning points, and

what was paid at each Redefinition is not the act of adding something. It is the act of deciding what to keep and what to let go. We set out the turning points together with what was paid. Turning point 1 — October 2021, the name What was given up was continuity of the identifier. A company name is also a vessel for trust. Replacing the vessel liquidates part of what has accumulated in it. Every subsequent correction would be exposed to public verification. Turning point 2 — from 2023, putting AI inside the existing business Behind the declaration, the substrate of recommendation and advertising was replaced. What was given up was the weight of human-designed operational tuning. The shift was not widely reported. But it is what supports the 2026 financials. Turning point 3 — June 2025, recomposing the research organization The creation of Meta Superintelligence Labs, and the $14.3 billion investment in Scale AI. What was given up was continuity in the existing AI research organization. Bringing core talent in from outside carries a judgment about what had accumulated inside. Turning point 4 — July 2025, personal superintelligence On July 30, 2025, the company published “Personal Superintelligence for Everyone” by Zuckerberg. The orientation is to put the power in individual hands. Around the same time, several outlets reported him saying, in substance, that not everything would be released. Turning point 5 — April 2026, Muse Spark The official announcement recorded provision in limited preview and a hope to open-source future versions. What was given up here was consistency in the open-weight position. What was the open-weight strategy, in terms of value capture? This is one of the core issues of the chapter. Release the weights of a model and the ecosystem thickens. Developers, researchers, and adjacent tools gather. But release itself captures no value. Value is captured on the side of the complements. This company’s complements are clear: advertising inventory and devices. So while it was not leading at the frontier, release was rational. For the pursuer, release is a way of thinning the leader’s advantage. Conversely, once the company judged it could lead, non-release became rational. The observable facts of 2026 are consistent with that switch. The general rule to draw is clear. Openness is a means, not a Purpose. Means can be withdrawn. But the cost of withdrawal is paid in trust, not in the financials. Developers and companies that invested on the assumption of openness receive the change of assumption as a burden. That is why the multiplicative formulas carry a Trust term. The risks this structure carries Celebration is the enemy of analysis. So is condemnation. We organize the risks visible in public information into four. First, concentration of revenue. Dependence on a single source transmits regulatory shifts and advertising-market swings straight into results. The company has itself disclosed in its annual reports that substantially all of its revenue comes from advertising. Second, capital brought forward. Capital expenditure guidance for 2026 was $130–145 billion, and total expenses $165–169 billion. With the timing of the return invisible from outside, only the spending is fixed in advance. Third, regulation and litigation. On April 23, 2025, the European Commission fined the company €200 million under the Digital Markets Act over its so-called “consent or pay” design. Criticism over privacy, content regulation, and effects on young people continues in many countries. On March 25, 2026, a jury in a US state court was reported to have found the company and another operator liable in litigation over effects on young people, with both indicating an intention to appeal. On the second-quarter 2026 earnings call the company itself referred to a $2.4 billion charge related to legal proceedings, and to the possibility that a youth-related trial within the year could produce a material loss. These matters are in dispute, and we assert nothing about their merits. What can be pointed out is only that the Trust term is under continuing examination. Fourth, the wearing down of declarations. The market and society remember the distance between the first declaration and reality. The second declaration is received through that memory. It is a question of the accounting of trust. All of these concern the future and none can be asserted. What can be said is that the strength and the fragility come from the same place. A single enormous revenue source makes an enormous bet possible and, at the same time, delays its verification.

