From commoditization, disintermediation or bundling, how AI is accelerated market shifts

The past couple of months, working with several founders, one thing became very clear: I have never seen software categories evolve so quickly. And I’m not only speaking about product and tech changes, but also about market shifts.

Some layers that looked established for years now feel like they’ll disappear soon. Some categories are splitting while others are emerging suddenly. In some categories, in less than 18 months I’ve already seen the second or third wave of new entrants (wtf!).

When the cost of building drops, when distribution can become highly viral (just look at how fast new AI tools can rise and fall in popularity), and when the tools to build software keep evolving in such rapid cycles (Cursor, Claude Code, OpenAI Codex…), market structures also change at a pace I have never seen before.

In the past two years, I have probably seen existing software categories evolve more than I did in the previous ten. From 2010 to 2020, working both on the investor and operation sides, I could see change happening, but it felt more stable, moving in directions we were expecting and at a pace that did not catch me off guard. The past two years have compressed that entire experience.

What I found interesting when I started modeling these market shifts, is that AI does to software categories what it does to technology itself: It accelerates them. AI does not necessarily create entirely new market dynamics. It compresses and intensifies the ones that have always existed, mainly:

• Commoditization
• Fragmentation
• Bundling
• Disintermediation
• Regulation & Tech driven shifts
• Verticalization

In the rest of the post I will concisely describe each of these market dynamics and how AI is accelerating them in many software categories.

Movement to look in your software category and how AI is accelerating them

Commoditization

Description:

• Commoditization happens when a product, feature, or entire layer that once carried clear differentiation becomes interchangeable. Value shifts away from “what you sell” toward price, distribution, brand, or scale.
• Before AI this rarely happened overnight. It usually creeps in while teams are still shipping improvements that no longer change buying decisions.

Typical signs:

• Messaging homogenization Competitors’ messaging starts to sound the same, with similar feature lists and promises.
• Pricing pressure increases. Discounts become normal, freemium tiers expand, and buyers push harder on price without asking deeper product questions.
• Innovation turns incremental / everyone ships the same things. New launches feel incremental and are quickly copied. Roadmaps converge across players.
• Buyers stop caring about how it works and only care that it works. Since many products are similar, what makes the difference for buyers is increasingly price and brand.

How AI is accelerating this market dynamic:

• Commoditization is accelerating in many software categories simply because LLMs make it much easier to copy or develop new features and also easier to copy the messaging and marketing tactics of others.

What can be done:

• Move up or down the value chain. A first option is to expand to an adjacent layer where differentiation still exists (For example orchestration, analytics, compliance, or outcomes).
• Product verticalization. Another possibility is to narrow the scope to a specific segment or use case where the problem is still painful and poorly served and verticalize your solution.
• Stop building features and focus on distribution. Brand, ecosystem, data, or switching costs often matter more than product depth at this stage.

Fragmentation

Description:

• Fragmentation happens when a market or a layer attracts a growing number of small, focused players, each addressing a narrow slice of the problem. Instead of a few strong competitors, value gets split across many specialized solutions.
• Fragmentation usually appears when barriers to entry drop or when a new opportunity becomes visible before a dominant model emerges. Innovation looks high, but power is diluted.
• A good example of such a market is the MarTech one.

Typical signs:

• Explosion of logo density. Over time many new players appear in the same layer, often with similar positioning but slight variations.
• Many small players appear and many die. Many bootstrapped startups are launched and often disappear quickly.
• Customers assemble stacks of software. Buyers start combining multiple tools instead of choosing a single platform.
• Plenty of subcategories emerge in a major software category. New subcategories keep appearing and overlapping with previous ones.

How AI is accelerating this market dynamic:

• Fragmentation is accelerating in some software categories because AI dramatically lowers the cost and time required to launch a narrow product. Small teams can now build and ship highly specialized tools in weeks rather than months.
• AI also makes it easier to target micro use cases with tailored positioning and messaging, which increases logo density inside already crowded layers.

What can be done:

• Start bundling. When customers are forced to assemble their own stack, bundling can become a way to remove friction. This can mean packaging adjacent tools or expanding your product (all-in-one software).
• Orchestrate. Another possibility is to position yourself as the “glue” between tools. Orchestration focuses on workflows, integrations, data flows, and decision logic, letting fragmented tools coexist.
• Become an acquisition target. Fragmented markets often consolidate at one point. A strong brand, a great product, and a clean cap table can make you attractive to larger players looking to assemble a broader offering quickly.
• Focus on branding. In noisy markets, perception matters. Clear positioning and strong branding often make the difference.

