Self-Service Business Analytics -- A Real World IT Operationalization Journey
Thanks to new technologies, building a good business, product and project models are faster and easier than it used to be. What used to take six months to a year can now be given the right information take just a day, a week or a month.
But product development books attest that getting models deployed is a different story. True. So how long does it take? In the most services industry, it may from six months to five years. Worse is that there are many organisations that have models just sitting on a shelf in a lab, unused. Now we wonder, “How do you take models out of the lab and introduce them to the real world?”
Here are five tips our experts offered:
- Make sure everyone has the same goal in mind.
A lack of agreement on or definition of the goal is a problem. Ask your team: “Is everyone on the same page?”, “Are you working toward the same ends?” Collaboration is working together toward a goal. So Step 1 should make sure everyone has the same goal in mind. There are various tools and processes such as scrum agile software that are available to make sure that when goals are defined, you have actual agreement on them so you’re all aligned. The process of prototyping in and of itself is a way to validate that team members share a common goal, before investing in a lot of work in a miscommunicated project.
- Get solid executive support.
Your strategy needs to be very specific for example optimising a certain decision and it needs to be clear how you’re doing it and what the expected outcome is. Although you shouldn’t expect quick change, change takes time.
A lack of structure will bring about colossal failure. In all types of collaboration, it’s important to have a way to resolve conflicts. This may seem simple as having a facilitator in a brainstorming session, or having a clearly defined hierarchy of decision makers and tasks to be done. So when the structure isn’t defined or doesn’t exist, there is a high possibility of working with unaligned goals, timeline and cross-purposes.
- Prove that the model can be a better decision maker than the person.
Keep in mind that the model can recognise patterns better than people. People who are accustomed to having the power to make or change decisions don’t like giving up control. For example compliance, in a very heavily regulated market, there are companies who chose not to allow overrides because there were too many questions about how to make sure they were done consistently and fairly. So the use of models or matrices can take out the human subjectivity, and aids in helping objective decisions.
- Educate, and be educated by, the stakeholders.
You need to explain to investors or stakeholders that models are based on what has actually happened. And that you developed a model that can best predict what will happen next. For example, you create a model that considers the distance between the customer’s residence and the store. Why? Because fraud is likely to be involved if someone drives a long distance to deliver a product. Clearly, this is just one of the attributes of having a model, and it can help assess risks for the company.
Also, don’t hesitate to supplement your analytic processes with the valuable insights of longtime experts. Let the experts assess the chosen attributes for their opinions are vital simply because they’ve been doing things successfully for years. At the same time, it also helps to melt the ice, so to speak. And by including them in the process of decision-making and presentation of the plan and results to stakeholders, they’re more likely to understand the valuable role analytics can play in their jobs.
- Connect the right and left sides of the business brains.
Adding business analysts or data scientists to the team – people who are knowledgeable about the data as well as how the business works. For example, inbound marketers can explain to the statistician what the executive means, describe what needs to change and help extract the right information to build relevant marketing models and activities for the company.
In agile development books you can use, it shows that iterative methods reduce the risk of failure, compared to traditional, deterministic approaches to planning and estimating in product development simply don’t cut it in today’s dynamic, change-driven projects.
You can also use tools such as SAS who had created capabilities and that let you build models with user-friendly interfaces. Since the interfaces are easy to understand and use, it's easier to illustrate how your prototypes and models have improved for instance user engagement or ROI.
When you make agile model development implementation is an important element in strategic and tactical decision-making processes of your organisation. Those decisions can be made with more confidence and then you can focus on activities that further boost your bottom line – pricing, positioning, profitability and analytics.