Skip to content
feelingstream
//Article

30 August 2022/Terje Ennomäe

From data to actionable insights: a practical guide

what are actionable insights and how to find them

Large organisations trying to improve efficiency, quality or sales are rarely short of data. They record calls, keep every email and chat, and run NPS and CSAT surveys. What they are short of is a reliable way to turn all that data into decisions. The journey from data to actionable insight is not a straight line — and skipping steps is why so many decisions still come down to gut feeling.

A better route is to start with the business strategy, ask the questions that strategy raises, then gather and analyse the right data until you reach an insight you can actually act on. A knowledge pyramid is a useful way to picture that climb.

Why isn't data enough on its own?

At the base of the pyramid sits data: a collection of recorded events. Every recorded call, email, chat and survey response is data. Organisations often have enormous quantities of it, and have spent heavily to collect it.

But data by itself creates no value. It is not having the data that pays off — it is using it well. And it is worth separating two ideas that often get conflated: an insight is something you have noticed; an actionable insight is something you can do something about. If you cannot act on an insight, it is of limited use. The aim is actionable insight that moves your strategic goals.

What are the levels of the knowledge pyramid?

The clearest way to see the climb is with a running example: using NPS (Net Promoter Score) feedback to find something worth acting on.

  • Data — the base layer. All your recorded customer communications. Valuable only once you do something with them.
  • Information — data processed in context. You collect NPS responses and work out your average score. Knowing the average is a step beyond simply holding the raw data.
  • Knowledge — information tied to the problem at hand. You read the free-text comments behind the scores and learn why customers rate you as they do — which processes, gaps or agents drive low scores, and what earns high ones.
  • Wisdom — the judgement to know whether your knowledge is sufficient. If you have analysed 10 of 100 negative surveys, can you safely apply that to all 100? Probably not. Wisdom is recognising the sample is too small and choosing to analyse more — ideally 100% — before deciding.

Each level adds context and confidence. The mistake is to leap from "information" (the average score) straight to a decision, without the knowledge and judgement in between.

How do you reach an insight?

Insight is the deepest level of knowing on the pyramid. Continuing the example: once you have analysed all the negative NPS results and found that, say, 20% of them stem from one specific cause — customers feeling misinformed about the terms of a service — you have found a pattern that drives low scores. That is an insight.

From there you weigh it up. If the pain looks small, you might simply monitor it — watching whether more low scores arrive, and how many calls and chats mention the same terms, to size the issue. With your strategy and the size of the pain in front of you, you can judge how much the insight is worth.

What makes an insight actionable?

Actionable insight — the top of the pyramid — is an insight that drives action, and it is typically more valuable than one that merely answers the question. Combine two findings from the example:

  • "20% of our low NPS scores are caused by customers feeling misinformed about the terms of our service."
  • "We receive around 100 calls a day about the terms of this service; they are unclear to customers."

Together these point to a clear action: review the questions customers ask, find the patterns (information missing from the website, or agents giving inconsistent answers), then update the information and coach the agents. Crucially, the work is not done at the decision. Once the change is live, you monitor future conversations and NPS results to confirm the issue is shrinking. If it barely moves, the action was insufficient and needs revisiting.

That closes the loop — from holding NPS data and an average score, to using the data at hand to make a change that measurably improves the customer experience.

Why analyse 100% of feedback?

The pyramid only works if your sample is trustworthy. Analysing 10% of negative surveys gives you a hypothesis; analysing 100% gives you confidence that the pattern is real. A conversation analytics tool lets you find patterns across all your feedback and conversations — not a hand-picked slice — which is what turns tentative knowledge into something solid enough to act on. Pairing that with clear strategic goals and well-framed business questions is the crucial part of the process.

Frequently asked questions

What is an actionable insight?

An actionable insight identifies a problem and points to a specific action you can take. It goes beyond simply answering a question — it tells you what to do and why, in service of a strategic goal.

How is an insight different from an actionable insight?

An insight is a pattern or finding you have noticed. It becomes actionable only when you can do something about it and that action supports your goals; an insight you cannot act on has limited value.

How does the knowledge pyramid work?

It moves from data, to information (data in context), to knowledge (tied to the problem), to wisdom (judging whether the knowledge is enough), and finally to insight and actionable insight — each level adding context and confidence.

Why analyse all customer feedback rather than a sample?

A small sample gives a hypothesis, not proof. Analysing 100% of feedback and conversations confirms whether a pattern is real and large enough to justify action, which is what turns knowledge into confident decisions.

Where to go next

Ready to turn the data you already collect into actionable insight? Book a demo and we will show you how on your own customer feedback and conversations.