6 What transfers, and questions for the executive

There are four things to take from this case. First, the rewriting of Purpose cannot be measured by declarations. The place to measure is the capital allocation table. Put the other way: if allocation moves, redefinition is under way even when the declaration is modest. Whether your own Purpose refresh is real can be judged by where it appears in next year’s budget. Second, do not mistake the collateral for a declaration. Do not stake the name of your existence on a bet about means. When the collateral is excessive, correction looks like defeat. Most delays in the decision to withdraw are matters of face, not economics. The phrase “betting the company” often builds the same structure. Third, redefinition can happen inside the existing business. This company’s substantive results came not in the territory it created but inside the old business of advertising. Incumbents tend to treat existing businesses as conservative ground. But the contact with customers and the accumulation of learning exist nowhere else. Fourth, do not misread the character of a decision to open. Releasing technology or data is easily spoken of as a statement of Purpose. In most cases it is a means belonging to the design of value capture. Whether you explain the possibility of withdrawal to stakeholders in advance changes how trust erodes. And one caution. The appropriate level on the Enterprise Redefinition Maturity Model differs by industry and by environment. Making the fastest possible arrival at Level 5 the objective is something the source papers explicitly warn against. Three questions to close. Each can be answered at your next executive meeting. Question 1 — On which line of next year’s capital allocation table does your new Purpose appear? If no line can be found, it is a declaration and not a redefinition. The question is not the size of the amount but whether the line exists. Question 2 — Is what you have decided to open a Purpose, or a means? If it is a means, the conditions for withdrawal have to be defined in advance. A withdrawal without a definition is received as a betrayal. Question 3 — How many years does your company allow between declaring and verifying? A declaration with no deadline becomes an institution without ever being tested. Once it is an institution, nobody can stop it. This returns to the question of Recognize, the first stage of Enterprise Redefinition: what assumptions about our enterprise are becoming obsolete? AI does not answer it. AI Optimizes. Humans Define. AI optimizes; humans define value, purpose, and direction. What Meta redefined is neither the metaverse nor the company name. It is the content of the old business called advertising. And something remains un-redefined: the place where the revenue is generated. The declaration was made in the Purpose layer; the result arrived in the Business layer. Five years of distance remain between the two layers. The lesson of this case fits in one line. Redefinition is not declaration. A declaration is only a starting point, and without passing through Learning and Redefinition it reaches neither Creation nor Enterprise Value. A company that skips the order sends only its name into the future, and leaves the business in the present.

In brief

What Meta redefined is neither the company name nor the metaverse but the content of the existing advertising business.
Within what public information supports, capital resembles a Continuous Redefinition Enterprise while the business still behaves like an Improvement Enterprise.
The value arose inside the existing business, on a path that passed through Learning and then Redefinition in order.
If the distance between declaration and reality persists, the Trust term wears down and the next declaration stops working.

Key concepts

Future Value Chain / Future Value Creation Capability / Future Capital / the Enterprise Redefinition Maturity Model / the Medium Pattern (→ Vol. VI, Ch. 059)

The chain of ideas

Purpose → Capital Allocation → Learning → Redefinition → Enterprise Value

Related first principles

Principle 1 — Purpose Precedes Profit. Principle 4 — AI Optimizes. Humans Define. Principle 6 — Enterprise Exists to Redefine Itself. Principle 8 — Trust Compounds Faster Than Capital.

Related chapters

Vol. VI, Ch. 059 “Cases of Enterprise Redefinition” — the definition of the Medium Pattern, and the order to follow when patterns overlap
Vol. V, Ch. 049 “What Does It Mean to Redefine a Brand?” — treats the weight of a company name as collateral
Vol. III, Ch. 028 “Does Purpose Change Enterprise Value?” — measuring the distance between declaration and capital allocation
Vol. V, Ch. 044 “What Is the Enterprise Redefinition Maturity Model (ERMM)?” — how to read levels that split by dimension

Papers and companion volumes

Kadowaki, N. (2026). Future Value Theory: A Management Framework for Enterprise, Capital, and Society in the Age of AI. VURA Working stract=7120980 / Paper. Zenodo: SSRN: https://ssrn.com/abhttps://doi.org/10.5281/zenodo. 21255662
Kadowaki, N. (2026). Enterprise Redefinition: Toward an Enterprise Evolution Theory for the Age of AI. VURA Working Paper. (Published on Zenodo; under review at SSRN)
100 Questions on Management in the Age of AI, #026 “Is Purpose Necessary in the Age of AI?”

Chapter 088 What Did Meta Redefine?