Good Better Best, for Agents
Plus: Updates from Hubspot, Datadog, Shopify, SentinelOne, and Hexnode.
Welcome back to Good Better Best.
Each week, we break down real pricing, packaging, and product moves from SaaS leaders and share the ideas worth stealing.
Today, we’re breaking down how Customer.io is packaging agents within their existing tiers, and why this bundled approach feels like the future for most SaaS companies.
Before we get there, our friend Kyle Poyar is running his annual State of B2B Monetization Report. Take a couple minutes to fill it out, and you’ll get access to the full report when Kyle publishes it. Check it out here.
On to today’s post.
🔌 PricingSaaS Partners power the next era of SaaS pricing
This Week in Pricing, Packaging, and Product
This week we observed 100+ changes. The highlights:
Datadog Cloud SIEM added a minimum of 50 Bits AI investigations [Link]
Projuris launched an ROI calculator to quantify law firm efficiency [Link]
Deepbrain launched Interactive Avatar plans [Link]
SocialPilot added MCP Server Integration across all four paid plans [Link]
SentinelOne rebranded all three plans around AI Security messaging [Link]
HubSpot Sales added Custom Channels API for messaging integrations [Link]
HubSpot Service added ad automation and CRM segments in beta [Link]
Contentful Adjusted usage metric capacity [Link]
Shopify added a max to how much Plus customers can earn back [Link]
Hexnode moved Web Content Filtering from Ultimate to Ultra [Link]
Check out more updates on PricingSaaS →
PricingSaaS Pulse Intelligence
Here’s what was top of mind in Pulse this week:
🔥 Hot Companies
Zendesk — 13 searches
Zoom — 13 searches
Notion — 12 searches
Tableau — 12 searches
Salesforce — 11 searches
🚨 Hot Topics
AI monetization credits vs seats SaaS 2025
AI feature packaging for standalone AI SKUs
Value metric for pricing AI-generated content creation
Average price increase percentage without feature additions
Credit-based monetization for agents
Good Better Best, for Agents
In the last couple weeks, both Anthropic and OpenAI have launched their own agents. Two weeks ago, Anthropic launched Managed Agents, which allow developers to build powerful AI workers:
Then yesterday, OpenAI launched Workspace Agents, which are designed to be AI employees that live inside your existing tools and help automate tedious tasks.
One pretty obvious pattern we’ve seen over the past few years is that when the frontier labs offer new functionality, it’s only a matter of time before it trickles down to the app layer. With that in mind, it seems like a given that we’re about to see a lot more agents on SaaS pricing pages.
A few weeks ago, I highlighted two examples of agent packaging: Notion and Hubspot [link]. Both companies bundle agents into their higher tiers, and use a credit system to monetize usage.
This week, I want to showcase another option, from Customer.io, who just launched their AI Agent in beta.
What I find interesting here is that they have broken the agent down into a set of features, with differentiated access by plan.
All plans get persistent memory and core execution skills, but you need to upgrade to get higher session limits, custom execution skills, scheduled tasks (routines), and higher volumes of scheduled tasks.
Same agent product. Three different levels of capability.
Why the tier-differentiator approach wins
While the add-on model works for companies with flexible agents like Notion, there are 3 reasons I think the tiered approach will be more popular among SaaS companies with more specific jobs to be done.
First, no separate sales motion. The agent sells itself as part of the upgrade path. Essentials customers who need Daily Routines don’t buy an “AI Agent Pro” add-on for $X/month — they move to Premium. The agent’s value gets absorbed into the tier decision customers are already making.
Second, the capability ladder matches how customers actually experience agent value. Agents aren’t a single feature — they’re a level of automation. A small team gets value from an agent that can be triggered. A growing team gets value from an agent that runs on its own schedule. A large org gets value from an agent with custom constraints. That progression maps cleanly onto plan tiers.
Third, credits create usage-based expansion inside the existing plan economics. Customer.io’s 100K LLM Actions credits on Essentials are a conversion trigger — hit the limit, and you either buy more credits or move up. That’s the same Hybrid-Led Growth mechanic we’ve been watching for a year, but now applied to the agent surface specifically.
Looking Ahead
When foundation model providers ship new infrastructure, the app layer tends to pick it up within two to four months. Usually as an experimental add-on first, then as a core part of the product once the playbook stabilizes.
Customer.io is one of the first vendors we’ve seen to have a packaged option, and they’re not going to be the last. Watch for the next wave in categories where agents can replace or accelerate seat-based work — categories like marketing automation, customer support, sales engagement, project management.
Thanks for reading! If you’re working on monetization and want to learn more about how we help, book time here.
Until next time,
Rob