I’m just about to blow a gasket over this “SaaS is dead” AI narrative. Tell me something: if SaaS is so dead, then why were the most compelling customer stories I heard this spring – with bottom line benefits and energized users – about moves to SaaS?
Yes, those stories typically had AI themes in there somewhere, but it leads me back to these conclusions:
- Modern enterprise software is pretty darn good, if you select and implement it correctly.
- The data discipline imposed by SaaS migrations turns out to be a darn good precursor to better AI. AI can inhale your word documents (be careful), but it can’t make sense of rogue spreadsheet chaos.
- It’s a whole lot easier to absorb AI functionality from your SaaS vendor of choice than to fine tune an LLM and do your own AI.
- Oh, and your SaaS vendor likely has solid security and role-based permissions – have fun trying to teach an external LLM about your company’s access permissions and hierarchies!
Ergo, there isn’t an a contradiction between SaaS projects, data quality and AI pursuits. From an organizational change and ‘AI readiness’ perspective, these can be parallel undertakings.
Few vendors nailed this down in their keynote as well as Planful. Why? Because Planful emphasized AI security in the keynote, taking attendees further under the hood of technical details that matter. Think finance leaders don’t care about security and risk? Why did so many vendors gloss over this in their spring keynotes?
Are FP&A maturity models sexier than AI prompts?
Consider industrial supplier Kimball Midwest. They are early adopters of Planful AI; they spoke about it onstage, and even showed up in Planful’s AI Innovations event press release.
But this all happens in the context of Kimball Midwest’s finance modernization push, driven by the results from Planful to date, e.g. less than two hours a month for monthly-close-to-reporting, 95 percent reduction in errors, and less than five percent of time on data consolidation and manipulation. And: getting budget prep and forecast slides down from seven days to one (Planful has posted Kimball Midwest’s team presentation, How Kimball Midwest Brought Its Entire Business Onto Planful, so you can check out how they got there).
I was at this presentation. While it was instructive to hear Kimball’s AI views, what really got my attention was their internal transformation model:
Now, maybe it takes an enterprise geek diehard like me to truly love a slide like this. Maturity models like this one aren’t as sexy to some people as magical AI prompts, but how do you make change happen if you don’t know where you’re going? So, during my sit down with Kevin Washek, Director of FP&A at Kimball Midwest, I asked him about these phases – and where Kimball Midwest is headed. It’s worth noting that Kimball’s FP&A transformation is happening in a century-old business. As per Kimball Midwest’s session description:
Kimball’s team will explain how they use Planful beyond FP&A and into sales, supply chain, HR, marketing, legal, and more. Kevin will also detail how it automated month-end packages with Spotlight, slashed reporting and variance analysis from a week to a day, and shifted FP&A to a strategic, insights-driven resource.
Budgeting – from Excel to Structured Planning to Dynamic Planning
Let’s start with budgeting, and how Kimball Midwest moved from Excel to the current state, Structured Planning – with Dynamic Planning on the horizon. I asked Washek: what does that push into Dynamic Planning look like? He explained: “It’s an opportunity to dig deeper into the business.”
However, Dynamic Planning is currently on hold, awaiting the taxonomy and hierarchy changes of an ERP transformation in progress (“Fun stuff,” jokes Washek). But in the meantime, Washek’s team is looking for “pockets” within the business where they can “drive some of the value back into the organization.” One potential target is putting the sales data into Dynamic Planning:
There’s a lot of detail behind those numbers, right? Are we selling the right mix of products? Are we targeting the right size of customer? A customer in this region, they typically buy fluid flow, right? So are we selling fluid flow in there? There’s opportunity to put that detail into Planful to help us get to that data quicker.
Forecasting – from “non-existent” to rolling forecasts
With Planful, Kimball Midwest has already moved from non-existent forecasting to quarterly forecasting. But what does the “rolling forecast” future look like? As Washek told me, Kimball Midwest’s teams have already gone through a huge amount of change getting to quarterly forecasts:
We started from Word and transition to Excel and then Planful. So it’s been a lot of change for the departments, and they’ve handled that really well, but we have to be mindful of that. From a forecast standpoint, last year was our first full year forecast. We’ve still got to get the teams comfortable with what that process is, and what we’re trying to generate from there – and I think we’re getting there.
But why is the rolling forecast compelling? Washek:
It would allow us to really understand, for example, if sales are super-accelerated, do we have the right right cost or the right expense or the right spend to support that business, or to support that growth going into the future? Doing it on a rolling basis or a monthly basis, would allow us to adjust those course connections.
