Data strategy: discovery & design

Discovery & Design establishes what data exists, where, and in what condition as a basis for designing or updating these structures needed to fit with the new Cloud environment but also the wider ecosystem, governance, transparency and security requirements as appropriate.

  • Information gathering & planning; once the frame is established during kick-off and related tasks, a broad brush sketch of the landscape and destination needs to be set out. This should include items like:
    • Incorvus will assess the client’s intended path and advise any early difficulties, e.g. between legacy and Cloud.
    • Landscaping: is the overall IT landscape and business environment understood in broad terms, including where data may be located?
    • Destination-scaping: are the current strategic plans of the business understood? Are future influences, trends, technologies understood? Are there any projects planned, likely to be affected by the work or have an effect? Is there a common view of the intended objectives and what the destination (and success) looks like?
    • Capacity management: assessing the amount of data likely to be moved and any network constraints that may cause service disruption for other systems or users in order to determine mitigation;
    • Resource management: are resources available for interviews, meetings, other foreseeable tasks? Are there any resources (human or otherwise) needed for the work to get underway?
    • Problem identification: have any likely problems already been identified? What about the ones that may not yet have been identified?
    • Documentation and reference material/data: are these available to the project team?
  • Data – find, (extract – where necessary) and evaluate: this determines what data and/or data sources you have (or not!) and in what condition; identify new sources and data that you may potentially have access to or be able to use, deploy or create:
    • Establish and locate the existing target data sources, including spreadsheets and legacy systems/stored archives;
    • Extract (if required – may be needed to understand exactly what is needed in packaged applications) and evaluate the sources, the data and the metadata to determine nature, quality, quantity or obsolescence etc. as a precursor to determining any remedial actions required. Incorvus uses Safyr to speed the understanding of data held within packaged applications such as ERP; and also to identify areas of particular interest to DPO’s such as sensitive personal data;
    • Discover: Moving data from an ERP/CRM system requires an understanding of both the ‘source’ and ‘target’ systems. If one of these is an ERP or CRM system, gaining an understanding of its data structure in order to map that structure to another system can be challenging. We use Safyr to speed the process of showing the user which tables exist in the packaged application, how they are related (and in most packages) show which application functions use which tables.
    • Reporting and advisory on legacy and consolidation issues.
  • Describe existing structures, models & taxonomies: to move forward, one must first establish, clarify and map existing data structures: metadata, access privileges, data dictionaries, data and table structures, field names, users, user rights, data flows, workflow, processes, content types, formats, relevance, accuracy and quality;
  • Identify remedial actions: produces a gap analysis of issues such as poor quality data or metadata, gaps in tables, discrepancies, dark data, data format, parsing, obsolescence etc.
  • Design and architect new data structures, models and taxonomies (suited to the new environment or Cloud), aligned to governance; new processes; new technologies; the relevance of canonical data; policies, standards and protocols;
  • Information governance advisory is offered to assist those organisations trying to align data structures and processes to new regulations.

This will establish the new, draft data structures, rules and principles while the Gap/Quality/Discrepancy analysis determines what data preparation actions are necessary.