I Want To Do Predictive Analytics – But How Can I Convince My Boss?
In today’s data-driven business environment, predictive analytics is an essential tool for companies looking to keep or gain a competitive edge.
You can think of a company that doesn’t use predictive analytics as a boxer who enters the ring wearing a blindfold. They don’t have a clear vision of what their customers want, how to optimize their operations, or how to stay ahead of industry trends.
In contrast, a company using predictive analytics has the advantage of insights and accurate forecasts, which allows them to make strategic decisions, maximize revenue, and outmaneuver the competition.
In spite of all that, convincing your boss to invest in and adopt a predictive analytics platform can be a daunting task, especially if they’re not familiar with the technology or its benefits.
As a data analyst facing that challenge, you’ll need to be able to explain what predictive analytics software is, and why it’s important. To help you with the first part, we recently published “How to Get Your Boss on Board With Machine Learning.”
To answer the “why,” you’ll want to consider potential use cases for predictive analytics and the impact they can have on your business. We’ll expand on this further in Step 3 below.
In this blog post, we’ll explore how you can strategically approach your manager and show them how investing in predictive analytics software is a worthwhile decision for your business.
Step 1: Identify the decision-maker
As you might expect, your first task is to identify the decision maker who needs to be convinced of the value of predictive analytics. If you’re integrated into a business team like marketing or sales or distribution, that person will typically be your boss.
But if you’re working in a separate department (like a data center of excellence), that person may be an external department head, such as a VP of Sales, Chief Marketing Officer, or Supply Chain Manager.
And in some cases, it will simply be a person that your boss (or another senior manager) takes advice from.
Let’s say you’re an analyst who wants to use predictive analytics to optimize your company’s massive online marketing spend. In this case, you may want to pitch the idea to a leader of performance marketing, growth initiatives, or customer acquisitiont, who will have no trouble visualizing and calculating the financial impact of better campaign ROI.
With your assistance, this leader will also be in a better position to pitch predictive analytics to the VP of Marketing, who might only have a basic understanding of campaign management – and thus trusts the expertise and judgment of the hired expert.
In cases where the impact is organization-wide or your success depends on the partnership of multiple stakeholders, you may need to cast a wider net or go further up the chain of command. For example: If you want to use predictive modeling as a way to optimize call-center activity and increase customer conversions, you’ll probably need to align with multiple managers from sales, marketing, and operations.
Get ready to show off your new predictive analytics capabilities.
Start a free trial →
Step 2: Understand the decision-maker
Business leaders have a lot of priorities to deal with, and as mentioned in our post about building a data-driven culture, predictive analytics is typically not at the top of that list. They are generally focused on high-stakes issues such as finances, operations, and product development. In addition, many leaders have an inadequate understanding of data and how it can be used to drive business decisions.
If you’re a data analyst looking to convince your boss (or another decision-maker) to adopt predictive analytics, one of the most important things you can do is understand that person’s priorities. In particular, you should be able to identify their most important key performance indicators (KPIs) and demonstrate how predictive analytics will address them.
In the marketing doman, their KPIs might include metrics like customer acquisition cost, customer lifetime value, and campaign performance. In the domain of customer experience, key KPIs might be time on platform, NPS score and customer retention.
But don’t worry – you don’t need to become an expert overnight. By scheduling interviews with relevant leaders or team members, you can get a better understanding of these KPIs and how predictive analytics can help improve them.
Once you have this information, you can determine the relevant use cases and show these leaders how predictive analytics will help them make data-driven decisions – instead of relying on gut feelings, business assumptions, and overly general best practices. We’ll explore an example in the next section.
Step 3: Build the business case
Among the key benefits of predictive analytics are improved decision-making and better business outcomes (a.k.a. money!) But it’s one thing to tell, and another thing to show. You’ll need to demonstrate how predictive analytics can provide actionable insights into customer behavior and other factors that drive business success.
