Jumping from lagging to leading in operationalising business-side AI at DataTorque
AI is everywhere. Most SaaS companies have started in product and engineering. But the bigger, messier challenge is what happens across the rest of the business.
Nick Steevens (CFO + VP Shared Services at DataTorque) put it perfectly when he described seeing a much larger category leader’s’ suite of tools: “It was a bit of an oh **** moment for us. We saw how far we were behind the leaders in this space and knew we had to move.”
That moment didn’t turn into a strategy deck. It turned into action — and the learnings are basically a trailer for what we’ll dig into at Operationalising AI for Business Teams.
1) Going from “dabbling” to “we’re doing this”Nick describes the early stage as a bit of internal tinkering, plus conversations externally as AI started “becoming increasingly a thing”.
The shift came when leadership got hands-on. DataTorque’s CEO (a software engineer by trade) jumped into the tooling himself and realised that the opportunity was immediate. There was real substance — so they started pushing it seriously.
That’s one of the big questions this workshop is built around: what does leadership conviction actually look like when you’re trying to land AI in finance, marketing, people, customer teams — not just engineering?
2) Business teams need more than tools — they need enablementDataTorque didn’t just buy licences and tell people to “use AI.” Nick says they went looking for who in the market was genuinely clued up and hired dedicated capability, bringing in an AI consultant who became Head of AI Enablement.
That’s the heart of operationalising AI on the business side: enablement, adoption, and change management, not just tech choices.
3) A real business-side example: tenders, pricing, and 1,000 requirementsWhen asked for a concrete example outside product, Nick went straight to business process — especially complex spreadsheets.
DataTorque prices tenders that can involve formal government procurement with up to 1,000 requirements, priced individually. They’ve built AI tooling that:
scrapes how their existing product is configured
ingests tender requirements
matches requirement-by-requirement
flags gaps
suggests how to word answers
generates the first tranche of pricing
Nick’s estimate on impact:
before: 5–10 FTE over ~6 intensive weeks (50–60 hour weeks)
now: two people running a bid, others mainly reviewing output
and they’ve cut ~60–65% of the groundwork
This is exactly the kind of ‘real value, not just noise’ example we want in the room — because it’s not theoretical, and it’s not polished. It’s a business team finding leverage.
4) The part people forget: morale, autonomy and momentumNick also shared something you don’t always see in AI rollouts: morale going up. They automated monthly reporting work that used to take two to three days and turned it into an automatic output in the same spreadsheet format — but without people having to touch it.
Nick says their shared services team survey hit the highest morale he’s seen since taking over, because people are automating repetitive work and getting “a lot of joy and satisfaction” from building things they previously couldn’t.
That’s the adoption flywheel: once people get to something useful, they start enjoying it — and momentum follows.
5) No one-size-fits-all: it’s culture as much as toolingNick’s biggest takeaway is that there’s no single answer to operationalising AI. Different cohorts are at different places.
Trying to force AI by handing out tools works for some and not most. Instead, he says you have to meet people where they’re at — workshops, demos, hack days, soft launches — “a lot of carrots”… and eventually “a little bit of a stick.”
If you’ve been wondering why business-side adoption feels harder than product/engineering adoption, this is the real conversation. Large KiwiSaaS members can join us on Wednesday 20 May in Auckland to hear first hand how leaders are navigating this shift. Register now.
