If you have lots of data in lots of places, but aren’t putting those data sources together to learn how the business works, you’re wasting a lot of money.
It’s fine to know how much stuff you made last month, or how much you sold, or the characteristics or your customer base. But until you learn how those things relate to one another, you’re missing out on information that enables you to make your business bigger and more profitable.
You don’t need to take a big, expensive leap of faith to benefit from data analytics.
Building substantial value from data analytics is a lot like knitting a sweater; the big task is made up of many little steps. Organizations that have big successes start with small, low-risk projects and gradually increase the number and scale as they build knowledge and confidence.
If you’ve got a lot of reports that tell you what’s happening, but not what to do about it, it’s time to consider combining the data sources behind those reports.
If your company is big enough to collect data, it's big enough to benefit by integrating sources together. Integrating data sources lays groundwork for data analysis that reveals how the elements of your business relate to one another, and how action can influence the performance of your business.
Here are some examples of high-value initiatives from my own experience, and how each depends on combining data sources:
- Retail: Identify new customers likely to be high spenders, so marketing can be appropriately tailored to maximize customer value to the retailer. Requires integration of loyalty program data with marketing campaign data.
- Operations: Compare processing times for similar administrative tasks performed in different facilities to discover which are most efficient, and identify best practices that can be shared. Requires integration of records across facilities and business functions such as customer service and accounting.
- Finance: Test alternative debt collection methods to maximize returns for creditors. Requires integration of payment and contact records, and customer information, such as credit history and demographics, with controlled test plans.
The math and information technology practices used are not necessarily new; value lies in stepping away from the way you’ve always done things to try something that is new to you.
Data integration calls for careful consideration of technical and business issues.
This is often a source of conflict between data analysts and IT staff. Data analysts are often unaware of compliance issues. In contrast, IT staff may be very resistant to allowing data access out of concern about consequences of inappropriate data use. So honest conversation among all stakeholders from the start is a necessity.
Don’t start a project before you define a goal, reach out to all stakeholders, and carefully craft a plan. When the plan is written and shared, it’s much easier to spot the legal and technical challenges that affect your own projects in your own workplace. You’ve got to spot them before you can address them.
Most pitfalls can be avoided with thoughtful planning.
For starters, you must be careful to make sure that what you do is legal. This is a basic, yet ignorance and disregard for data privacy concerns is still a big problem.
Privacy laws and contractual agreements set boundaries on what you may do with data that pertains to your clients and partners, whether they are individuals or organizations. Before you touch any data, go over your plans with a data compliance professional to verify that what you intend to do is within the law and your company’s policies.
Don’t expect, as many managers do, to hand over the data analytics work to outsiders.
Outside help can be wonderful for helping you to implement a new tool, do a few starter projects to prove value, or help out when workloads are exceptionally high, but your own people are the greatest experts in your business and have the greatest commitment to your business.
Your internal staff knows more about the meaning and implications of your data than any outsider can. So train internal staff in analytics, and give them the resources they need to do the work.
That means paying for training, providing real data analysis software (no spreadsheets, period) and adequate computing resources. Yes, you’ll have to pay for those things, just as you pay for all the other things that produce value to your organization. You get the best value for your analytics investment when the staff knows what it’s doing and has the right tools for the job.
Most businesses still don’t make much use of analytics. That means every step forward you take with data analytics is an opportunity to build advantage over your competitors. What’s more, many analytics methods that are valuable for business have been around for decades, so you can use tested techniques and still be ahead of the competition.
The first move is asking a question that matters to your business, and sitting down with your people to make a plan for finding the answer.
