AI is making it easier to build software.
That does not make the product decisions easier.
As development gets faster, the harder questions move upstream and downstream: what should actually be built, how should analytics help customers act, and which capabilities are valuable enough to charge more for?
That is what this edition is about.
We look at how AI is changing the role of software engineers, why embedded analytics becomes more valuable when it moves closer to execution, and how to package analytics inside a tiered pricing model.
Let’s get into it.
📰 Upcoming in this issue
The Future of Building Software Products
How Insight-to-Action Analytics Drives Retention
How to Package Embedded Analytics for Tiered Pricing
📈 Trending news
AI Coding Agents Are Becoming a SaaS Architecture Problem
Why SaaS Teams May Need Less Autonomy and More Coordination
AI Makes SaaS Easier to Build. Product Judgment Gets Harder.
The Future of Building Software Products
AI is making software faster to build, but that does not necessarily mean software products will become dramatically cheaper.
Many SaaS teams are already using AI to complete coding work in hours or days that once took weeks. But coding is only one part of building a durable software product.
As AI increases the speed and volume of code production, the harder problem becomes managing what gets built and keeping systems maintainable.
That shifts more value toward:
Architecture and technical standards
Testing, observability, and security
Reviewing AI-generated code
Managing technical debt and complexity
Making better product and platform decisions
The bottleneck is moving from writing software to governing software.
Senior engineers will spend less time producing every line of code themselves and more time setting guardrails, reviewing designs, controlling complexity, and making decisions that hold up over time.
AI does not eliminate software engineering. It changes where the most valuable work happens.
Key Takeaways
AI makes development faster, but expectations rise with it.
Architecture, testing, security, and governance become more important.
Senior technical judgment becomes more valuable as code production gets cheaper.
How Insight-to-Action Analytics Drives Retention
Most embedded analytics helps users understand what is happening.
The stronger products help them decide what to do next.
That is the difference between analytics that informs and analytics that becomes part of how the business actually operates.
Take inventory management. A traditional dashboard might flag that stock is running low. The user then has to review the issue, decide what to do, and manually start the replenishment process.
Insight-to-action analytics can go further. It can calculate what is needed, prepare the order, and route it for approval inside the application.
That closes the gap between seeing a problem and acting on it.
The closer analytics gets to execution, the more valuable it becomes. It can trigger workflows, generate recommendations, prepare transactions, or surface actions when certain conditions are met.
The goal is not to remove human judgment. It is to make the next step easier, faster, and more consistent.
When analytics becomes part of everyday workflows, customers rely on it more deeply. Removing it no longer means losing a dashboard. It means disrupting how work gets done.
Key Takeaways
Analytics gets stickier when it supports execution, not just reporting.
Closing the gap between insight and action creates more value.
The closer analytics gets to core workflows, the harder the product becomes to replace.
Pricing & Packaging Analytics: Part 1 of 3
How to Package Embedded Analytics for Tiered Pricing
Over the next three editions, we’re looking at three ways SaaS companies can price and package embedded analytics:
Part 1: Tiered pricing
Part 2: Usage-based pricing
Part 3: Seat-based pricing
This week, we’re starting with tiered pricing.
Tiered pricing works well when analytics capabilities become more valuable as customer needs get more sophisticated.
The key is separating what customers already expect from what they are willing to pay more for.
Basic dashboards, KPI tracking, and operational reporting usually belong in the core product. Higher tiers can then unlock capabilities such as self-service analytics, custom metrics, AI-assisted analysis, predictive insights, workflow triggers, and advanced governance.
A simple structure could look like:
Starter: standard dashboards and reports
Growth: custom dashboards, cohort analysis, and self-service analytics
Scale: predictive alerts, AI-assisted recommendations, automation, and advanced controls
The strongest packaging follows customer maturity. Entry-level customers need visibility. Growing customers need flexibility. Larger customers need automation, governance, and more control.
Avoid differentiating plans only by report counts or dashboard limits. Customers are more likely to pay for new capabilities than simply more of the same.
Key Takeaways
Keep expected reporting in the core product.
Use advanced analytics to create meaningful reasons to upgrade.
Package around customer maturity, not arbitrary limits.
Revisit what belongs in each tier as expectations change.
Why It Matters
Faster development is useful, but speed alone does not create a stronger SaaS product.
The bigger advantage comes from making better product decisions, building analytics into the workflows customers depend on, and packaging those capabilities in a way that matches how customer needs grow.
AI may make more software possible. The harder work is deciding what deserves to be built, how it creates value, and how that value should be priced.
See you in the next edition,
Arman Eshraghi
The Embedded Intelligence Brief
Connect on LinkedIn
P.S. Qrvey is hosting a live panel on Building SaaS with AI on October 21. If you are building with AI coding tools and hitting friction, you won’t want to miss this. Save your spot here →