Data Strategy … Is it Transformative ?

Often we have heard of data strategy & transformation , depending on where an organisation is in its journey, one could adopt a strategy that could transform the way of working.

While i could go on about what is the right strategy, irrespective of the approach for a single source of truth or multiple version of truth that fits your business goals the three key challenges

Challenge 1 : Is my customer your customer ? with multiple versions of customer databases or Product SKU versions depending on localisations, there are chances that each functions/market teams are dealing with the same customer or products under different names. With no single owner, often these require a new ways of referring a single source of truth (SSOT) for consistency and scalability of the business

Remediation suggestion : There is a need to agree on a single source of truth for core data consumption across the organisation, BU’s for consistent sales performance reporting or pricing or forecasting.

Challenge 2 : Local on-prem vs Globalised platforms . With localised data is inevitable for transactions (for eg: Distributor data or POS data ) often a lot of processing of the data is done in local on-premise systems or databases which is local to one country/zone/cluster/region. Therefore, when this data is used for central report for finance whether it is global or regional , there is a lot of dependency or waiting on on the local teams.

Remediation suggestion : There is a need to embrace cloud or one dev platform for data Ops across the organisation to govern dependencies and knowledge management. The need for a standard DevOps is often underestimated and usually leaves a big overheads footprint in terms of the ‘shadow’ IT functions.

Challenge 3 : While your business may be well intentioned to do the transformation, often its the mindset of the people who are at the grassroot level which holds back getting the outcomes in spite of calories spent on the transformation. While mindset is the first stumbling block, what often follows is the know-how to use the new ways of consuming information. I.e imagine consuming a power BI dashboard containing all commercial analytics needed for a role vs, accessing 8 different reports separately to ascertain business performance.

Remediation suggestion : There is a need to change mindsets, starting with leaders who need to lead by setting examples, promoting and supporting a culture of self-service and data stewardship.

BONUS : Here is a very insightful revelation of how an AI data project looks like in action ! Given that the pursuit if getting more value from Data, often people think AI is single switch that can do the magic, where it is the combination of Data engineering (Data engineers ) , Machine learning (Data Scientists) and Operationalising it (Cloud Ops) by effectively handling all the Security, Legal, Regulatory and Ethical issues if any. Often the backend work of clearing the constraints ( i.e the road mines) is the biggest challenge for the success of such projects or initiatives. Often, teams fail here as they think just have a set of technical resources can do this job for them, but the fact that business needs to wake up to this and understand this is not a one time activity but a new ways of working where there is overall need to upskill their understanding and ways of working to make an effective attempt to make AI work for their business. If traditionally , if a Sales Ops team could take care of all this manual work in a shadow IT setup , in the modern context this would be a set of ML Ops, Data Ops, Cloud Ops team together with Data )Product) owners who own , drive utilisation of the data and the insights generated based on business cases.