The transition to making business decisions using data, rather than by gut instinct and experience alone, makes analytics projects as relevant as ever — without them, organizations would suffer from a dearth of insight with which to make these decisions. However, analytics project success can be elusive. Furthermore, organizations are navigating an evolving landscape of analysis tools owing to the proliferation of AI-driven offerings, as key findings from our Data & Analytics, Data-Driven Decisions 2022 survey illustrate.
AI has penetrated most tech sectors, analytics included. Indeed, our survey results show that AI-driven analytics tools will become the dominant offerings used to make future data-driven decisions. However, our results also show that companies won’t automatically ditch the investments they have made in existing analysis tools, such as dashboards and spreadsheets. The shift to AI-driven analytics could result in a better user experience for business decision-makers by letting machine learning take the strain, which could also address the skills gap — found to be the second most common barrier to analytics project success.
Further market education and evangelism are required to convince some analytics professionals of the value of a specific form of AI-driven analytics known as a “decision intelligence platform.” Some of these individuals regard decision intelligence as nothing more than the latest buzzword for business intelligence/analytics, even though the lion’s share agree with our definition of it as a new type of software with analysis and data management capabilities, built on a foundation of machine learning and designed to improve data-driven decision-making.
Summary of Findings
Project success in analytics is never guaranteed. Organizations on average see 11%-25% of their analytics projects fail. For 29% of respondents, more than one-quarter of their organization’s analytics projects have not been successful.
Project cost is the main barrier to success. Of all the reasons cited for why projects failed over the last 24 months, cost is the most frequently chosen factor at 31%.
A paucity of skills is another major hurdle. Cost is followed by a lack of skilled resources (27%) as the next most common reason for analytics projects to fail. A lack of appropriate skills could also be a major reason why Al-infused augmented and automated analysis tools are in ascendency for data-driven decision-making. More than half (53%) of respondents cite automated/augmented analysis tools (e.g., anomaly detection and root-cause analysis tools, AutoML platforms, natural language query and natural language generation tools) as the products their organization plans to use to make data-driven decisions in the next 24 months, followed by dashboards at 52%. By comparison, data-driven decision-making in the last 24 months was led by dashboards (56%), followed by spreadsheets (53%) and AI-driven analytics tools (47%).
Decision intelligence is interpreted differently depending on whom you ask. Most respondents (72%), all of whom are analytics professionals, agree with our definition of a decision intelligence platform as a new type of software with analysis and data management capabilities, built on a foundation of machine learning and designed to improve data-driven decision-making. However, slightly more than one-quarter of respondents (26%) believe decision intelligence is just the latest term for business intelligence/analytics.
Data scientists head up most analytic projects. For 59% of respondents, more than half of their analytics projects are led by data scientists. This leaning toward data-scientist-led analytics projects is particularly strong among organizations that we have identified as “tech leaders” (i.e., those that are early and fast-acting technology adopters). In contrast, organizations that we have identified as “tech laggards” are more inclined to have projects led by non-data scientists (i.e., business analysts, marketing and sales professionals, and other business decision-makers). Indeed, 30% of tech laggards say more than three-quarters of their analytics projects are led by non-data scientists.
Do you have your finger on the pulse of tech trends? Join the 451 Alliance for exclusive research content on industry-wide IT advancements. Do I qualify?