Memo
Market Reset
The 2026 agent market is splitting into two different businesses: heavy enterprise replacement systems and lightweight brand-owned conversational layers. At the same time, buyers are getting tired of "agent washing", and broad "ask anything" chat is underperforming compared with agents placed at high-intent moments such as pricing, onboarding, and decision pages.
Legacy Platforms (CX, IT/HR, Enterprise)
- Today’s model: seat-based SaaS, services-heavy enterprise contracts, or support-suite expansion.
- Challenge: capable agents weaken seat logic, while bundled AI lowers visible price but does not solve brand ownership, speed, or content freshness.
- Implication: cannibalization risk stays high, so incumbents pull AI back into copilots, governance layers, and contact-center replacement projects.
- Typical responses: bundle AI into suites, sell observability/governance, or reposition toward enterprise workflow control.
Model / Platform Players
- Today’s model: usage-based APIs, builder studios, or platform subscriptions.
- Challenge: they capture compute value but rarely own the customer-facing brand surface; pricing can feel opaque to operators and finance teams.
- Implication: platform gravity is strong, but business differentiation is weak unless they move closer to distribution and workflow ownership.
- Typical responses: launch agent builders, marketplaces, enterprise hosting, and reference apps on top of the model layer.
Where HEA Fits
- Category: brand-owned conversational infrastructure for websites and selected messaging touchpoints.
- Packaging: per-HEA subscription plus pooled usage, with agency workspaces and later governance / analytics upsell.
- Retention design: deploy HEAs where intent already exists; do not depend on generic open-ended chat as the core usage pattern.
- Non-goal: do not compete head-on as a contact-center replacement or a multi-month services-led transformation.
Go-to-Market Logic
- Self-serve wedge: creators and SMEs adopt quickly when deployment is measured in minutes and pricing is predictable.
- Agency flywheel: one agency can repeat deployments across multiple client websites and verticals.
- Corporate-side path: later expansion comes through selective regulated or enterprise-side deployments once references, analytics, and command-center controls are proven.
Risks To Manage
- Novelty decay: many agents are launched once and then ignored if they are not tied to a job-to-be-done.
- Bundle pressure: incumbents can market AI as "included", even when the experience is generic.
- Ownership ambiguity: unclear responsibility across marketing, product, and support can make agents go stale.
- Over-compliance: safety features that create too much friction will reduce adoption even if governance improves.
Takeaway
HEA wins if it owns the brand front door, keeps pricing predictable, proves repeat usage at intent-rich touchpoints, and expands through agencies before moving deeper into governance-heavy enterprise use cases.
Porter’s 5 Forces — HEA-World
1) Threat of New Entrants — Medium–High
- Foundation models, OSS, and agent builders keep technical barriers low.
- The real moat moves toward distribution, packaged onboarding, and repeat usage.
- HEA angle: fast go-live, brand-ready templates, agency reuse, and EU-friendly governance.
2) Bargaining Power of Suppliers — Medium
- Model vendors, hosting, auth, and messaging channels can pressure margins and roadmap flexibility.
- Supplier power is moderated by multi-model options and falling infra costs.
- HEA angle: multi-model routing, retrieval, and caching reduce dependence on any single vendor.
3) Bargaining Power of Buyers — High (SMB) / Medium (Enterprise)
- Buyers face many options and growing confusion about what is actually an "agent".
- Finance teams increasingly dislike opaque usage pricing or hidden seat taxes.
- HEA angle: keep packaging simple, show funnel lift, and tie value to owned digital touchpoints.
4) Threat of Substitutes — High
- Search, AI answer boxes, live chat, messaging apps, and bundled copilots all compete for the same user moment.
- HEA angle: own intent-heavy surfaces on the brand site with better context, CTA guidance, and measurement.
5) Rivalry Among Existing Competitors — High
- The market is crowded, but much of it is converging around contact-center AI or workflow control.
- Competition intensifies through bundling, fast-follower productization, and category confusion.
- HEA angle: stay focused on the brand-owned conversational layer, not the call center.
Summary: Win by owning high-intent brand touchpoints, proving repeat usage, and keeping pricing predictable while others drift toward enterprise complexity or infra commoditization.
Matrix
2×2: Heavy enterprise systems vs lightweight brand-owned agents
Positions are directional. Update the points array as the category keeps moving.
