Three core principles for the demand intelligence age
The new normal is here and we’re learning what’s returning to pre-pandemic norms, and the many factors that aren’t. Demand patterns, customer expectations, staffing and inventory management have all changed drastically. We’ve been catapulted into the demand intelligence era faster than any of us expected. Let’s talk about what’s next.
The pandemic accelerated a major trend that had been going on for more than a decade: the digitization of business, and in particular the orienting of core business operations to be truly data-driven. From supply chain management to demand forecasting to labor optimization, businesses are rushing to enable the powerful advantage of building their forecasts and plans on objective data that brings new insight and efficiencies to their BAU.
Enabling businesses to become truly data driven is something PredictHQ has always been passionate about. But we noticed several key focuses of ours were shared by the other four businesses that Gartner® named as their Cool Vendors for Data for Artificial Intelligence and Machine Learning in late 2021. We were honoured to be named a Cool Vendor™, and I want to use this opportunity to talk about the new era of business operations.
Principle 1: Big data is dead, long live smart data
For decades, companies have been investing in data sources and strategies, while the volume of available data burgeoned. The amount of information out there is overwhelming and we’ve all learned the hard way that more data does not equal smarter business.
The demand intelligence age is characterized by the rise of high quality data that is verified and enriched for a particular purpose and is supported with prebuilt features. It’s not just information, it’s intelligence. The trend to seek out smart data and context is one Afraz Jaffri and team note in their Gartner Cool Vendors report:
"By 2025, 70% of organizations will be compelled to shift their focus from big to small and wide data, providing more context for analytics and making AI less data hungry.”
The report continues: “Data and analytics leaders want to improve the delivery of AI results with data innovations. AI teams are expanding their focus from model development to data that makes these models effective, and many of them are looking for AI-specific data offerings to improve and simplify their data-related efforts."
This evolution of teams seeking out the very best data intelligence, and the many specialist sources like PredictHQ and others providing it, is a trend growing so rapidly it’ll soon become how everyone does business, like the maturation of the cloud from cutting edge to core.
Principle 2: No more data grenades
Not only is the data getting smarter, the manner with which demand intelligence companies deliver it is too, and not a moment too soon. It’s no longer acceptable for demand intelligence or data companies to just send over a tranche of data and expect companies to invest months to get it working.
No one has that kind of time these days. Margins are tighter, teams are smaller, and everyone is trying to do more with less. Yet even before the pandemic, demand intelligence companies weren’t just throwing data at their customers and partners and hoping they could make sense of it.
The demand intelligence era is being defined by companies that are not only transforming data into intelligence, but building the tools and infrastructure to make the most of it. Whether that’s through specialised API endpoints such as feature APIs or unique feature engineering to mitigate gaps in similar data sets, the move from “have a bunch of data” to “here is exactly what you need to get started, here's how it scales and we’re with you every step of the way” is long overdue. Perhaps most interestingly, many companies are investing in building the tools that quickly identify the connection between their intelligence and their customer’s data, to expedite customer journeys to value.
With so many data sets out there, the demand intelligence era will be defined by businesses that can quickly prove the value of new data, see where it should be integrated and what critical outputs can be generated to support their teams in a more adaptable and real-world manner.
Principle 3: Free your data scientists from data cleaning so they can drive value
These first two principles lead to one of the most transformative: the mainstreaming of higher quality data that’s built for purpose, enabling data scientists to spend more time building models and discoverings insights, and less time cleaning and verifying data.
Survey after survey of data scientists puts the time these teams spend finding, fixing and cleaning data at ~50% and more. Even the more conservative surveys, such as this one, that found data scientists spend 45% of their time cleaning data, and only 11 to 12% of their time on model selection, model training and scoring, and model deployment.
This means these highly qualified team members are spending a minority of their time doing what they earned multiple degrees to do: build and refine models that can sort the signal from the noise and help your business make the best decisions.
What happens next
The demand intelligence era and the rise of smart and well architected intelligence and features means this huge cost should rapidly become simply a bad memory. Seeking out intelligence over simply more data means data scientists can invest more of their time and expertise in models and initiatives that move the needle for your business, and less time sorting through data.
This can have major implications for businesses, especially as they respond to a world where demand patterns have changed. Data is the only reliable guide through the chaos and it’s never been more competitive out there.
Unfortunately the disruption of the pandemic has only made this inevitable evolution more urgent than ever. As Mckinsey’s Chief Data Officer Mohammed Aaser put it well, the pandemic “provided an example of just how relevant external data can be. In a few short months, consumer purchasing habits, activities, and digital behavior changed dramatically, making preexisting consumer research, forecasts, and predictive models obsolete. Moreover, as organizations scrambled to understand these changing patterns, they discovered little of use in their internal data. Meanwhile, a wealth of external data could—and still can—help organizations plan and respond at a granular level.”
Applying the above three principles of the demand intelligence era, it’s not simply external data that will be the defining feature of best practice in demand forecasting, planning, staffing, supply chain management and pricing. It’ll be data and intelligence that's smart, fit for purpose and built for integration, letting data scientists do the best work of their lives for your company.
The era of demand intelligence is here. The only question is where your business will land on the adoption curve: as an early winner, or a business that loses years playing catch up.