In military theory, it is said that no battle plan ever survives first contact with the enemy. In business, hopefully, there is no enemy, but plans still have to change quickly as the competitive landscape changes. Planning cycles tend to be too long, and companies struggle to change plans in response to real-time events.
At the recent Anaplan Hub Conference, Anaplan CEO Frank Calderoni (pictured), laid out a solution to the problem of obsolete plans in the face of changing conditions—kill planning. “Planning as we know it is dead,” Calderoni declared.
That’s a bold statement from a company that sells planning software. Founded in England seven years ago, this San Francisco-based firm has grown to a valuation of over $1.4 billion by dedicating itself to the concept of “connected planning.” Connected planning is the concept that, when using traditional methods, companies not only have data silos but also planning silos. Often the HR plan and the supply-chain plan and marketing plan don’t talk to each other or even use the same data. Connected planning is about using single sources of data and plans that are connected to each other to destroy those silos (for more on connected planning, see Anaplan Hopes to Launch Era of Connected Planning).
So, if planning is dead, what is Anaplan going to sell?
Calderoni is selling a new vision where “the distance between planning and action is zero. Zero months, zero days, zero hours.” Think of it as connected planning on steroids. The steroids are machine learning and artificial intelligence. AI and machine learning were conspicuously absent from any mention at the 2017 Anaplan Hub conference, where the focus was on a “customer first” and community-oriented strategy. This year wasn’t exactly a pivot away from people, but toward a user experience driven by intelligence.
AI for the People
Calderoni provided a demo of a near-future look at connected planning that featured a fictitious VP of marketing named “Ana” who discovers that a competitor has released a product shortly before her own expected launch of a similar product. In the demo, Ana interacts through voice, tablet, smart screens, and even an automated car during her commute and is able to change the launch date of her new product in a matter of hours to better compete.
In the process, Ana interacts via email, text, and instant message with her team. But the whole thing is coordinated by a digital assistant that actually suggests (before Ana thinks of it) that moving the launch date will increase sales. In another demo, the software suggests product giveaways for a cosmetic product, as people tend to buy cosmetics more often after getting to use them. Anaplan also demoed a dashboard that showed supply-chain difficulties and made suggestions for how to make up for the loss.
The goal here is to reduce planning cycles and the time it takes to change those plans to as close to the point of action as possible. Take the launch of a new product: with traditional planning techniques, the planning cycle for the manufacture, marketing, shipping, and sale of a new product could take weeks or months. Change one variable, and it could take significant time to remodel and rebuild the plan. Anaplan believes machine learning and AI will allow for planners to play “what if” with their plans and do it on the fly.
What if we moved up the launch by two weeks? Could the supply chain get the product on the shelves in time? What would it cost? Is marketing in line? What types of weather would interfere? What other market conditions could make this a good or a bad idea? How would it affect the entire plan? Anaplan says that once all the tools are in place (and executives were a bit coy on the exact timing for all of it), all of these questions could be answered in real-time, in natural language queries, putting a great deal of power in the hands of planners and decision makers.
A Curated Experience
In an interview with Computer Economics, Anaplan’s Daniel Eyre, senior product manager, platform technologies, said that the best way to think of the suggestions in the demo is as “curated content.” The goal of the machine learning and artificial intelligence isn’t merely to provide numbers and data but to look at context and provide different data stories to different users. He compared using data to any other consumer experience, whether it is with a voice activated digital assistant, a mobile device, or even possibly a smart and connected kitchen product. Eliminating the friction for the user allows for better work and better decisions.
“As decision making begins to blend into your daily life and [Anaplan] does more heavy lifting and starts generating content and starts being prescriptive, the skill set and the work of people in the enterprise will change,” Eyre said. “When you look at data in a silo you fall into the trap of ‘I’m working with data right now.’” But Eyre and Anaplan believe that once you get out of the concept of, “I’m working with data”, you can reduce the space between planning and action, and that will lead to better business decisions.
In other words, if you can stop asking if the numbers add up and the model is correct and start thinking about your business, then you get out of the modeling and planning business and into the business of running your company.
Obstacles in the Path
Discussions with customers indicate they are pleased with Anaplan’s direction. Anaplan talks about a “cascade effect” where customers buy Anaplan to solve a single pain point and it quickly spreads to usage all over the organization. That matches the description from the customers themselves, most of whom have stories about other departments adopting Anaplan after they see it.
However, an executive at a Fortune 500 financial services and manufacturing company told us that she did have one problem with Anaplan. She only had one developer who had significant experience with the product. She had just hired a second, but the second wasn’t yet up to speed. “I can’t let my genius go,” she said. “I have to share my developer and it just isn’t enough. We could do more if we had the developers.”
The lack of developers was a major recurring theme, and Anaplan is attempting to solve the issue. Anaplan is training and certifying developers, and the company has been establishing a more mature customer care unit. It is also building a developer community to foster the exchange of information and ideas. “The big system integrators are now seeing us as a revenue builder for the first time, as well,” said founder and CTO Michael Gould. “That is at a tipping point. We also have a good growth in boutique partner consulting firms that want to specialize in Anaplan.”
Another major factor is that Anaplan still has a long way to go in spreading the message of connected planning. Not all Anaplan Hub attendees that we spoke to had a firm grasp on what connected planning means to them, and Anaplan executives admitted that their sales team often has to explain who Anaplan is and what it can do for potential customers.
Last, there is the technology aspect of Calderoni’s vision. Anaplan insists much of the technology for their planning at zero hours is already in place. The market is saturated with voice assistants, for example. And, Anaplan is using Google’s TensorFlow for machine learning and AI. The challenge, according to Anaplan, is simply fitting all the pieces together in a coherent, intelligent, and useful way. Many vendor demonstrations AI revolve around similar voice-activated digital assistants. Most are extremely limited right now. Serving up the kind of context-specific curated content Anaplan wants to produce is easier said than done. Making it easy to use without the need for highly-experienced developers will be a trick.
Connect Planning at Something Greater than Zero
We’re likely not very close to the point where the distance between planning and action is zero. However, the journey is valuable in its own right. Anaplan added 40 capabilities to its offerings this year, including improved workflow and collaboration tools. It has also delivered Optimizer, which allows planners to work with some of the “what-if” scenarios that may have previously taken weeks or months to work through. This is how Anaplan describes Optimizer:
Through the use of advanced mathematical modeling and algorithmic problem-solving, Optimizer determines pathways to ideal outcomes for any challenging, multi-variable decision with enormous speed. Optimizer can determine preferred outcomes for many complex questions, from pricing and staffing to capitalization, asset utilization, and much more. Users can define objectives, such as revenue, profit maximization, or cost reduction, then set multiple variables or constraints to guide the planning process.
These offerings are building toward the vision of connected planning put forth by Calderoni. Or as founder Michael Gould said when asked about how close they were to zero, “We aren’t there yet. But what we are doing is enabling decisions, and we are dramatically reducing cycle time. At least we are heading in that direction.” It’s the right direction, as long as the company can continue to scale developers familiar with Anaplan, spread the concept of connected planning, and keep the whole process clean and simple to use.
(Photo credit: Peter Prato)