Every minute of every day, an increasing number of businesses collect vast amounts of data and attempt to leverage value from their estates. These organizations expect these swathes of data to support key strategic change, such as shifting gears from revenue-driven to profit-driven and enabling opportunities within new markets.
But in reality, companies are too focused on measuring the past and comparing data sets with Key Performance Indicators (KPIs) rather than asking the important questions that can unlock a deeper understanding of the value of the data. Without this questioning approach, businesses will lose insight and waste capability, even if they leverage increasingly intelligent technological tools at their disposal.
The “just store everything” data model, promoted so heavily by IT functions, cannot provide the most valuable knowledge. If companies are to deliver operational benefits from the extensive and expansive investments in data, it is time to step outside the box and ask the important questions.
Letting go of the past
The number of businesses adopting data collection technology is broad, and in almost all cases this is fronted by a data-driven culture where collection for “KPI analysis” is the main goal. However, while this is admirable, it’s far from what the businesses are achieving. By measuring past data sets and comparing KPIs, they’re simply tracking performance.
This is a completely backwards-focused approach, with no comprehension of the future or ability to create smarter strategic change. This system of monitor-and-measure offers businesses no significant perception of the future.
Data must be used proactively, with a shift from the tracking of KPIs to the strategic relevance of Key Performance Questions (KPQs). This can reverse the mindset of “monitor and measure” into a focus around “how did we meet our targets?” and instead of “how many products have been sold this week?” it becomes “are we selling to all the right customers?”
This shift looks to drive change for the future. Rather than simply monitoring the past against targets, it questions whether the business is going in the right direction and identifies further opportunities for that successful profit-driven approach.
Barriers to insight
Why does this closed-circuit model exist in the first place? It’s down to the basic fact that data scientists and vendors have fuelled the myth that data can do anything and solve any problem as long as it has access to unlimited computing power.
But this isn’t the case. Businesses need to move beyond KPIs and analysis of the past. Adding more data sources without that critical direction only incurs cost without adding business benefit. A recent 451 research determined that the biggest barrier to successful machine learning deployments was a lack of skilled resources, followed by the challenges in accessing and preparing data. Simply put, the more data-driven the organization, the bigger the issue is.
Businesses need to move beyond the strategy of adding more tools to analyse the past; the barrier to data value must be overcome through KPQs and in turn this can identify the best people to discuss and implement the resultant insight. Suddenly, rather than looking at 25, 30 even 100 data sources, the KPQs may require analysis of just five or six.
This form of data analysis clearly requires more thought, but when it’s applied intelligently into the real world, the benefits can be transformative to a business.
Take the example of a travel business that decides it wants to move up-market to offer more luxurious packages that come at a higher price and offer a better margin. This business could just look at KPIs to determine whether their sales are matching desired levels.
But then it would miss out the core element of identifying customers that would be attracted by a more up-market offer, where they would like to go and what compels them to buy. Without asking these KPQs in the first place, the data will simply answer a set KPIs and provide nothing further. It would be a waste of expensive and expansive cloud-based technology that has no real strategic focus.
By implementing smarter thinking and questioning the value in the data, you unlock strategic knowledge, such as “does marketing reach out to the right audience to enable this transition?”
Now consider the example of a logistics company managing key sections of global supply chains for large retailers. The KPIs typically exist around speed and profitability per section, whether that’s delivery or time taken for stock to get to the distribution center and back out.
By transforming that into KPQ terms, the question moves from “how fast can stock be moved through the chain?” to “who else can we sell our most profitable service to?” This opens up new transformative business opportunities.
This process not only advances businesses in the moment, but leverages future innovation opportunities through experimentation, building and usage of artificial intelligence to investigate deeper into the answers to KPQs.
This approach to data obviates the need for a data scientist in many cases because it relinquishes the control that scientists had through a valuable combination of the right KPQs, the right experts to hand and strategically-prepared data. This enables businesses to improve business acrumenu at transformative speeds, and allows for trends to be seen and understood immediately, with the right context and KPQs in place to ensure this has an immediate business benefit.
How is this done? AI can blend internal valuable data with external resources, such as market data, to deliver this in real-time and provide the advantage of business opportunities. This is an area where that computing power does have a valuable role to play. This form of data landscaping provides immediate information about the existence of mathematically identifiable connections between data outcomes.
If that connection does not exist, then the business is either looking at the wrong data, or that data is incomplete. And this is an issue that companies will need to embrace: KPIs measure existing performance, based on internal data sources. KPQs may well demand additional external data and computing resources, but provide information on opportunities that can promote business change for sustainable success.
For instance, in a food retail business, this could manifest as AI acknowledging outside data resources, such as changes in food trends, against stock plans and current product sales, to offer a smarter way forward in which experts alter and adjust stock levels to match the AI-produced, expected demand levels for alternative products, e.g., vegan alternatives. This approach would identify where the business is currently missing out on potential customers, and where marketing needs to be focused to ensure a successful implementation of stock changes.
Too many companies are simply sitting back and hoping that simply collecting vast quantities of data and measuring past performance will provide valuable information. Unfortunately data doesn’t work that way, regardless of the intelligence of AI or machine learning. There has to be a clear strategic change to focus on insight generated not from past performance but from ongoing and future aims and opportunities.
Businesses must move beyond being data driven to become insight driven with direction and with the right expertise on hand to deliver the resultant hidden value. Proactive questions—not measurements—will result in crucial transformative change.