Software creation
The cost of creating a digital interface may fall sharply, while the cost of operating it reliably, securely and within regulation does not fall at the same pace.
The Portfolio at a Glance
The following overview summarizes the strategic building blocks, position directions, and allocated assets for this research theme.
Long
Small-business operating platforms
These companies own business relationships that include domains, hosting, commerce, checkout and payment history. The case rests on commercial attach, not on a proprietary model claim.
Long
Payment authorisation and settlement
Payments infrastructure can benefit if automated commerce requires more authentication, tokenisation and dispute handling rather than bypassing established rails.
Long
Electricity and connection
Cable and copper express the physical requirements of more connected computation, although both remain exposed to the industrial cycle.
Long
Conformity and verification
This is an option on durable regulatory evidence for systems that affect personal data, money or safety. It should not be valued as a software business.
Long
Identity and operational evidence
Machine identity and observability matter operationally, but the revenue opportunity remains unproven until distinct pricing and durable usage are disclosed.
Short
Offshoring to low-wage countries and generic software risks
These are paired or defined-risk expressions of AI substituting routine digital work. Low valuations, borrow costs and takeover risk require modest sizing.
The analysis
The investable consequence of cheaper software creation is often misunderstood. It is more likely to turn much of the application layer into a feature of a few established platforms.
The relevant distinction is between generated software and an operating relationship. A generated landing page, workflow or internal tool can be replaced. The domain, merchant account, customer history, payment credentials and regulated record behind it cannot be replaced without cost, disruption and liability. This favours companies that already sit in those relationships.
The framework also separates a real economic effect from a tradeable one. Machine identity, conformity assessment and payments triggered by AI agents may create large volumes before they create large standalone revenue pools. Investors should look for disclosed pricing, transaction volume and recurring revenue rather than assume that activity alone produces a new listed winner.
The investment argument
Generated software is not necessarily an ownable asset
Software that is produced for a specific task and discarded afterwards has little reason to become a durable product, marketplace listing or fee-bearing ecosystem. Its economic value is more likely to be captured by the platform that authenticates the user, hosts the state, processes the payment or supplies the electricity.
Builder platforms can retain value where they become the operational environment itself. The burden of proof is high: it must demonstrate corporate retention, recurring usage and a cost advantage that is not quickly available to competing platforms.
Small-business platforms may benefit from free creation
Website and commerce platforms are usually described as vulnerable to AI-generated sites. That view overlooks their more durable functions. Businesses pay them for a domain, hosting, billing, checkout, payment acceptance and operating history. Lower-cost creation can increase the number of sites and stores that enter this commercial system.
The important monitoring figures are transaction volume, commerce attach, bookings and average revenue per customer. In-house model claims are secondary. A platform is validated by evidence that more customers use and monetise its operational services.
Agents require trusted payment and permission layers
Even with automated purchasing, the question remains who was allowed to act, what recourse applies and who bears a loss if the action was not authorised. The first transactions may be small and contained within closed platforms, which keeps their scale modest for now. Established authorisation and dispute resolution remain necessary.
Standalone identity vendors deserve greater caution. Industry surveys indicate that machine identities can substantially outnumber human identities, but cloud providers may bundle agent governance into existing licences. A new revenue opportunity exists only where vendors publish a distinct pricing unit and customers accept it.
The final constraint is physical
When software creation becomes cheaper, the scarce inputs are less likely to be code and more likely to be the authority, record and electricity needed to put generated software to work.
AI compute still has to run on connected capacity. Grid-connection queues, transmission projects, cable lead times, municipal approvals and cost-allocation rules determine where that capacity can be built. These constraints are local, slow to change and economically meaningful even if application software itself becomes abundant.
This makes grid equipment and electricity infrastructure a longer-duration expression of the theme than the first generation of AI builders. They remain cyclical investments, however, and a financing shock to data-centre construction could affect them before the physical shortage is resolved.
The value chain
| Rank | Layer | Rationale |
|---|---|---|
| 1 | Customer base and business data | Domains, customer records, checkout and payment history are embedded in a business relationship and are expensive to move or rebuild. |
| 2 | Payment authorisation | Tokenisation, fraud controls, reversibility and dispute rights remain necessary when an automated system moves money. |
| 3 | Record and conformity evidence | Activities involving regulated data or products can require durable documentation, testing and proof of compliance. |
| 4 | Electricity and interconnection | Grid connections, transmission equipment and power supply remain physical inputs with long development cycles. |
| 5 | Builder and application layer | The layer may grow rapidly, but its returns depend on retention, distribution and a durable operational role rather than on code generation alone. |
The weakest position is a business whose only asset is a generated interface, a marketplace fee on temporary applications or a per-seat feature that customers can recreate locally.
The likely setting
In the early 2030s, generated software may be plentiful while the ability to operate it remains constrained. Local power supply, permission to act, durable records and established payment relationships become more important than the code used to create an interface.
That setting supports a limited number of regional small-business operating platforms, incumbent payment rails and service providers that translate regulation into testing and evidence. Higher valuations do not follow automatically. Most returns should come from improving usage, cash flow and resilience rather than from a broad AI multiple expansion.
