5 C's: AI STORY-TELLING IS MUCH MORE THAN JUST DATA STORIES

---

As per our 2020 surveys on AI adoption across end-user enterprises-

• more than 25% i.e. a quarter of senior-level respondents (Director and above, business and IT functions, 60%+  Northam respondents) mentioned that the AI initiatives have moved from being operations-focused, tactical point solutions and cost-saving initiatives, to strategic and transformative programs. 
• Only less than 15% respondents at this level said said they were still using AI just for operational efficiency gains and internal cost controls. 

Why must data storytelling mature into AI storytelling?

While 2020 saw enterprise AI adoption mature from point initiatives to strategic capabilities, only a small percentage of senior leadership respondents (less than 10%) suggested that they were satisfied with the level of maturity that their AI CoE's and teams have reached, in terms of AI strategic impact assessment, metrics definitions, articulation and communication. 

90%+ senior leaders admitted that neither their own AI narratives nor the ones from their tech/ service provider partners have matured into strategically relevant business-impact based stories.  

The dire needs to cultivate the art and science of AI storytelling becomes clearly manifest in these numbers. Good AI storytelling practices are really not visible, and often not even well understood by the leaders and AI practitioners themselves! This is true for a majority of the strategic enterprise-wide AI adoption initiatives, be it in technology & service providers landscapes or in end-user businesses and AI capability centers. Most of these organizations have got varied degrees of maturity when it comes to data storytelling, having invested in multiple data visualization tools. 

How to move up the maturity ladder of data storytelling to enterprise-grade strategic high-impact AI stories?

Data storytelling and AI storytelling have core similarities in terms of technical capability requirements and usage of visualization & narrative generation tools etc., but there are a few key differences across critical parameters like scope, capabilities & components, granularity, focus and impact:

1. Coverage/ Context/ Scope (What's covered in the story?): While the scope of data storytelling is usually tactical by definition and is limited to specific data science/ AI-ML projects, the scope of  AI storytelling is much broader. AI storytelling involves strategically important, core change narratives & AI adoption programs @scale, spanning across and touching the entire enterprise, either directly or in a cascading manner. This way, the context of data stories is mostly focused on and confined to the as-is, current-state narratives, whereas AI narratives, being strategic, are trend-setting, forward-looking, futuristic, can be speculative, simulated/ extrapolated or stretchable into the to-be unknown. 
2. Citizens/ Audience (For whom is the story built?): Data story-telling is mostly internal-facing i.e. the target audience is within the organizational functions, levels, business lines and partner ecosystems. AI storytelling is predominantly external-facing because it influences the key strategic messages of the enterprise, in the external market & competitive landscape. Also, the target audience for data storytelling are managers and business process leads and users at specific process/ task levels. The audience for AI storytelling are CXO-level strategic business leaders who are responsible for the key business KPIs of the company e.g. financial metrics like shareholder value, market valuation, growth, profit & operating margins etc. 
3. Capabilities (What are the key skills?): Key capabilities required for effective data story-telling are data science techstacks, modelling & visualization tools know-how and good interface, UX design skills. AI storytelling requires a deep as well as broad understanding of the overall enterprise business context, the market context and the competitive landscape, so that the scale of change brought by AI adoption can be seen in the exponential curve. 
4. Components (What are the main ingredients?): Data storytelling practices and tools center around the specific data science projects, the algorithms & model outcomes and its usage & efficacy in the specific problem contexts/ decisions/ actions. The critical components of AI storytelling are the long-term impact metrics and connections between the AI & data strategies to the core enterprise business strategy. 
5. Coarseness/ Granularity (What's the level of detail?): Data storytelling techniques and tools may allow detailed-level granularity e.g. slice-dice functionalities, for every data science project. For effective AI story-telling, terse, crisp, high-impact messaging is key, given that the CXO level audience may not have enough time to go through all the micro-details but would like to sum it all up in the Big Picture. 

Net net, enterprise leadership in end-users and/ or tech/service provider companies must mature from only data storytelling to more holistic AI storytelling, to move up the exponential AI impact/ value maturity curves. Only then the clarity & visibility of the road ahead will improve the certainty of their strategic decisions in the mid to long-term horizons.