To provide you with a framework for building your own business case, let’s use an example. Imagine you’re an analyst who wants to pursue predictive analytics as a more accurate and efficient way to improve forecasting for customer service needs. These are the steps you might take with your boss:
- Identify and articulate the problem. In this scenario, the company may be struggling to allocate service staff and equipment effectively to different locations. This is resulting in increased costs, inefficiencies, and reduced customer satisfaction. Part of the problem is that the company is relying on simple trends from historical data to make business decisions, which is proving to be inaccurate and inefficient.
- Highlight the benefits of predictive analytics. Your next step is to demonstrate how predictive analytics software can address these issues and improve the forecasting of customers’ service needs.
Using the current example, you could plan for a predictive model that analyzes a variety of data sources (such as historical call volume,customer demographics, and even weather). This model can leverage a massive volume of data to identify patterns and trends that would not be visible through traditional methods or simple hand-crafted statistical models. This would help the company anticipate customer call volume more accurately, leading to better resource allocation and improved customer satisfaction.
- Provide ROI projections. Crunching some numbers will allow you to provide a clear estimate for the return on investment (ROI) your company can expect to achieve by using predictive analytics software. This will typically include metrics like revenue, customer satisfaction, churn, inventory write-offs, call-center performance, customer acquisition costs, and so on.
In the case of the Canadian Automobile Association (a travel, insurance and roadside assistance organization), adopting predictive analytics led to a 30% reduction in time spent generating forecasts, optimized scheduling across all service facilities, and improved customer service and satisfaction – all without additional data-science resources.
- Create a detailed proposal. The last thing you may want to do is create a proposal that outlines the problem, the limitations of your current approach, the benefits of predictive analytics software, and the projected ROI. This proposal could also include a plan for implementing the software, training staff, and measuring success over time – although you won’t need to do much of that when using Pecan.
If you’ve done a good job up until here, it shouldn’t be much of a hard sell. But if you’ve got a lot of red tape to cut through, or want to take a more formal approach, you might want to check out the article “How to Build a Business Case for Your Software.”
Be ready to discuss every aspect of your new predictive strategy — and demonstrate how you'll bring real impact.
Sign up for a free trial of our low-code platform today.
Step 4: Emphasize ease of adoption
The next point may be our last, but it certainly is not “least.” You may actually want to raise it quite early in the conversation with your boss. Using predictive analytics software doesn’t require the efforts of a whole team, or even of a highly trained data scientist. All you need is you – a data analyst with a desire to make a significant impact in your role.
To begin using predictive analytics software, all you should need to know is what data you have, and where it’s stored. Pecan provides native connectors so you can easily connect your data sources to our machine learning algorithms. And your raw data is all we need – Pecan automatically cleanses that data, builds and evaluates features for high-performing machine learning models, and provides SQL templates for each business use case. All you have to do is query your data to fuel the type of model you want to create.
Your data is automatically fed into Pecan’s AutoML system, which is where all the magic happens – from model building to algorithm selection to model tuning, deployment and monitoring. This saves tremendous time and energy, and overcomes many of the most common stumbling blocks for organizations that want to adopt predictive analytics.
What your boss may be most impressed with is that you can export predictions directly into your existing business systems, such as your CRM, ERP, or BI tools. With accurate predictions integrated alongside current data, your colleagues can instantly start making better business decisions without changing the way they work.
On top of that, Pecan’s dashboard clearly communicates the performance of your model, both metrics-wise and in comparison to benchmark models, so you can easily assess business impact. It also reveals the factors that contribute to a model’s prediction, and to what degree – details that you can use to inform business strategy and further optimize the model.
Altogether, this makes it super easy to communicate the value of predictive analytics and the impact of your predictions. And once your first model is up and running, you can even show your data-savvy colleagues how they can tweak existing models and start experimenting with models of their own to address still more business challenges.
Ready to share the love?
By identifying the decision-makers in your company, understanding their priorities and KPIs, and demonstrating the benefits of predictive analytics, you can make a compelling business case for adopting predictive analytics. And once your leaders start seeing results, they’ll become strong advocates for this powerful technology.