Our recent article on 5 critical success factors to turn data into insight, written by our Founder and CEO Kelle O’Neal, indicated there should be concern around the not-so-successful outcomes of many analytics and big data projects.
In her article, Kelle also covered five capabilities that contribute to the success of analytics-based initiatives.
Business alignment – Determine context and value of using information
Data understanding – Seek to better understand data assets and manage accordingly
Data quality – Define accuracy for the purpose for which data is being used
Data-centric processes – Increase understanding as new data is created, used, managed and measured as part of operational processes
Data-centric resources – Embed data-oriented knowledge and skills throughout the staff
In this article, I will explore business alignment techniques, such as Business Information Requirements BIRSM or “Line of Sight,” that support these capabilities and further secure a foundation for the success and sustainability of the analytics effort.
Analytics vs. Business Intelligence
When doing analytics, context is key. A lot of companies begin thinking about analytics and immediately start to address operational and managerial reporting, trapping themselves in the “low-hanging fruit” metaphor. This is not analytics in the current context. It is good old-fashioned business intelligence (BI). And while there is nothing wrong with BI, applying nouveaux technology to old BI problems is expensive.
There is a common thread across any technique to beef up data capability ─ that is, gaining a clear understanding of what is needed by the organization, and using the requirements from this to establish analytics capabilities. The techniques are applied during business alignment, requirements definition and the rest of the systems development life cycle and then are carried through the other capabilities.
Business alignment through Business Information Requirements BIRSM or “Line of Sight” technique
The business purpose of analytics is often overlooked. After all, analytics will reveal new insights ─ so why do we need to understand business purpose? The problem with that is business purpose drives context of data use. Data meaning will change based on context.
Therefore, some idea of context is required or the data will not be viewed in the proper light. It is far too common to reply to business demands and supply technology without defined purpose or context. And these efforts tend to fail.
Alignment techniques work across all five capabilities referenced above. At a high level, alignment is accomplished via the following steps:
1. Understand business strategy – If all expectations of analytics are delivered, what is different? What does the bottom line look like?
2. Decompose strategy into how data will be used to help meet strategic goals – Far too often we deliver what a constituents asks for versus delivering what the business needs. Strategies need to be expressed as business needs – not wants.
3. Decompose data usage into the critical component parts of data element, metrics, dimensions, lists and values – Every single “requirement” for data to be managed can be expressed as either a metric, fact or contributor to that metric or fact. Doing this keeps the requirement in a business context.
4. Understand patterns of use – Some data usage might require high volumes of data and high velocity, i.e., real-time analytics. Others might require data that is pre-calculated and stable, but drives interactions, such as a customer satisfaction score. Either way, there are patterns of use that ─ once understood ─ make it pretty easy to determine if you are really doing analytics or just really good BI.
Context and purpose
The data elements and other components and actions required to fulfill strategy will provide a sense of context and purpose, and will be carried through into all of the other capabilities. The best way to understand this is to look at one example of each.
Understand business goals and objectives - Examine your organization’s strategies, goals or even incentive plans. What measurable changes are on the horizon? More customers? More products? Lower cycle times? These all need to be measured.
Example: ACME wants to increase market share in the millennial demographic by four percent.
Decompose goals and objectives into levers, usage cases or other statement of business requirement – What do you need to use data for to achieve the goal?
Example: Understand buying patterns in millennial households.
Derive KPI, metrics and BIRs – To understand the buying patterns, you will need to do some data analysis. That means metrics, algorithms and slicing and dicing the data.
Example: Millennial household by income levels, region and education level need to be understood. (This is a juicy requirement!) We need households, data dimensions of regions, income levels and maybe even facts of individual incomes.
Analyze/categorize KPIs, metrics and BIRs by characteristics and features – Decompose those cool requirements into their bits and pieces. Data elements, ranges of dimension values, and even describe the algorithms.
Example: Millennial household income by region will use government-reported census data sliced by age, and will use regional economic data from the Federal Reserve Bank. Household will need to be a bit of reference data that ACME has to build.
Group KPIs, Metrics and BIRs into patterns of framework elements and analyze for technical requirements
Example: The data will need to be accessed daily for ACME to generate offers to these households. We also want the customer record flagged when a member of this data set contacts ACME, so we can apply special handling to keep them engaged at various customer touch points. So while this data is calculated, it has an operational aspect. These patterns will drive how we design and deliver the BI or analytical solution to ACME users.
There are clear benefits for defining techniques to manage the analytics process. First San Francisco Partners has applied this technique to many of our clients, and here are benefits they shared with us:
- Much more rapid development of data strategies and delivery architectures. A typical scenario will have an organization determine a data strategy in a third of the time of similar efforts where informal alignment techniques are used.
- Deeper levels of comprehension of data strategies by ultimate users
- More economical acquisition of tools and higher levels of satisfaction with solutions
- Higher awareness of meta data and data glossary requirements
- Quicker realization of benefits from analytical and BI solutions
Remember, business alignment techniques support your process so you are able to prioritize BI/analytics efforts to meet business needs; reach consensus on business vocabulary and definitions; craft solutions that meet all customer needs and deliver value; manage costs and risks embedded in current reporting and BI approaches; and partner to expedite “time-to-market” (learn and share).
Do not lose site of the ultimate goal: Align your business to appropriate BI and analytical solutions.
This article is published as part of the IDG Contributor Network. Want to Join?