Think about every time you’ve heard a product boast about “machine learning”, “AI”, “cloud analytics”, or “big data”. “Actionable insights” are just one in a long list of tech phrases that get thrown into the marketing-jargon mix. But is there something important hidden under the surface?
Like every buzzword, there is a reason for the term. That doesn’t mean that every product lauding “actionable insights” actually delivers. But it’s important to understand why a term was developed in order to look for meaningful outcomes that can change how you do business.
“Actionable insights” describes contextualised data analysis. It’s a piece of information that can actually be put into action — supported by the analysis, context and communication tools required to get the job done.
Here, we are going to look at this definition in more detail and explore what actionable insights actually means. We want to help you focus on the right thing — precisely what actionable insights are supposed to deliver.
What are actionable insights?
Over the last five years, most companies have taken steps to go from being data-aware to data-driven. They collect operational and transactional data sets from applications; process the data into data lakes, data warehouses or data marts; and use analytics tools and business intelligence (BI) to understand what the data is telling them.
The process described above is being “data-driven”. But are these enterprises transforming the signals they get from the data into actionable insights?
Information can be thought of on three different levels:
- Data: the component pieces.
- Analytics: sorting data into meaningful groups — letting you start to make decisions.
- Insights: contextualising analytics with real-world information — letting you use analysis to take actions and predict outcomes.
Actionable insights hone in on the outcome portion of “insights” and take it even further. Actionable insights:
- Focus the right people on the right insights.
- Include exploration and problem definitions — they aren’t just about decision-making.
- Are supported by communication channels that help guide outcomes.
Actionable insights must be:
- Practical: what to do, rather than explaining or understanding.
- Conditional: purposeful and thorough, focused on what happens when something occurs.
- Verifiable: qualitative and quantitative, focused on outcomes over speculation.
Fundamentally, actionable insights need to help you transform data analysis into outcomes. That’s why they are useful and should be considered the defining characteristic.
When it’s a buzzword and when it’s not
The problem with defining any buzzword is that over-use will reduce the power of the phrase. So now that we’ve defined actionable insights, let’s look at how to use it.
So far, we have been technology-free in our definitions, and it’s worth making the point that “actionable insights” does not equate to expensive IT. You need data storage and software able to guide analysis and present insights. But, at its core, going from data collection to action does not need to be complicated — or driven by technology. It’s simply that the complexity of that task scales with the amount of data used, so it’s very difficult to achieve manually.
Critically, actionable insights are not “data-driven decision making”. It’s all about the presentation of data in a way that makes taking action easy. Fundamentally, actionable insights are about focusing on the right information at the right time. It’s less about how to process data and more about supporting the human activities of planning and doing.
Incorporating these human activities is when the understanding of goals and KPIs become important. From this perspective, there are two things to keep in mind:
- An “insight” should represent an uncovering of significant potential to alter the outcome of a course of action.
- Insights are “actionable” because you can do something about it.
Actionable insights require business intelligence tools. But you really want to create a communication, action and data generating feedback loop that can show, measure and drive success.
An example of actionable insights
The prerequisites
Creating actionable insights isn’t just a data collection and presentation challenge. It’s more of a change management problem. You need to:
- Identify a change that will move the needle on your KPIs
- Pilot, research and instigate the change
- Measure the change
- Communicate constantly
- Improve
- Rinse and repeat.
The challenges are many — here are some typical real-world examples:
- Multiple data sources and lots of KPI data
- Distributed performance data — often spreadsheet-based
- No central place to store KPIs
- Lack of clarity on where to focus
- Delays and issues with data supply
- Lack of confidence in data accuracy.
Strategies to help: Centralisation of data and cloud capabilities are now making it possible to see not only what’s happening across a company, but also what you can do about it. Some of these prerequisites will need to be identified by hand. However, the more data that you can get into the cloud the easier it becomes to manage and share.
The process
Setting up your KPIs
Picking the right KPIs is important, and will provide the focus you need. The risk is using the data to reinforce long-held views, pet projects or the status quo. KPIs, once agreed and validated, have to be reinforced, clearly displayed within the organisation, and provide a visual representation of how you are doing.
At a glance, decision-makers need to understand performance, trends and what’s required — drilling down into the data as required.
Typical KPIs can be:
- Sales revenue: How much revenue does a new product or marketing campaign generate?
- Customer lifetime value: Which customers generate the most revenue and must be retained at all costs?
- Cost per lead: where is spending wasted in lead generation?
- Organic traffic: How many new leads are coming to you through online searches?
- Customer success: Are customers seeing the benefits? This is often represented by surveys or NPS scores.
Strategies to help: Your KPIs should align with your business goals and priorities. Ultimately, this is only a decision that you can make. However, the tools you deploy should help you execute and focus your strategy. For example, KPI dashboards within a business intelligence tool are critical to tracking success and communicating priorities.
Creating actions as part of an action plan
Performance improves through targeted actions and activity. For example:
- Managers and field teams create an action
- The action is allocated to team members
- Reminders are set
- The action is closed once complete
- Move on to the next
This focus on action invariably sees rapid performance improvement. Easily sharing that action with other areas of the business breaks down silos, creates and shares best practice.
Strategies to help: Some good points to keep in mind are:
- Setting up Scorecard/focus areas: Consistency of performance across your business or network is an important goal. A balanced scorecard with clear rankings by KPI and overall performance helps achieve this. For each team, you need to highlight the areas giving the best chance of increasing performance.
- Centralised planning for field teams: If you have distributed field teams, you need to provide the feedback to plan, manage and follow up. Ensure any customer or sales contact is efficient and productive with no misunderstanding.
- Ensuring consistency: Customers want consistency wherever they experience your brand, anywhere in the world. Audits become a vital part of the process. Assessing and maintaining standards can give visibility of compliance across your business. You can consider using standardised assessment templates or custom assessments completed centrally. Specific action plans can be generated from each assessment and KPIs created from assessment scores.
- Reporting/data analysis: Reporting functions need to tackle both searchability and recovery. Scalable data storage and reporting should expand as your company grows while still providing the tools to interrogate and find recent versions instantly.
Action is the difference between data and outcomes
The main problem we see in business intelligence is the gap between analysis and action. Filling this void was a central purpose of our own tool — Loop. We’ve never been obsessed with the term “actionable insights”. But, realistically, Loop captures what this term is supposed to look like in action, and is a good example of a piece of software that puts “turning data into actions” at the centre of what it does.
If you want to learn more, get in touch. The whole system has been built around focusing your attention on the right things, and partnering communication tools with analysis software to ensure the right actions are actually taken and then tracked.
However, as we have shown, actionable insights are as much about process and strategy as the technology itself. Technology only makes the outcome easier. This is a big reason why “actionable insights” really is a buzzword. Taking “action” comes down to you. The key is to look for tools that will help you (and your process) actually take effective steps to transform data into outcomes.
With that said, the future of “big data” has as much to do with data presentation as data volumes. It’s far too easy to get lost in a sea of information. Business intelligence tools that focus you on relevant and contextual analysis are critical to managing the current economy — and are an important business intelligence trend. When deployed correctly, “actionable insights” describe that exact set of features. However, you need to analyse the validity of that claim on a tool-by-tool basis — just like you need to look at actionable insights and then actually take action.
