Make Data Management a Hedge Against Uncertainty

The rapid supply chain changes of 2020-2021 are barely in the rearview mirror, and with trade barriers in flux the world is changing fast once again. In a changing strategic landscape, organizations need to be able to shift strategy quickly. Decision-makers need to put their attention outward, focusing on the changes that are out of their control and taking or preparing for action. An effective data management program supports leaders’ ability to do this.

The challenge is that many data programs are built like utilities: effective, reliable, deliberate and slow to change. Thinking of data like a fire department can help: fast, coordinated, and empowered. This is what “contingency bureaucracies” do, and that thinking can make your data program able to flex when you need it most.

In a typical organization, the reports that executives use the manage the business day-to-day are well curated. Follow-up questions are predictable and routine. The “Utility Model” appears both effective and efficient. However, when a typical organization confronts atypical, unpredictable market changes, its Data Management program is put to the test.

Here are three things that are built into contingency bureaucracy organization models that can be applied to your Data Management program to make it flexible:

Standardize Core Tools for Interoperability

IT and data teams often mirror the organizational model that they support. While this is a good thing when it comes to ensuring fit-for-purpose service delivery, behind the scenes in a crisis it is helpful if certain foundational functions of Data Management are consistent. For example, often a multi-divisional enterprise will have independent Data Management teams by division and a Data Management team at “corporate.” The corporate team handles reconciliation across divisions on topics relevant for the reports senior executives use but does not oversee the day-to-day work of the divisional teams.

When there is a sudden strategic shift and data previously only relevant at the divisional level is needed at corporate, having a consistent foundation will accelerate the data team’s ability to deliver new insights. If each division is using different tools for their dictionaries, metadata, lineage etcetera, it will simply take longer to wrangle the data: disambiguate definitions, identify relevant data and confirm gaps, etc.

Small companies benefit from a “consolidated core” as well. While it may seem financially prudent to run the governance program using spreadsheets in standard times, a crisis is not a good time to discover that the 4 data stewards have been using their “standard” spreadsheets in very different ways.

Cross-Train Data Teams Ahead of Them Needing to Collaborate

A mostly standard service delivery model, where role definitions align etc., will facilitate rapid communication in much the same way that standard tooling will. Along with this, consider more cross-functional teams even under circumstances where they are not strictly necessary.

This will serve two purposes. Much as “cross training” application support engineers ensures there is someone knowledgeable about even small systems when the primary engineer is out of the office, cross-training Data Management stewards and engineers means that you have more people with broad knowledge. When the scope of Data Management is changing and expanding, it will be easier for a team with “broad knowledge” to come together than one of head-down specialists.

The second reason is simple human nature. It is easier for a team to pull together when people already have worked together. While “Communities of Practice,” “Knowledge Sharing Sessions,” and similar interventions are meant to facilitate this, they cannot replace the tacit knowledge about collaboration that people gain from working together towards a goal. Even a sophisticated Data Governance Operating Model that interlocks divisional and corporate committees will not do this. If the organization is siloed, these will operate in a “you mind your stuff, I’ll mind mine” kind of way.

Make some important projects truly cross-functional so that knowledge is dispersed and the team is comfortable working together under pressure.

Clear Ownership of Data Management for the Crisis

Contingency bureaucracies combine the shared tooling and active teambuilding of a traditional bureaucracy with the dispersed decision-making of task-oriented organizations. In a crisis situation, it needs to be clear to senior management who is “on point” for data delivery and within the Data Management community, that individual needs to be entrusted with the authority to get what they need on behalf of management. Senior executives should not be faced with competing views of the reality within their own organization. During this time their attention needs to be outside-focused.

Summary

Data Management can help hedge against uncertainty if it is structured and managed with the inevitability of crisis in mind. During a time of rapid strategic change, the commodities in shortest supply are time and information. You can reduce friction by:

·      Standardizing tools across Data Management teams

·      Cross-training the data organization before broadening knowledge appears relevant

·      Clear “on point” leaders who have authority to act across organizational boundaries

Building these capabilities into your standard operating model ahead of any a disruption will position the data team to respond effectively during a crisis.