RevOps could only use structured data
RevOps has been tying together different teams with different tools and systems and data layer stuff for 15 or 20 years. The problem was that the only data it could use to fix these problems was structured data.
AI readiness
Is your company ready for AI?
By Kevin Stout, founder of Revductive · Updated September 2026
Short answer
AI readiness is whether your systems, processes and data are in a state where AI can build on top of them. Nine times out of ten, when someone’s talking about using more AI in their business, they mean the go-to-market side, not their product, and that leads back to operations. They need their systems and processes and data in a place where the foundation is usable. If it’s not, you’re just amplifying a mess.
Most readiness frameworks score you on infrastructure, governance and model management. For a B2B SaaS company under $50M ARR, what stops the project is almost always that the process underneath it was already broken.
Book a 30-min call →02 / Why the projects fail
What the failures have in common.
Look at all of the horror stories. We implemented AI, we invested this much in AI in our go-to-market stack, and everything failed, or it didn’t work, or we’re not getting good results. If you really zoom in, the whole customer lifecycle infrastructure was a mess in the first place. You can’t plug something that accelerates a process into a process that’s already broken.
It’s frustrating and boring, I think, for a lot of people. They want to move fast, they want to do the new thing. And I’m saying hold on, we need to slow down for a second, because if we just plug this into what you have right now, it’s going to make things worse.
The teams I see getting the most benefit out of AI on the go-to-market side are the ones that already had a good foundation in the first place.
03 / What gets checked
Five things to check first.
These are in the order I usually run into them, and only the last one is about tooling.
01
Whether the documented process is the one people follow
Whatever’s documented, if it’s even documented, is never the thing that’s happening. An AI workflow gets built against a description of the process, so when the description is wrong, what you build is wrong in the same place.
02
The handoffs between teams
Marketing to sales, sales to onboarding, onboarding to CS. That’s where the leaks are in a lot of software companies, because every department gets its own KPI and it’s hard to figure out what the shared KPI is during the handoff. Routing is usually the first one to break.
03
Who owns the space in between
A lot of companies are structured so everybody owns one part of the funnel. What ends up happening is there are always in-between stages. It’s not perfect boxes, it’s a lot of gray areas in between it all, and nobody owns those, because nobody’s incentivized to care about something outside their own number.
04
Whether anyone can describe the business in specific terms
Ask what happens at a particular point in the process. If the answer is that you should probably talk to somebody else, those two teams don’t have a shared definition of how they hand work to each other. That’s what the AI would have had to encode.
05
Whether the systems capture enough to learn from
Unless your processes and systems are built in a way that keeps track, you’re not going to know what’s working. You can’t iterate quickly, you can’t try new things quickly, you do the half version of it, and then you make a decision on gut feel instead of data.
04 / What changed
What’s changed in RevOps since AI.
Unstructured data is usable now
It’s the same toolkit RevOps has been using for 20 years, except unstructured data is useful now, and in a lot of cases you can turn unstructured data into structured data.
So the constraint moved
When the constraint was a clean trigger, the fix was better fields. Now it’s whether anybody wrote down how the work gets done, which is a people and process question.
Working out the role side of this instead? That has its own page.
GTM engineer vs RevOps →05 / How it gets assessed
I interview people before I open the systems.
A questionnaire gets you one person’s version of how the company works, and it’s usually a manager’s.
01
Interview the people doing the work
I started off as a journalist, of all things. Journalist, then editor. So I’m good at interviewing people across a lot of different skill sets, and I can have a conversation with an engineer as well as a salesperson.
02
Go into the systems
I take those interviews and dig into the systems to work out what the workflow really is. The ICs say one thing, the managers say another thing, the data says another thing, and you can usually glean what’s going on.
03
Find the handoff that’s costing the most
If a company is generating leads that aren’t converting into money, and nobody can say where it stops, they’re describing a lack of visibility into their own funnel.
04
Say what to fix before anything gets automated
You get an order of operations instead of a tool list. What has to be true before AI is worth pointing at this, and what you can point it at today.
Both audits run this process. The Systems & Lifecycle Audit walks the customer lifecycle stage by stage, and the AI Effectiveness Audit walks the org function by function.
06 / The order
Where the order goes wrong.
The version of this advice that does damage is the one that turns into a data hygiene project you have to finish before you’re allowed to touch AI. That’s the opposite failure, and I’ve watched companies use it as a reason to do nothing for a year.
It’s per workflow. Most companies have something clean enough to point AI at this month and something that would be a bad idea to automate at all, and what you want out of this is knowing which is which.
Cleaning records while the thing that dirties them is still running gets you a clean CRM for about a month, which is why the process comes before the cleanup. What the cleanup involves.
07 / Where to start
If you want somebody to run this.
The five checks above are the same questions the audit asks. What you’re buying is somebody running them against your systems who has seen the answers at other companies.
Not ready for a monthly engagement
Start with the two-week audit instead.
An inventory of what your team is currently using and how, who is using it effectively, and a 6 month plan for what to automate. The full $1,200 comes off month one if you continue.
$1,200
Flat
$0
If you continue
08 / FAQ
Questions that come up
What is AI readiness?
AI readiness is whether your systems, processes and data are in a state where AI can build on top of them. Nine times out of ten, when someone’s talking about using more AI in their business, they mean the go-to-market side, not their product. That leads back to RevOps, because they need their systems and processes and data in a place where the foundation is usable for AI. If it’s not, you’re just amplifying a mess.
Why do most AI projects fail?
Look at the horror stories where a company invested in AI across its go-to-market stack and got nothing back. If you really zoom in, the whole customer lifecycle infrastructure was a mess in the first place. You can’t plug something that accelerates a process into a process that’s already broken. The teams I see getting the most benefit out of AI on the go-to-market side are the ones that already had a good foundation.
Do I need to clean up my CRM before using AI?
Usually you fix the process first and the data follows. If you clean records while the thing that dirties them is still running, you get a clean CRM for about a month. So the order is to find where the workflow differs from the documented one, fix the handoff that keeps breaking, clean the data that handoff was corrupting, and then automate. Most companies skip to the automating step.
How do you assess AI readiness?
I interview people first and go into the systems second. I started off as a journalist, so I’m good at interviewing people across a lot of different skill sets, and I can have a conversation with an engineer as well as a salesperson. Then I take those interviews and dig into the systems to work out what the workflow really is. The ICs say one thing, the managers say another thing, the data says another thing, and you can usually glean what’s going on. Whatever’s documented, if it’s even documented, is never the thing that’s happening.
What is the difference between AI readiness and an AI strategy?
A strategy says what you intend to do. Readiness says whether you can currently do it. Most AI strategy work skips the second question, so you end up with a plan written for a company that can’t execute it, and then the plan gets blamed when nothing ships.
Is my company too early for this?
The moment this usually matters is going from founder-led sales to building a team. Before that the founder is the contact person throughout the entire process, so nothing shows up. Then you hire the first rep, everything that lived in the founder’s head has to be translated to other human beings, and that’s when the handoffs start leaking.
Working out whether you need training or building? Enablement vs implementation. Ready to build and pricing it? Internal AI workflows on Claude or ChatGPT.
Who wrote this
Kevin Stout
Fifteen years in B2B SaaS operations and growth, and co-founder of a healthcare SaaS product. First operations hire at Pearl, a dental AI company, and stayed five years while it grew from the early days to a couple of hundred people. He has talked through this work on the HabitStack podcast and on Show Me Your Stack, and Supered and SaaSGrid have both published case studies on it. He runs Revductive, a fractional RevOps and AI enablement service.