Marketo's co-founder is now building an AI rival to it. The shift is from rules to reasoning, but rules still own consent
A group of former Marketo builders say AI can replace the rule trees marketers maintain by hand. Here is what that means for your CRM workflows.
What happened
In a MarTech column, marketing operations practitioner Moni Oloyede describes a set of platforms positioning themselves against Marketo: Paminga, Inflection.io, Conversion.ai and Phave. Inflection's site says "Replace Marketo, finally." Paminga offers a Marketo Translator for migrations. Phave was founded by Marketo co-founder Jon Miller and former Marketo product executive Nick Bonfiglio, and Inflection's co-founders also worked at Marketo.
Miller describes the change as moving from rules to reasoning. Instead of marketers encoding every scenario as a workflow, newer platforms evaluate customer context and signals and pick an action within objectives and constraints. The column notes rules do not go away: consent, compliance and anything that needs predictable behavior still runs on them.
Oloyede also shares a classic rules engine trap. She changed a Wait step from three days to five, and people already in the step kept the three day wait. Only new entrants got five. Adobe, for its part, has added Marketo AI skills and an MCP server.
My take
This is not just a Marketo story. The same rule trees live in HubSpot workflows, GoHighLevel automations and Zapier chains, and they break in the same quiet ways. The Wait step example is the one I see most: a change looks applied, but records already in flight follow the old logic.
How I would split the work between rules and AI in a CRM:
- Rules: consent and unsubscribe checks, data residency, routing that must be explainable, SLA timers, anything legal or financial
- AI: picking the next nurture message, summarizing intent from emails and calls, scoring fit from messy notes, suggesting the next step for a rep
- Logs: every AI decision written back to the contact record with the reason, so someone can audit it
You do not need a new platform to start. An AI step inside your current workflow, wrapped in hard rules on both sides, gets most of the benefit. Migrate later if the platform itself is the bottleneck.
More posts
- Uber says AI agents booked four rides in its busiest week of October. Its app takes about 50 seconds to book oneNov 5, 2026
- A 2B parameter model beat frontier LLMs at pulling fields from invoices and IDs in its makers' benchmark, at 6 to 147 times lower costNov 5, 2026
- The best AI agent finished 29 of 45 real AWS tasks. About 80% of failures came from stopping too early, not from missing knowledgeNov 5, 2026
