Digging Back Into AI ROI
Hosts: Trisha Merriam, Chris Carolan, Erin Wiggers
September 2, 2026
Chris Carolan, Trisha Merriam and Erin Wiggers return to AI ROI three weeks after the series closed and find it harder. Chris shows the custom HubSpot object that catches value when someone says it out loud; Erin's calculator attributes revenue and hours saved, then names its own limit.
Moments from this episode
2 short cuts from this conversation.
Key takeaways
AI does not always give time back — Trisha Merriam's HubSpot audits still take the same amount of time to deliver, but they go far deeper now that the manual data-pulling is offloaded.
If you do not actively help someone understand the value you delivered, they will guess at it, and most of the time they will guess wrong.
A client who has never received a good version of your deliverable has no baseline to judge the one you just handed them, which is why depth of expertise so often goes unpriced.
Chris Carolan built a custom object in HubSpot called the value entry so that the moment someone says out loud that something landed becomes a record, instead of dying in a notes field or buried in a transcript.
The value entry's stages run foundational value, capability shown and multiplied value: a website built is foundational, a website people actually use is a capability, and people sharing their own success with it is multiplication.
Agree what value looks like at the milestones up front, rather than finishing a project and asking how it felt, by which point the answer is unrecoverable.
Insight abundance can create negative value — it is easier than ever to surface findings, and a client who cannot act on all of them ends up feeling worse, so ship a prioritised plan of what to fix first alongside the findings.
Erin Wiggers's ROI calculator attributes value three ways: AI-assisted revenue from the share of a deal's tracked time spent on AI activity, AI-retained revenue where an AI alert prompted a save, and time saved as the gap between scoped and actual hours priced at a blended rate.
The ROI number that comes out of a well-instrumented system is usually so high that people assume it is broken, which is why Erin stopped showing the total on its own and broke it into buckets to keep it credible.
The person who built the dollar-based ROI view named its own limit on air: it is objective, it does not incorporate the value side, and she wishes it did.
The most useful ROI question is not how to split the difference but what happens if you do nothing at all — measure against the alternative, not the compromise.
Before trying to measure AI value, get visibility over the whole story: if your data and context are not connected, connecting them is the work that comes first and it activates almost everything else.
Resources from this episode
Transcript
This transcript was generated by AI and may contain errors. It is provided to help you skim and search the conversation.
Keep learning
You just watched the work. Now build it.
The next step is doing it yourself — live, with people on the same path.
Get the next conversation in your inbox
One signal-dense note when a new episode lands. No noise.