Reporting – from manual to Office 365 integration to data storytelling
Moving from “manual/time consuming” reporting to Spotlight (Planful/Office integration) makes sense. But what about the data storytelling goal? Washek says data storytelling is about two things: getting the foundation of data trust in place, and building that deeper collaboration with the business. The time saved from the nitty gritty of manual tasks is what makes that relationship-building possible:
Spotlight has been a game changer for us, and I completely say that over and over. My team has done a fantastic job of getting everything linked and fluid within Spotlight. And now, to your point, we’re able to spend more time with the business, understanding the reason behind certain variances, and that helps shape the story. So I think ultimately, the good FP&A professionals, as storytellers, have to have the information and the understanding of the business to really shape and tell that story.
Planful customers and the “AI readiness” conversation
Washek’s three person FP&A team is always looking for ways to punch above their weight, and offer more value to the 45 budget owners they support. AI certainly qualifies – but what type of AI? Planful AI is always in the conversation. Why? Because it checks all the security boxes. Washek:
For us, security is the top priority… Financials are a little bit restricted throughout our company. Having that ability, and knowing that Planful’s AI is going to follow the initial security parameters that are originally set up is also very refreshing to hear. So if I’m in marketing, I can only get the answers back from the AI that would be in my security clearance.
In his keynote appearance, Washek talked about why rolling out AI in parallel to other FP&A functions makes sense:
As the business grows, more and more is being asked of the team, and so we’re really excited about the prospects of technology and AI and where we can leverage that to supplement some of our work, to continue to remain an efficient, valuable resource across the organization. So for FPA, we’re bought in.
When we think more broadly across the organization, we take more of a methodical approach, and introduce tools in parallel with the current processes. So for instance, we’re leveraging Planful Predict in parallel to our current forecasting process to forecast out our variable expenses, to really highlight that ease of use, and show the value across the organization.
In my Planful Perform AI review, I shared a similar view from Rocket Software’s Luis Martinez Luna. As Luna told me, consuming AI through Planful is changing Rocket Software’s ability to bring AI to business users:
The vendors that have the capability and the expertise in the technology space, they’re just embedding AI into the user experience. What that does for us is we no longer have to worry about building a data science team, and we can keep our functional experts, whether it’s marketing or finance or whatever organization, focused on what is core to them. They don’t need to go necessarily go learn a new skill set; they just need to learn how to interact with these new capabilities… Something like a Planful solution is just rolling out the capabilities as part of their monthly update cycle.
The push for better finance AI – what we’ve learned
In the push for better finance AI, we’ve checked off some key boxes:
- SaaS software imposes a useful data discipline (and much-needed security/role access in all AI interactions).
- Predictive tools for anomaly detection and variance analysis gets the ball rolling.
- Early adoption/co-innovating with your SaaS provider launches the iterative process and org change that AI requires.
But there is another topic to track: will rolling out Planful new AI assistant for help FP&A teams serve their constituents better, by giving business users a heightened ability to self-service their own finance queries? The customers I talked to issued an emphatic yes. Here’s what Washek told attendees:
Every FP&A professional knows that there are times we get slammed with questions, and this is a ripe opportunity for Planful’s AI assistant to take a lot of those questions off of our plate, and really be the one-stop-shop for our business partners to ask questions, and get those answers back about their business.
Can AI help Washek’s team become truly strategic partners to the business? He says yes:
It’s only going to get better as time goes on. Some of the stuff that [Planful] was sharing earlier, that’s going to allow FP&A to be more pro-active – and be that strategic business partner that businesses need.
The wrap – savvy customers ask better AI questions
When I hear eagerness to roll out chatbots to business users, warning bells go off. Even though assistants like Planful’s are much more built-for-purpose than say, asking ChatGPT for financial projections, there is still the potential for inaccurate information for the query at hand. (Planful’s CTO Sanjay Vyas has found some very interesting ways to minimize this architecturally – he goes so far as to say “no hallucations.” I’ve shared some of those architectural details, and I plan to share more). In other cases, the user might ask the wrong question, even if the AI assistant delivers the information correctly.
But I was impressed by the savvy of Planful customers on this. They were well aware that AI chatbot output isn’t something you would hand directly to an auditor. Another danger: users can trust the tools a bit too much, and need a bit more prompting of their own, to cross-check where needed. (Example: the AI generates several different scenarios on the fly, but the numbers in the chosen scenarios are then validated before acting upon them).
Kimball Midwest also has good things to day about Workforce Pro, Planful’s new workforce planning enhancements. To me, that’s another sign that modern software still has big impact, with or without AI. During Kimball’s team presentation, I was struck by the team chemistry between Washek and his fellow team members, Jillian Channell and Abby Brown (check the session, and I think you’ll see what I mean). I’ve said it before, but it’s the way that technology changes human collaboration that really stick with me.
And with that, Planful Peform London is about to kick off. Time to hand the coverage baton to my colleague Phil Wainewright.
