If you lay a strong foundation for data management that is based on good, reliable rules, which are managed and governed, you will do well in providing good information to assist decision making. Makes sense, right?
- Business must lead
What is the goal of the project, why are you running this project, what will you achieve, what are you trying to solve? Most if not all of these will be business challenges: effected by accuracy of data, inaccurate reports, incomplete customer information, poor marketing and logistics information. The owners and stakeholders of these processes and procedures are the business people who must lead the projects. They have a vested interest (hopefully) helping success be achieved.
Each process, data item and dataset (Dataset definition: a collection of related data. Most commonly a data set corresponds to the contents of a single database table, which may be accessed individually or in combination or managed as a whole entity) within the organisation needs to be owned, defined and governed.
Process Owners – enables effective management of adherence to the process.
Data Stewards – day to day accountability for care and maintenance of the data.
Data Owners – owner of data, bears ultimate responsibility for the security, quality and lifecycle management of the data.
More detailed explanations can be found in my earlier blog “Data Governance Roles” http://wp.me/p3QDdI-t
- Think big – start small
“Don’t boil the Ocean” – a familiar phase in the industry. If you want a cup of tea, would you try to boil all the water in the ocean? J
Know what you want to achieve as a business, it could be better satisfied customers / reduced logistics costs / better predictive analytics to assist accurate and faster decision making. Whatever it maybe you cannot change everything all at once, for one the business needs to continue running effectively while any change takes place.
Start small – demonstrate a quick win to gain support and buy-in. Plan which domain/area will be next. This is a journey, not a big bang!
Understand how your data management, data quality and data governance programmes combined in the enterprise. Is your data quality programme part of your customer engagement initiative or could it be?
- Meta Data
Increasingly important & available we are creating metadata each time we transact on our mobiles. GPS / Time and Date stamp etc., on a photograph.
Why is your metadata important? Good, accurate, well thought though metadata has value for creating associations and relationships between items and users across one or more sources of data or related applications, such an ERP (Enterprise resource planning) or CRM (Customer relationship management) system, as well as its benefits for instilling consistency in the way information is generated, used, its value, limitations, how it is stored and shared.
- Measuring Problem Cost
You know your problem that needs to be solved, or you should do. Do not start a Data Management or Big Data project without knowing what you are trying to solve. How will you demonstrate that your project was a success? Sentiment these days is often measured, great but what if the journey was painful sentiment will still be negative. You need cold hard facts and figures to support your efforts and maybe even that the project should commence initially.
How can you measure the problem?…Is it in time wasted, costs being spent on rework or duplicated effort, regulatory costs.
Whatever it may be baseline the cost/time out the outset of the project. Issue regular progress improvement measures, showing where you came from and where you are now. i.e data was showing 50% duplication’s now down to 45%. Target 5%.
How can the measurement be shown in monetary terms? i.e 5% reduction in duplication’s leads to 20% increase in accuracy or reporting and a 10% in excess stock holding and customer satisfaction improvements due to more accurate billing.
- Best Technology Platform
Although this is a business led project, this is the point where your IT team can demonstrate their expertise. Consider what systems you already have – are they being used to their full extent – can they be developed, extended, integrated?
What systems are being used within your industry? Or by an industry you are taking the lead from. Consider Airline solutions for booking tickets in other industries i.e Train transport.
Seek Gartner & Forrester studies for industry leading and niche providers.
How are the systems rated? What are their strengths and weaknesses?
Analyse the strengths and weaknesses of each system carefully against your requirements. Nothing will fit 100%.
What is the best fit? What is a “must have” vs a “would like”? How will the system be futureproof to the changing needs of your industry, regulation and customer needs?
- Organisational Governance & Change Management
Develop your governance policies and procedures with input from the business. Best way to achieve this is to form a Data Governance Committee with key areas represented by your stakeholders and other area representation invited as required.
Do not underestimate the time needed to instil new processes and change for your people. Data is about the people, without them the processes and technology will not succeed. Updating and correcting data and systems always takes longer than expected, especially if it is done to the required standard (which is what you do want) – you do not want to end up with a Toxic Data Dump!.
Evaluate and plan the time needed for the journey, to achieve your various phases, demonstrating progress and success along the way, with consideration for down time, sickness and holidays.