Durable Sessions AI UX vs Coltivar Strategic Problem: Which?

// TL;DR

These two frameworks solve completely different problems and never compete for the same use case. If you are building or fixing an AI product's streaming architecture — handling disconnections, multi-device sessions, or agent control — use the Christensen Durable Sessions AI UX Framework. If you are a business owner stuck at a revenue plateau or preparing a strategy session, use the Coltivar Strategic Problem Framework. One is engineering infrastructure; the other is business strategy. Pick based on your problem domain, not preference.

// HOW DO THEY COMPARE?

DimensionChristensen Durable Sessions AI UX FrameworkColtivar Strategic Problem Framework
Best forEngineering teams building or auditing AI chat/agent product experiencesBusiness owners and leadership teams diagnosing growth stalls or planning strategy
DomainAI product architecture and real-time streaming infrastructureBusiness strategy, financial diagnosis, and constraint identification
ComplexityHigh — requires understanding of streaming protocols (SSE, WebSockets), pub/sub systems, and agent architecturesModerate — requires financial literacy and willingness to confront hard truths about the business
Time to applyDays to weeks for full architectural redesign; hours for an initial auditOne to two focused sessions for diagnosis; ongoing loop for execution and adjustment
PrerequisitesAn existing AI product with a streaming delivery layer (SSE, WebSocket, etc.)Basic financial data (revenue trends, margins, churn) and an existing or planned business
Output typeArchitectural audit, gap map, and redesigned streaming infrastructure using Durable SessionsA single Strategic Problem statement, three interrelated strategic choices, and two to three prioritized initiatives
Creator backgroundMike Christensen (Ably) — real-time infrastructure and AI engineeringColtivar — business strategy consultancy focused on small and mid-size businesses
Core mental modelDecouple agents from clients via a persistent shared session layer (pub/sub)Identify the single binding constraint and make all choices revolve around solving it
Failure mode addressedFragile AI UX — streams break on disconnect, no multi-device sync, no live agent controlDirectionless activity — initiative sprawl, vanity frameworks, ignoring financial reality
Iterative processValidate three capabilities (resilience, continuity, control), then layer on additional featuresScientific loop: problem → hypothesis → experiment → measure → adjust

What does the Christensen Durable Sessions AI UX Framework do?

The Christensen Durable Sessions AI UX Framework diagnoses and fixes the streaming architecture behind AI chat and agent-driven products. It was created by Mike Christensen of Ably and presented at AI Engineer.

The core insight is that most AI products fall into the Single-Connection Trap: the health of the AI response stream is coupled to a single client connection. If that connection drops — the user switches networks, changes tabs, or moves to a different device — the stream dies. The framework introduces Durable Sessions, a persistent shared layer between agents and clients built on pub/sub principles. Agents write events to the session; clients subscribe to it. Neither holds a direct pipe to the other.

This architectural inversion unlocks three foundational capabilities: Resilient Delivery (streams survive disconnections), Continuity Across Surfaces (sessions follow users across tabs and devices), and Live Control (users can steer, interrupt, or cancel agents mid-generation). The framework also solves the SSE Resume-Cancel Conflict and the Orchestrator Dual-Purpose Problem in multi-agent systems.

The workflow is a 10-step process starting with auditing your current streaming model, scoring it against the three capabilities, designing a Durable Sessions layer, redirecting agents and clients to use it, replacing SSE with bidirectional transport where needed, and validating the result.

What does the Coltivar Strategic Problem Framework do?

The Coltivar Strategic Problem Framework replaces unfocused planning sessions and popular-but-shallow business frameworks with a strategy rooted in one true constraint. It targets business owners, leadership teams, and anyone preparing for a strategy session or stuck at a revenue plateau.

The central principle is the Plan vs. Strategy Distinction: a plan says what you will do, while a strategy is an interrelated set of choices about where you will compete, how you will compete, and how you will win. If your document does not answer all three, you have a plan, not a strategy.

The framework insists you begin by diagnosing the financial reality — revenue trends, profit margins, churn, return on invested capital — and then naming the single Strategic Problem, the one constraint holding the business back. It explicitly warns against survivorship bias (copying Nike's origin story without understanding your economics), initiative sprawl (listing 30+ priorities), and setting growth targets that exceed your Sustainable Growth Rate.

The workflow is a 9-step scientific loop: audit existing plans, diagnose financials, calculate sustainable growth rate, name the Strategic Problem, make three interrelated strategic choices, form a hypothesis, reduce initiatives to two or three, connect mission and vision to strategy rather than the reverse, and review on a defined cadence.