The Investment Selection
The selection groups listed companies by the economic function they perform. Company ratings on the individual cards come from Leeway’s general equity analysis and are independent of this theme.
The Business-Rating scores the quality of the business model, independent of the share price. The Market-Fit-Rating tests eighteen fundamental figures for how well the company currently fits the market; a negative reading implies expected negative performance. The Cycle-Rating places the valuation in the stock’s own history: a higher figure means the shares are historically cheaper. The Leeway-Score combines the three in equal parts. How the ratings are calculated
The recommendations since publication
From 4 September 2026 to 19 October 2026: Portfolio-0.3%ACWI-2.3%
Small-business operating platforms
LongThese companies own business relationships that include domains, hosting, commerce, checkout and payment history. The case rests on commercial attach, not on a proprietary model claim.
Role in the thesis: Small-business operating platform with hosting, commerce, checkout and payments.
Investment case: The central case is that lower-cost site creation expands the population of businesses that can use Wix’s recurring operating services.
Invalidation risk: Payments attach and bookings must improve. If generation moves into browsers or operating systems without increasing Wix’s commercial relationship, the thesis weakens.
Role in the thesis: Domain and small-business operating platform with substantial recurring customer relationships.
Investment case: GoDaddy’s domain base, billing relationship and commerce services can benefit if cheaper creation increases the number of businesses that become active online.
Invalidation risk: The domains base could erode, and commerce attach may not be sufficient to offset slower legacy growth.
NAVER Corporation
Long
Role in the thesis: Korean search, commerce and payments platform with a domestic operating relationship.
Investment case: Naver offers a regional version of the thesis on customer base and business data through merchant participation, commerce activity and payment volume.
Invalidation risk: Competition in Korean commerce and platform regulation could outweigh improvements in merchant and payment activity.
Your knowledge edge
Understand the real drivers of our portfolio
Look behind the scenes: see not only what we buy, but why. Get the complete view of our strategy.
- Every position revealed: Access each holding in our portfolio.
- Understand the investment thesis: See the exact role each security plays in our return.
- See the connections: Analyse our long and short strategies in direct context.
Timeline and review points
| Window | What happens | What it means for the portfolio |
|---|---|---|
| September to December 2026 | Wix and GoDaddy report bookings, payment activity and commerce attach, while infrastructure companies update their order outlook. | Commercial activity at operating platforms matters more than AI feature announcements. Weak attach would keep the core long thesis unconfirmed. |
| 2027 | The first audited filing by a major AI builder, pricing decisions for AI agents and enterprise software renewals provide better evidence on retention and monetisation. | Corporate retention and a distinct agent price would strengthen the application and identity cases. Bundling or weak cohorts would favour incumbents over standalone tools. |
| 2028 to 2029 | Power constraints, interconnection rules and financing conditions determine the pace and location of compute expansion. | Grid and cable demand can remain resilient, but a financing event would create a broad drawdown before relative returns become visible. |
| 2030 to 2034 | Higher-value automated commerce and the regulatory treatment of autonomous decisions become clearer. | Established payment and authorisation layers benefit only if the volume reaches regulated transactions rather than remaining inside closed, low-value loops. |
Three Possible Outcomes
| Path | Weight | What happens |
|---|---|---|
| Favourable: operating relationships deepen | 25% | Small-business platforms show accelerating commerce attach and payment volume, regulated evidence becomes a visible service line, and automated commerce uses established authorisation rails. Grid investment extends without a financing disruption, while earnings from offshoring to low-wage countries weaken structurally. |
| Base: gradual divergence | 50% | Software creation becomes cheaper, but application economics remain mixed. Operating platforms improve slowly, commerce triggered by AI agents stays low-value and contained, and identity receives attention before sustainable revenue. The theme works through relative earnings and selection rather than a broad re-rating. |
| Adverse: broad de-rating, everything falls together | 25% | A credit or power shock delays compute investment, and application generation becomes embedded in browsers or operating systems. Small-business platforms retain their customer base and business data but do not gain meaningful attach, while regulation is softened and the conformity option fails to develop. |
Counter-arguments and Risks
The strongest counter-case is that specialised builders retain a compounding advantage. A large proprietary dataset of real build, test and repair traces could make their models materially better and cheaper for a narrow task. In that outcome, a few builders become the operating environment rather than a feature inside another platform.
Public evidence is not yet sufficient to settle this. The decisive indicators are audited revenue quality, retention by customer cohort, independent performance comparisons and the share of work performed by proprietary rather than general models. Until those are available, private builder valuations should be treated with caution rather than extrapolated into listed-market conclusions.
Unresolved Market Factors
The first open question is whether regulation forces generated software to leave a durable record. Requirements affecting AI systems, cyber security and operational resilience can create documentation and assessment work, but implementation and enforcement remain politically contested.
The second question is whether agent authorisation is a priced product or a bundled platform feature. Cross-system delegated authority is a genuine operational problem, yet a standalone revenue pool appears only if customers pay separately rather than receive it within cloud or productivity licences.
What investors should monitor
- Wix transaction revenue, payment volume, bookings and commerce attach relative to subscription growth.
