How can you get more value from your data, more quickly, more often?

Most organisations have never had as much data as they do today, or needed to make more decisions at speed. The world of the data analyst typically revolves around reports and metrics. However, organisations that want to gain competitive advantage should expect analysts to do much more. So, how do you ensure your data has maximum impact on business outcomes?

The Adaptavist Group (TAG) comprises a relatively diverse range of companies. Perhaps the simplest view of what we do is that we either deliver services (such as digital transformation or migration), deploy our own products (add-ons and integrations that enhance our clients' platforms and tools), or manage clients' use of platforms (such as Atlassian Jira, monday.com, and GitHub).

Let's take our product business as an example. We have approximately 30 apps on the Atlassian Marketplace. The data team supports the product teams by collecting all data generated by the apps and providing insights into user behaviour, adoption of new features, conversions, and other user journeys. We're tasked with distilling usage and activity to measure, monitor and, most importantly, understand how effective our products are. How do people use them? What do they do before and after key 'events' like using particular features?

Going beyond reporting

There are, of course, typical reports and analyses that teams want to see, such as product usage, regional sales analysis for planning and budgeting purposes, marketing effectiveness, and resource utilisation. Doing this at TAG also means looking through multiple lenses—professional services and solutions, products, license resale, and partner channels. Where data analysts can add the most value is by exploring the company's data and understanding its meaning. Ultimately, there is limited value in just producing reports. Our focus is on answering questions and illuminating the rest of the organisation. To do that, we must understand the data’s context and be rigorous in how we analyse it.

How do we take advantage of data?

Without clarity on the meaning and context, it's hard to be sure that we're producing helpful answers. It's easy to create reports, but are they useful? Do they help to initiate change and improvement?

Two fundamental challenges tend to prevent organisations from taking advantage of their data: understanding it and trusting it. Why? Well, suppose data and analysis are provided without appropriate context or explanation. In that case, people will struggle to trust what it is telling them. Similarly, if the data is inconsistent or incorrect, confidence is quickly lost.

It's easy to take data at face value. People often place much more faith in numbers, especially compared to qualitative information. It often feels more robust, more factual, though it isn't necessarily the case. It's essential to consider how the data was captured and whether the definitions used are consistent across the organisation. A simple example is how people in different teams might understand the word 'revenue'. Are we sure that we define it consistently everywhere the term is used? It's like the different ways 'revenue' is used are correct in a particular context, but without that context mistakes and misunderstandings will arise.

Can we trust the data?

Being able to provide justification for your findings is essential. Decisions are increasingly made based on data and its analysis, so analysts need to be able to provide context, establish the data's provenance and share that understanding.

Variance in data can be both accidental and deliberate, arising from input errors or changes to how it’s captured. The downstream impact of such changes can be dramatic. Analysts, we need to be acutely aware of this possibility and check for such errors. In fact, I would argue that data analysts have a responsibility to work with the data creators to help remove such inconsistencies wherever possible. The people using your analysis are likely to need to present, explain and justify the conclusions and decisions they make based on the data. If we don't have confidence in it, how can anyone else?

What traits do good data analysts share?

In addition to providing answers, data professionals should take responsibility for building trust and understanding of data within their organisation. They should be prepared to evaluate and interrogate the data as thoroughly, if not more so, than anyone else. When findings unexpectedly change from month to month or year to year, without explanation, confidence in the data decreases.

In addition to identifying the reasons why, you should have a role to play in helping to improve data quality, establishing clear ownership of the raw data, and implementing versioning, and documenting any assets that depend on the data. Definitions need to be clear and precise, and we must be careful not to select data that fits a hypothesis rather than discovering what the data is truly telling us.

In my view, data analysts should possess some of these traits.

Endless curiosity and dogged attention to detail. Data analysis is all about answering questions and discovering what's happening. Analysts should explore odd patterns and ask (themselves and others) 'why' until the answer is clear, sharing and collaborating around their findings to help remove or highlight inconsistencies and inaccuracies.

They take ownership. If the data looks wrong, they investigate. They make it their mission to ensure anomalies are understood, exploring how the data is collected, fixing the pipelines, closing gaps, and documenting their findings.

Comfortable with ambiguity. Despite being focused on accuracy and data quality, good data analysts are pragmatic and remember that they're there to answer questions. If the data is incomplete, they make the best call they can and explain their reasoning.

Context is key. It pays to understand the rhythms, models and definitions of the organisation you're working in. Context has a significant impact on how data is interpreted and helps you pick the right place to look for answers.

Clear purpose and people skills. Stay focused on answering questions of the organisation and help the business to access the findings. If the brief isn't clear, ask what is required so your recommendations can be valid and valuable. As I often say, "SQL is teachable, empathy is harder."

Don't skimp on presentation. How people receive information matters. Make it crisp, clear and consistently branded. It makes it easier to digest and builds trust in the output. If people have confidence in your work, they'll be more confident about taking action as a result.

How can we level up the use of data?

Good data analysts become trusted advisors to their organisations. They combine context, curiosity and clarity to help make key decisions. I know we're making an impact when leaders in the business want an analyst in the room when they're making decisions.

In terms of development goals, I want to bring more proactive insights to the teams at The Adaptavist Group. If we can share what is changing, how fast and in what direction, and why it matters, we'll be helping the business to evolve, pivot and optimise. We're implementing automations and dashboards for established lines of questioning, improving our ability to deep-dive into complex business questions. We work ever-more-closely with the teams that create the data, from CRM and product development, to HR and finance.

Ultimately, data analysis is about making decisions, and the pace of change means only one thing: making more decisions at a greater speed. However, without accuracy and understanding, organisations will struggle to make the correct calls.

Want to continue the discussion?

Connect with me on LinkedIn to chat about how organisations can unlock more value from their data.

Let's talk