Seven Reasons Why Timely Analytics Is So Hard

Seven Reasons Why Timely Analytics Is So Hard

February 26, 2015 0 Comments

Rich Julius introduces 7 data preparation and data blending challenges that delay timely analytics.

The driving force behind business intelligence and analytics is a need for information. Every day, we ask questions such as, “How good are we at converting leads?” Or, “Which of our marketing campaigns is the most successful?” Finding answers to these seemingly simple questions, however, is much easier said than done. The people asking the questions just want answers, and often don’t understand why it takes so long to get them.

From Apples to Zombies: Seven Data Problems that Hold Up Analytics

Answers are found in data, and data is found everywhere. What’s worse is that data sources run the gamut of forms from relational databases to warehouses like Hadoop. Data scientists need to be able to use data in all its forms, whether it is as tasty as an apple, or as scary as a zombie. To do that, they must take time to pick it apart, blend it and analyze it before they can get insights. But by the time that is done, the project is running behind and the boss is upset. So what is the hold up?

A red delicious apple representing siloed data1. You’re comparing apples to zombies!

Business intelligence relies on data gathered from multiple systems, but you can’t just mix data from a CRM system with data from your marketing campaign data (not to mention your sales automation system and Google Analytics) and expect everything to make sense. What each of these systems store and the way they store it is so different that it’s hard to know what you’re looking at. To get answers, you need to find common keys between these systems. You might get away with comparing apples and oranges; at least they’re both fruit. But…zombies?

2. You have whole apples in one system and pureed in another

Have you ever tried blending data from a marketing campaign system with Google Analytics? Or cross-referenced Salesforce data with an ERP system? Or teased out the flow from call center logs to compare with other customer satisfaction metrics? If so, you know that the data stored in one system doesn’t always match the format of data in another, and throwing whole apples into a puree only makes a mess. This problem has led to a legion of data wranglers who are forced to spend time learning Excel tricks instead of doing real analysis. Most of the week, they manage string conversions and complex formulas and vlookups, so that on Friday they can finally do some analysis and achieve insights—if they aren’t zombies by then.

3. Zombies spell their names differently in each system

She’s Margaret in your ERP system, Molly in Salesforce, and Madge in your marketing campaign system (where her address is in Disneyland and her email is Your job? Determine her Lifetime Customer Value and likelihood of purchasing in the next 6 months. Good luck! Without a data quality solution, you don’t know what records match, what records are imaginary and whether Madge is even a current customer.

4. There are more apples and zombies than you can manage

Why do they call it “Big Data”? It’s enormous data! Your data changes constantly, and there is a lot of it coming at you from all directions. You need to slog through thousands or millions of rows of data before you can turn any of it into insight. The sheer magnitude of that task often makes Excel seem like a teaspoon approach to an oceanic problem, so you turn to IT and ask, “Can you build me a data warehouse?” By the time that’s done, however, the business has evolved, the questions have changed and the warehouse needs to be re-tooled.

Zombie data attachks

Preparing data all day can turn you into a zombie.

5. IT is defending against the zombie apocalypse

The pundits have proclaimed self-serve access to data for business intelligence as a key factor for business success. But how will the IT team open the doors to self-service access when their task is to defend the gates? And do line of business users and analysts want to contact IT every time they need data? What about the ever growing list of data sources? Moreover, data analysis is an iterative process of discovery, so you don’t always know what to ask for—initially.

This brings us to the next point…

6. Does an apple a day keep the zombies away?

How do you even know that you’re asking the right questions? Sure, you can establish some formulas for key performance indicators and you can come up with some clever questions, but they will only be relevant today. Business never rests, and the questions will keep coming. Business intelligence is the result of collaboration between line of business users, analysts and IT. You can’t ask all the right questions until you can see some of the data. Sharing, collaborating and iterating results in better questions and better insights.

7. It looks like pumpkin; is it fruit or is it the living dead?

One of the toughest challenges in business intelligence is figuring out what your data is saying. It’s hard to know what you’re looking for until you see it, and it’s hard to recognize what data to blend until you know how it fits together. It’s hard to paw through the guts of all the various system APIs, databases and Excel files that make up the data landscape. There has to be a way to more easily identify data so that it can be quickly integrated.

Zombie Repellant

Of course, there is a solution. Progress® Easyl® provides a self-serve data access, integration and blending platform that empowers all users, from line of business to IT, to move from data to dashboard with speed, reliability and security. It is collaborative, simple, powerful and cloud based, with no infrastructure to manage. Easyl is your perfect defense against the zombie data apocalypse, and the ultimate blender for those data apples. Best of all, you can get started with a free trial today!

Rich Julius

Rich is the Principal Product Manager for Progress Easyl, a cloud-based data integration and blending tool that enables you to create reporting data sets and data marts “on-the-fly” quickly and easily. Rich has spent two decades working with data-driven applications and marketing automation, and served in senior management roles at a number of Silicon Valley companies before settling in North Carolina. Rich has led consulting engagements for Fortune 1000 clients including Cisco, Microsoft, Seagate and McKesson, and is an alum of PeopleSoft, Oracle, and Informix. Rich was nominated in 2015 by President Barack Obama to serve on the Internal Revenue Service Oversight Board.

Comments are disabled in preview mode.
Latest Stories
in Your Inbox

Subscribe to get all the news, info and tutorials you need to build better business apps and sites

More From Progress
2020 Progress Data Connectivity Report
2020 Progress Data Connectivity Report
Read More
Getting Ahead of the Hybrid Data Curve
Read More
Creating Quick, Codeless Connectivity with Autonomous REST Connector
Read More