- GoDaddy applications-and-commerce bookings, customer monetisation and retention in the domains base.
- Naver Pay volume, commerce activity and the number of active merchants.
- Published pricing and revenue disclosure for machine identities or independently authorised agents.
- Payment-network and processor disclosures on tokenised, delegated or agent-initiated transaction activity.
- The revenue mix, customer retention and gross-margin evidence in the first audited filing by a major AI builder.
- EU implementation of AI, cyber-security and operational-resilience obligations, including any independent conformity designations.
- Interconnection queues, cable lead times, transformer availability and grid cost-allocation decisions.
- Credit spreads and financing structures for compute capacity and data-centre projects.
- Headcount, utilisation and revenue per employee at IT services and staffing companies.
What would disprove the thesis
The analysis should be revised if these conditions persist.
- Small-business platforms fail to increase payment activity, commerce attach or customer monetisation despite much cheaper site creation.
- Independent evidence shows that specialised builders retain a durable performance and cost advantage that established platforms cannot reproduce or distribute.
- Power supply expands quickly enough to remove interconnection scarcity while regulatory evidence and machine identity are broadly bundled without incremental revenue.
How this analysis is produced
The assessment is produced in several steps. Independent model families answer the same question separately and then attack the results. What you read here has survived several rounds.
- Two independent first theses. The same opening question goes to several model families that cannot see one another. Disagreements are kept, not averaged away.
- Dated evidence. Every claim that depends on facts is broken into individual search questions and answered with dated, sourced web research. Question, answer, sources and timestamp are logged and remain traceable.
- Adversarial review. Several review roles attack the thesis from different angles: one hunts for the strongest refutation, one for the awkward edge cases, one tests whether a path from thesis to share price exists at all, one checks the timeline for contradictions. Each role raises its own questions, which are again answered with evidence.
- Merge, then the next round. The surviving theses are merged into one and attacked again. The counter-position and the unresolved tension on this page come out of that step. They were not bolted on afterwards to look balanced.
- Back to the start. The process runs again until there is a clear result and a list of tradable companies with structural advantages.
Any analysis can be wrong. That is why the falsification criteria and the counter-position sit on the same page as the thesis, not in the small print.
The numbers shown against individual companies do not come from this process. The Business-Rating scores business-model quality, the Market-Fit-Rating eighteen fundamental figures against the current market, the Cycle-Rating the valuation against the stock’s own history. They are documented under the Leeway scores.
Frequently asked questions
Does this analysis predict that all AI builder companies will fail?
No. It distinguishes activity from durable ownership. A builder can succeed if it becomes an operational environment with recurring corporate usage, strong retention and a defensible distribution advantage.
Why are small-business platforms included as long ideas?
Their durable assets are not only website templates. They include domains, hosting, customer relationships, checkout, payment acceptance and operating history, which are costly for a business to replace.
Why are payment networks relevant to automated commerce?
An automated transaction still needs credentials, authorisation, fraud controls, reversibility and a framework for disputes. Existing payment infrastructure already performs these functions at scale.
What is the central risk to the portfolio?
A financing or power shock could cause a broad decline across infrastructure and technology exposures before the intended divergence appears. The short positions do not fully remove that correlation risk.
Review the analysed companies with the three Leeway ratings
The Business Rating measures the quality of a company’s business model, the Market-Fit Rating assesses its fundamentals in the current market regime, and the Cycle Rating places its valuation in the history of the stock. Review each company with Leeway’s general equity analysis.
Company Valuation and Fundamental Analysis
The data is recalculated on a weekly basis and depends on the current market value of the company and the balance sheet figures of the annual financial statements. The market value changes continuously with price changes, the balance sheets are created annually and change the valuation massively. The time of the annual financial statements and the metrics used can be viewed under "Metrics". Further information on how the analyses work can be found as tooltips directly on the analyses as well as in our explanations.
General
The information is intended to support you in your independent decision-making in implementing investment strategies and serves purely informational purposes. Past performance is not a reliable indication of future performance. No warranty can be given for the completeness, correctness and accuracy of the listed content. The information does not constitute specific investment recommendations. We neither know you nor your financial situation and do not provide investment advice. Only licensed investment advisors with knowledge of your personal circumstances may do this. PWP Leeway UG (limited liability) is not an investment advisor and does not collect any personal data for the purpose of investment optimization. PWP Leeway UG (limited liability) is a provider of investment recommendations and investment strategy recommendations. As such, it is registered with and supervised by the Federal Financial Supervisory Authority (BaFin). The legal notices on the use of the website and the General Terms and Conditions of PWP Leeway UG (limited liability) also apply.
All price data are closing prices of the respective stock exchanges. Price information and master data are provided by an external service provider. Furthermore, public trading data, such as provided by Finra.org, is used to analyze market behavior. No warranty can be given for the completeness, correctness and accuracy of the listed content.
Disclosure of interests: To the best of our knowledge, Leeway and the persons responsible for creating or publishing the respective content do not hold a net long or net short position exceeding 0.5% of the issued share capital of any issuer discussed in that content at the time of publication. There are no other interests or business relationships that could conflict with objectivity, unless expressly stated otherwise in the respective content.