Bundling

Description:

• Bundling happens when previously independent products or layers start collapsing into broader, integrated offerings. What used to be a best of breed stack becomes a single product, suite, or platform.
• Bundling is usually driven by customer fatigue. As stacks grow more complex, buyers start valuing convenience, integration, and accountability over depth in any single feature.
• A good example of that is the customer service software market where complete suites emerged (like Intercom).

Signs:

• From specialized software to “swiss army knives”. Competitors that once started as specialized tools become all-in-one software by adding plenty of features over time.
• The rise of the software suites. Product announcements increasingly focus on new modules, suites, or end-to-end capabilities rather than depth in one feature.
• Customers start to privilege platforms. Customers talk less about tools and more about platforms. They actively reduce the number of tools they use, even if these individual tools are better (convenience vs performance).

How AI is accelerating this market dynamic:

• AI accelerates fragmentation by enabling small teams to ship narrow products quickly, but it also accelerates bundling (I know it sounds contradictory). The same technology that makes it easy to launch a focused tool also makes it easier for larger players to add new features and extend their scope.
• Instead of building entire modules from scratch, companies can layer AI capabilities on top of existing products, expanding horizontally at a much faster pace.

What can be done

• Bundle deliberately. If bundling is happening, either commit to broadening your offer or stay sharply focused on a niche or a specific customer segment.
• Differentiate on depth/focus. When others bundle broadly, there is sometimes still room for specialists that go significantly deeper (verticalization).
• Prepare for pricing pressure. Bundles often reset willingness to pay. Anticipate this by adjusting packaging, usage models, or value metrics before the market forces it on you.

Disintermediation

Description:

• Disintermediation happens when a player higher up the stack finds a way to connect directly to the bottom layer (or vice versa), bypassing one or more intermediate players entirely. Basically bypassing middlemen.
• This move is often enabled by scale, data, regulation, or technology. What used to require several specialized layers suddenly becomes “native” to a dominant player.
• A good example of that is the eCommerce software category where a platform like Shopify destroyed many middlemen (they even “own” payment now).

Signs:

• Upper layer players expand downward. Platforms start building native capabilities that replace entire categories below them.
• APIs and middleware dry up. Intermediaries decline in importance.
• Customer stack simplify. Buyers skip software that used to be mandatory (and often do not miss them).

How AI is accelerating this market dynamic:

• In some categories, middleware and internal workflow tools are increasingly exposed because they basically exist as a middle layer between companies and developer tools.
• With AI coding assistants and automation agents, employees can now build simple integrations, scripts, and internal automations themselves without relying on specialized vendors.

What can be done:

• Become infrastructure, not middleware. If you cannot win the end user interface, go deeper. Infrastructure that is hard to replicate, regulated, or deeply embedded is harder to bypass.
• Pivot or exit the category. Most disintermediation waves are structural and irreversible. When a dominant player absorbs your role and you have no durable leverage left, the rational move is not to fight but to pivot. It’s hard, but it’s often the best solution…

Regulation disruption / change

Description:

• Regulatory disruption happens when new laws, standards, or enforcement regimes change how a market works.
• Regulation can sometimes enable new models to emerge and sometimes kill existing ones.
• A first good example is GDPR and how it made data protection and consent a core product constraint, weakening data-extraction business models while favoring platforms and vendors that could turn compliance into trust and differentiation.
• Or recent EU green regulations such as CSRD and parts of the Green Deal that forced companies to measure and report environmental impact, but are now being slowed, softened, or selectively rolled back as economic pressure and competitiveness concerns take priority (which is hurting many companies born out of this regulation btw).

Signs:

• The compliance topic enters sales conversations. Buyers start asking about certifications, audits, data residency, or legal guarantees.
• New software categories appear in the market. Entirely new blocks emerge around compliance, reporting, or risk management.
• Often incumbents gain momentum. Larger players move faster than expected because regulation reinforces their scale, legal teams, and balance sheets.

How AI is accelerating this market dynamic:

• AI has accelerated the concerns around data privacy, data residency, model training, and intellectual property.
• Anecdotally, here in Europe I hear more and more software users explicitly asking where their data goes, whether it is used to train models, and whether AI features are compliant with local regulations. Data sovereignty is becoming a real thing for customers. And it has implications both from a sales but also a product pov (where do I host my data).
• This increases compliance pressure across entire categories and can quickly reshape competitive dynamics, often favoring players that can offer clear guarantees and strong governance.

What can be done:

• Turn compliance into a feature. Compliance itself can become a differentiation. Make the regulated path the easiest path for customers.
• Target the newly constrained buyers. Regulation often creates new decision makers and new pain points. Selling to legal, compliance, or risk teams can open doors that did not exist before.
• Exit or pivot. If regulation structurally destroys your business model, speed matters. The earlier you pivot, the more optionality you preserve.

