Three Steps to Planning a Successful Business Intelligence Project

Three Steps to Planning a Successful Business Intelligence Project

Companies that try to launch a new business intelligence (BI) initiative with the sole goal of quick wins and fixes forget one of the most important steps in the process: You must align the needs and expectations of the BI professionals, business executives, and what the organization aims to achieve. Guest article by Hila Kantor, BI Consulting Team Leader at Sisense BI professionals should take just a day or two before plunging into schemas, graphs, and KPI dashboards, to gather their users and pinpoint the most relevant KPIs for the project at hand. Remember: The metrics need to fit the organization, and not the other way around. So keep in mind that the goal of BI should always start with the business and serve the business, and all metrics have to be flexible enough to accommodate the BI solution that the business actually needs, rather than the one that’s easiest for an analyst to create. Step 1: Meet with Stakeholders, Make a List of Business Requirements Time to completion: 1-2 days Deliverables: Summary of business requirements Written specifications are important, but don’t proceed to the next step until you’ve had meetings with all relevant executives, managers, and business users who will look at the data and dashboards on a daily basis. This will allow BI professionals to hear from the horse’s mouth exactly what it is their users expect to measure. Remember, people can only improve what they measure, so if an executive plans on improving sales by quarter, but doesn’t have that metric set as a KPI, there will be a disconnect at the end of your BI project. If users need help narrowing down their...
Unleashing the Power of Analytics in the Cloud

Unleashing the Power of Analytics in the Cloud

With low or no IT footprint, minimal or zero up-front capital costs, and ubiquitous access to the tools, a cloud-based approach to analytics can be enticing. However, as some find out the hard way, extracting the potential of the cloud is not a “plug-and-play” proposition. Companies able to see results with cloud analytics have a measured approach that combines organizational capability with the right data-driven philosophy. Aberdeen Group’s recent report demonstrates that cloud users have a commitment to analytics beyond their counterparts taking an on-premise only approach to business intelligence (BI). These characteristics include the following. Strong support from senior management Aberdeen’s July 2015 cloud BI report demonstrated that those taking a cloud approach were able to distribute analytical capability to a wide variety of business functions. From finance and operations to sales and marketing, cloud users enjoyed a more pervasive deployment of analytics. This pervasiveness would not be possible without the authority and cloud to sidestep cross-functional red tape, and cloud users are able to garner the executive support needed to make this a reality. Heightened analytical mindset BI technology has evolved into many different shapes and sizes, closely matching the needs of a very diverse workforce. However, Best-in-Class usage of analytics has always rested upon factors beyond the speeds and feeds of the technology. Top companies build adoption and engagement, partly by delivering the right tools to the workforce, but also by fostering an environment of analytical skill through the use of training and other talent development programs. Expanded usage of data So many “born-in-the-cloud” or mobile applications and data sources are purpose-built to handle a diversity...
Three Reasons Why Your Users Should Care about IoT data

Three Reasons Why Your Users Should Care about IoT data

Data is data, right? It varies in quality and origin, but it all winds up in my dashboard. It couldn’t have a material impact on the way I make decisions…right? At a typical data-driven company, the decision process is fickle and fast-changing. Technology plays an important role in the way we transform data into insight, as does organizational maturity. Companies that have the right processes and internal capabilities in place are simply better positioned to exploit the potential of analytics and make better decisions. However, Aberdeen Group’s research suggests that the type of data used for analysis can also impact the decision process and boost user satisfaction. According to recent research exploring companies that use Internet of Things (IoT) data frequently, there are several factors of decision-making that strongly correlate with the usage of this type of information. 1. Timely information Research shows that IoT shops have a greater urgency for information, yet they’re more likely to get it on time. The volume and constant propagation of IoT data can certainly create challenges for companies, but it also creates opportunities to get real-time data into the hands of users faster. 2. Data quality To a substantial degree, issues with data quality are born out of human error. Erroneous entries, missing fields, multiple versions of the same data, etc. Due to its nature of being machine-generated, IoT data is less prone to these types of quality issues and Aberdeen’s survey respondents validate that claim. 3. Analytical firepower Any company using IoT data frequently is most likely engaged in analytical activities beyond just simple static reports. When handled properly, this type of...
Alteryx Inspire 2016: Is There a Cure for Data Hate?

Alteryx Inspire 2016: Is There a Cure for Data Hate?

Returning from Inspire 2016 –the annual user conference from Alteryx – as I write this, I can’t get away from one burning question rattling around my brain: Why do people hate their data so much? Before delving into that one, let’s provide a little background here. Alteryx likes to distinguish itself as a “platform” –as opposed to a point solution – for self-service data analytics. Including capabilities for data blending (some have called it integration), data preparation, enrichment, and predictive/advanced analytics, the solution is geared primarily toward data scientists and business analysts, especially those that deftly balance a modicum of technical acumen with a healthy dose of business expertise. Accordingly, the Inspire conference is chock full of folks whose primary job function involves extracting meaningful insight from data, a task monumentally easier said than done. Through casual conversations, and after listening to several “before and after” style customer accounts, a common thread that emerged was a deep dissatisfaction with the raw material used to create insight. So, why do people hate their data? There are several reasons (in no particular order): It’s gross. Riddled with corrupted, duplicated, incorrect, or absent fields, the typical data source has major quality issues. To make matters worse, these issues arise regardless of data source, be it an application, an operational data store, a data warehouse, or just a spreadsheet. You need the Keymaster of Gozer just to find it. In the heat of analysis, with all the brain-burning and number-crunching, how often does someone need data that isn’t immediately at their fingertips? All. The. Bleeping. Time. For many of these poor souls, adding a field to a data set, or, God forbid,...
Three Key Criteria When Selecting Data Visualization Tools

Three Key Criteria When Selecting Data Visualization Tools

So how many rows of data did you say this tool can ingest? It’s backed by a massively parallel processing engine? It runs in-memory? And…why should I care about any of this? If flexibility, firepower, and adaptability were the only considerations for technology implementation, these decisions would be easier. The fickle business user community, though, needs more convincing than raw numbers on paper. When it comes to tools like interactive visualization, for those that survive or thrive based on business user adoption, connecting with a broader community is absolutely vital. Recent Aberdeen research explored the impact of data visualization and found that, in addition to greater analytical engagement, these users shared three common characteristics of satisfaction. 1: Ease-of-use Interactive visualization is all about exploring the data behind the data — the “why?” behind what is presented. All too often though, the tools don’t resonate with the business users typically asking those questions. Ease-of-use can be difficult to describe, and varies from company to company, but the general concepts apply broadly. Users need intuitive, drag-and-drop, easy drill-down capabilities in order to explore the data. 2: Data connectivity Nothing kills the momentum of analytical activity like hitting the invisible wall of data absence. In the heat of an analysis, users need the ability to pull in data from sources not necessarily presented to them in the original dashboard or visualization. The ability to connect to, and ingest information from, other sources is a key enabling factor of interactive visualization. 3: Line-of-business fit Closely tied to ease-of-use, users need tools that fit the logic and taxonomy of their business area. Sometimes that involves...
Page 2 of 1912345...10...Last »