How do they compare?

These frameworks operate in entirely different domains and solve entirely different problems. The Durable Sessions framework is an engineering architecture pattern for AI product teams; the Coltivar framework is a business strategy methodology for leadership teams.

They share one philosophical trait: both reject superficial approaches in favor of identifying and fixing a root cause. Durable Sessions rejects the assumption that model quality alone creates a great AI UX — it is the infrastructure that separates a fragile demo from a resilient product. Coltivar rejects the assumption that vision statements and popular frameworks create strategy — it is the financial diagnosis and constraint identification that separate activity from direction.

Beyond this philosophical overlap, they do not intersect. You would never use Durable Sessions to decide business priorities, and you would never use Coltivar to fix a broken streaming architecture. An AI product company could use both — Coltivar to identify that their strategic problem is poor user retention caused by broken UX, and then Durable Sessions to solve the specific technical constraint.

Which should you choose?

Choose the Christensen Durable Sessions AI UX Framework if you are an engineer, product architect, or technical leader working on an AI product and your streaming delivery is fragile. Specifically, if your streams break on disconnect, your users cannot resume sessions across devices, or your stop button is unreliable, this is your framework.

Choose the Coltivar Strategic Problem Framework if you are a business owner, founder, or leadership team member who needs to make strategic decisions about where the business is headed. Specifically, if your revenue is flat, you have too many priorities, or your planning sessions produce activity lists instead of strategy, this is your framework.

If you are an AI product company, you may need both — but at different organizational levels and for different audiences. They are complementary, not competing.

// FREQUENTLY ASKED QUESTIONS

Can I use Durable Sessions and the Coltivar Strategic Problem Framework together?

Yes, but they serve different audiences within the same organization. The Coltivar framework is for leadership and business strategy — it might identify that your strategic problem is poor AI product retention. The Durable Sessions framework is for your engineering team to fix the underlying technical architecture causing that retention problem. They are complementary layers, not alternatives.

Is the Durable Sessions framework only for AI products?

It was designed specifically for AI chat and agent-driven experiences, but the underlying principles — decoupling publishers from subscribers via persistent sessions, resumable streams, and bidirectional transport — apply to any real-time product with similar requirements. However, the framework's workflow, audit criteria, and failure modes are tailored to AI agent architectures specifically.

Does the Coltivar framework work for startups or only established businesses?

It works best for businesses with enough financial history to diagnose — revenue trends, margins, churn rates. Very early-stage startups without revenue data will struggle with Steps 2-3. However, the discipline of naming one Strategic Problem and making three interrelated choices is valuable at any stage. The Sustainable Growth Rate calculation becomes critical once revenue exists.

What technical skills do I need for the Durable Sessions framework?

You need to understand streaming protocols (SSE and WebSockets), basic pub/sub architecture, and your current AI agent stack (e.g., Vercel AI SDK, LangChain). You should be comfortable with concepts like sequence numbers, message buffering, and connection state management. It is an intermediate-to-advanced engineering framework, not a beginner tutorial.

Does the Coltivar framework replace EOS, Scaling Up, or OKRs?

Coltivar explicitly warns against adopting popular frameworks as 'flavor of the day' substitutes for real strategy. It does not replace execution systems like EOS or OKRs wholesale, but it insists that strategic direction — the Strategic Problem and three choices — must come first. Execution frameworks can then serve the strategy rather than replacing it.

How long does it take to implement Durable Sessions in an existing AI product?

The initial audit (Steps 1-3) can be completed in hours. Designing and implementing the Durable Sessions layer (Steps 4-8) typically takes days to weeks depending on the complexity of your agent topology and the maturity of your infrastructure. Validation (Step 9) adds another cycle. Expect a meaningful architectural refactor, not a quick configuration change.

What is the biggest mistake people make with business strategy according to Coltivar?

Confusing a plan with a strategy. Most businesses create lists of activities, assign owners, and call it strategic planning. The Coltivar framework insists that strategy must answer where you compete, how you compete, and how you win — and all three answers must reinforce each other. A list of initiatives, no matter how detailed, is not a strategy.

Can the Durable Sessions framework help with multi-agent AI systems?

Yes, this is one of its strongest use cases. It directly addresses the Orchestrator Dual-Purpose Problem, where an orchestrator is forced to both coordinate sub-agents and relay their progress updates. With Durable Sessions, each sub-agent writes directly to the shared session channel, eliminating the relay bottleneck and simplifying orchestrator code significantly.