How to Create a Data Analysis Strategy With Impact

Many leaders of companies owning a ton of GB of data face a similar challenge:

“We’re having all this data, how can we get insights out of it?”

Is a common complaint I’m hearing among business leaders. If you’re also one of them, answering key strategic questions can be frustrating.

“How can we make users return to our app?”
“What client segment has the highest growth?”
“How can we improve customer satisfaction with our brand?”

Finding answers to these questions becomes impossible. Either data is not accessible or you’re lacking the specialized staff to analyze it.

To make customer data more accessible, companies decide to automate their reporting, using so-called business intelligence or data analytics solutions.

If you’re one of the leaders managing data analytics in a company that it’s just starting out on this path, prepare for some challenges ahead:

Challenge 1: Manage communication with engineers

You’re hearing What exactly do you need? every time you present an idea for a new reporting solution. If you don’t have solid answers to technical questions, you may not convince the engineers that the project is viable. Hence, your reporting idea does not become reality because it is not backed by a solid implementation plan.

Challenge 2: Make sure the reports will actually be used

Otherwise you spend $$$ on reports no one uses. If you don’t have a plan on how employees will adopt the reports in their workflow, it’s gonna be hard to gain the trust of the sponsor and receive funding.

I want to show you how to overcome these challenges and build a data analysis strategy in 4 steps.

Data Analysis Strategy

1. High level tasks
2. Desired business outcomes
3. Strategic goal
4. Metrics

Let’s break down all these steps:

Step 1: High Level Tasks

High level tasks are what employees are currently doing – or aim to do – while working with data.

To uncover these tasks, ask:

What are you trying to do with data?

When the answer leads to a clear business task that can be explained in a few words, note it down.

For example, let’s say you create a data analysis strategy for a company that owns a platform helping freelancers manage their business. 

You’re going to everyone from the company working with data generated by the platform and ask:

What are you trying to do with data?

It may lead to answers like:
Track engagement of videos from the platform
Review visits to the customer relationship module

After you have written down several such answers, the next question should be Why?

Let’s see where it leads you.

Step 2: Desired Business Outcomes

Everyone is not wasting their time in vain.

Asking Why? helps you identify what they’re hoping to achieve.

What are you trying to do with data?

Consider the example of the company helping freelancers manage their business.

You’re asking:
Why are you tracking the engagement of videos?
Why are you reviewing the visits to the customer relationship module? 

You find out that:
We’re aiming to grow adoption & quality of video series
We want to identify the freelancers using the customer relationship module

Are these really the ultimate goals? The goals that once achieved, you can open the champagne?

Step 3: Strategic Goal

To create a data analysis strategy with impact, you need to elevate 10 feet over what everybody is doing and discover the single, most powerful goal of the teams. It helps you create reports geared towards making decisions that support the overall direction of the company.

To uncover the strategic goal, ask:

If you achieve goal X, is this goal alone a success?

Consider the example of the company helping freelancers manage their business.

You’re asking:

If all videos from the platform are completed, is this goal alone a success?
If all freelancers are using the customer relationship module, is this goal alone a success?

The answers may lead to uncover the strategic goal of the company:

Helping freelancers grow their business

Having the strategic goal written down helps you keep the eye over the big picture. It helps create reports supporting decisions with impact. Otherwise, it will be hard to justify the ROI of the investments in these tools.

Something is still missing though:

Not sure how to measure all that – I’m hearing often from analytics leaders.

Step 4: Metrics

Data reports show … well … metrics. Some call them KPIs, some call them metrics. Metrics can be:

% of deals closed
or
Number of service centre tickets closed on time

You get the idea. A metric quantifies something.

Until now, after applying the steps 1 to 3, you uncovered the workflow of employees: What they’re doing with data, why and with what strategic goal. As an analytics leader, your task now is to decide how to measure all that. After all, it’s called data strategy for a reason, right?

Ask yourself: How do I measure X?

Consider the example of the company helping freelancers manage their business

How do I measure if a freelancer is using the customer relationship module?
How do I measure the growth of a freelancer’s business?

The way you quantify all that needs to be relevant & accurate. Otherwise you’re building castles on sand.

Let me show you what I mean.

An accurate metric

Consider measuring if a freelancer is using the customer relationship module.

How can you accurately identify those freelancers?

You can’t just count the users logging to the platform. They may simply watch the videos.

But if you, for example, count the users who clicked the Create Lead button, and track the click down to the user, then you’re into something. You accurately identified the freelancers who are managing leads.

An accurate metric assumes you have data to calculate it in a precise way.

A relevant metric

Consider measuring the growth of a freelancer’s business.

“If the revenue of freelancers increased after joining the program, we prove we helped them.”

Is revenue a relevant metric?

Freelancers could apply all learnings from the program and still not grow the revenue. Or, even if they would, it would be incorrect for the firm to take credit for that. Revenue depends on many other factors, not only on lead generation.

Instead, you may consider the number of leads per freelancer, before and after the freelancer joined the program.

A relevant metric is correlated with what you’re trying to measure

The Complete Data Analysis Strategy

With this strategy written down, you can tackle both challenges outlined at the beginning of the article:

Challenge 1: Manage communication with engineers

When engineers ask:
What exactly do you need?

You can respond:
I need these exact metrics:
The number of users who clicked the Create Lead button, tracked down to the user.
Number of leads per freelancer, before and after the freelancer joined the program.

In this way it will be easier to gain approval of the engineers for the feasibility of the project.

Challenge 2: Make sure the reports will actually be used

To avoid spending $$$ on tools no one uses, you can open discussions with the team who will be using the reports and gain their approval for your vision.

After you convince the engineering team and the users that your reporting idea is valuable, it will be easier to secure funding from the sponsor.

Summary

Use these questions to help you create data analysis strategy with impact, while getting the others share your big vision:

Step 1: High Value Tasks
What are you trying to do with data?

Step2: Desired Outcomes
What is the desired outcome of your tasks?

Step 3: Strategic Goal
If you achieve X, is this goal alone a success?

Step 4: Metrics
How do you measure X?