Technological change

Description

• Technological change happens when a new technology alters the structure of products, not just their performance. It changes how software is built, how it is used, and how value is priced.
• Good examples of such disruption are the recent AI wave or the emergence of cloud computing in the 2000s.

Signs:

• New product categories emerge inside existing markets. Technology shifts often give birth to categories that could not exist before, such as mobile apps enabled by smartphones or cloud native software enabled by SaaS.
• The way users interact with products suddenly changes. Usually interfaces get lighter, workflows shrink, and human involvement decreases. This is the overall direction that software has taken the past 50 years.
• Often, economic assumptions break. Real technology disruptions often change business models and impact the unit economics. Like the transition from on-premise licenses to the subscription one. Or more recently how AI is impacting software margin.

How AI is accelerating this market dynamic:

• In the past eighteen months alone, we are already entering a second product paradigm wave. Eighteen months ago, most companies were focused on launching AI assistants that helped users work faster inside existing workflows.
• Today, the shift is toward more autonomous agents and workflow level automation, where the product does not just assist but executes.

What can be done:

• Identify what is being removed. Every major tech shift eliminates something. Installation, configuration, manual steps, training, or decision making. Map what disappears first.
• Redesign the product around the new constraint. Do not just add features. Change workflows, defaults, and assumptions to match the new technology.
• Or start a new product from scratch. This is the approach Intercom took with its Fin AI agent, letting the core business run while dedicating a separate team to rebuilding a product designed for today’s agent driven interaction model rather than retrofitting the old one.

Verticalization

Description:

• Verticalization usually happens when a large horizontal market gives birth to industry specific products that are more tightly integrated and better adapted to specific use cases or custom segments. Instead of one horizontal tool serving everyone, multiple vertical plays emerge, each optimized for a narrow context.
• This usually occurs once the horizontal layer is mature enough that “generic” is no longer good enough.
• A good example is how Salesforce gave birth to plenty of verticalized CRM players.

Signs:

• New entrants lead with industry specific products. These products serve directly specific use cases, regulation, or workflow rather than abstract capabilities.
• Vertical products integrate deeper into their customer’s stack. Vertical tools embed billing, compliance, reporting, or domain specific logic that horizontals can sometimes struggle to abstract.
• Sales cycles change. Buyers are willing to pay more for tools that “just fit” their industry without customization.

How AI is accelerating this market dynamic:

• LLMs adapt particularly well to domain specific data, which lowers the barrier to building industry focused products.
• Smaller teams can now deliver highly specialized solutions that feel deeply integrated into a vertical workflow without building massive custom stacks.

What can be done:

• If you have no product yet: Pick a vertical and ride the wave. When you see verticalization starting, choose a segment where pain is high, workflows are complex, and willingness to pay is strong and double down on it.
• Defend against vertical attacks., decide which verticals matter most and build opinionated paths for them before others do.
• If you are horizontal: Become the platform layer. When vertical products win on depth and specificity, horizontals should avoid feature-by-feature competition. Instead, focus on being the flexible platform that connects tools, data, and workflows across verticals. Strong APIs, integrations, and extensibility let vertical solutions plug into you rather than replace you.

Practical Framework: How to track structural shifts in your software category

Now that we’ve seen how things are accelerating, here’s a quick practical guide on how you can assess the situation in your market.

1. Modelize your market structure

• Start with your direct competitors. List the companies solving the same core problem for the same target customer.
• Then map your indirect competitors. These are alternative solutions or adjacent tools that compete for the same budget or outcome.
• Finally, map the value chain around you. Who is above you, below you, before you, after you. These players might not compete directly, but they influence your positioning, pricing power, and strategic options.

This structure should give you a clear overview of the market you operate in.

2. Observe what is actually moving

• Check what each player has done over the past 12 to 18 months.
• Are new entrants accelerating? That might signal fragmentation.
• Are incumbents expanding horizontally and shipping new modules? That could indicate bundling.
• Are some categories shrinking or disappearing? That may be disintermediation.
• Are new regulations changing the way you operate? That can signal regulatory shifts.
• Are product paradigms changing? Are AI agents coming for your meal
• etc.

It is by monitoring these concrete moves that you start to see the big picture of where your market is heading.

3. Reassess frequently

• Revisit it every six months. As I mentioned above, in some categories, we are already at the second generation of new entrants in less than two years.
• Cycles are compressing. What looked stable last year might already be changing, not only tech is accelerating, market shifts as well.