Applied Data Analytics for Impact

 Introduction Topic modelling can surface recurring themes in customer text, but on its own it often raises a bigger question: Which customers are actually driving these topics? Without segmentation, topic analysis treats all feedback as equal. A concern raised by a small, high value segment and a one off comment from an infrequent user appear side by side. That makes prioritisation difficult and insight shallow. The challenge is combining what customers say with who they are . Why this combination matters Customer segmentation already helps analysts understand differences in behaviour, value, and engagement. NLP adds context by explaining why those differences might exist. When topics and segments are analysed together: issues can be prioritised by segment importance messaging can be tailored more accurately engagement strategies become evidence based rather than anecdotal This combination turns text analysis into a decision support tool rather than a descrip...

Introduction Identifying topics in customer text is useful, but it’s only half the story. What most analysts really need to understand is how those topics change over time . A theme appearing once is interesting. A theme growing steadily over several months is actionable. Without tracking trends, topic modelling risks becoming a one off exercise rather than a decision support tool. Customer feedback is dynamic. Concerns, expectations, and language evolve as products, services, and external conditions change. If topic analysis is static: emerging issues are spotted too late resolved problems continue to receive attention teams struggle to prioritise what matters now Tracking topic trends introduces a temporal lens that allows analysts to distinguish between noise and meaningful change. Adding time to text analysis Once topics have been assigned to text records, the problem becomes familiar to most analysts. You now have: a topic label a timestamp optional ...

 Introduction  CRM systems don’t just store numbers. They store words. Open ended survey responses, feedback comments, support notes, and free text fields often contain the most honest signals about customer experience. Yet these fields are usually underused because they feel messy, subjective, and hard to analyse at scale. Many analysts either ignore text entirely or rely on manual tagging, which doesn’t scale and introduces bias. The challenge is turning unstructured customer text into structured, analysable insight . Problem explanation When customer text is left untouched: important issues remain hidden in long comment fields patterns are detected too late or anecdotally decision making relies on summaries instead of evidence Basic NLP techniques allow analysts to surface themes, track changes in sentiment or concerns over time, and complement quantitative metrics with qualitative context. This doesn’t require advanced machine learning. It requires disci...

Introduction CRM data is rarely stored in a single, analysis ready table. Customer details, interactions, transactions, and campaigns are usually split across multiple datasets, often with inconsistent keys and repeated records. As a result, analysts frequently encounter inflated metrics, broken joins, and confusing totals. These issues are not caused by SQL itself, but by how joins, aggregations, and deduplication are applied. Getting these fundamentals right is essential for trustworthy CRM analytics. Most CRM metrics depend on combining tables correctly: counting unique customers attributing interactions to campaigns summarising behaviour over time If joins are misaligned or duplicates are not handled deliberately, metrics quietly drift. Dashboards may look correct but tell the wrong story. Intermediate SQL skills allow analysts to express clear analytical intent , not just retrieve data. Thinking before writing SQL Before writing queries, I focus on three questi...

Introduction Many dashboards fail quietly because the transformation logic is scattered, manual, or undocumented. Analysts often clean data “just enough” to make visuals work, without designing transformations that are structured, repeatable, and auditable. In BI environments, this leads to fragile reports. A small schema change breaks refreshes. A new column introduces inconsistencies. Over time, confidence in the numbers erodes. Power Query sits exactly at this fault line between raw data and analytics. Power Query is not just a prep tool. It is an ETL layer embedded inside BI workflows . When transformations are well designed: data refreshes become predictable logic is transparent and reviewable models remain stable as data evolves downstream DAX stays simple When they are not, analysts compensate with complex measures and manual fixes, increasing technical debt. Structured transformations reduce that debt. Intermediate technical explanation: how to think abou...

 Introduction CRM datasets are rarely analysis ready. They often contain duplicated records, inconsistent text fields, missing values, and dates stored in multiple formats. While tools like Power BI and Excel can handle some cleaning, analysts frequently face a point where repeatable, scalable data preparation is required. This is where Python becomes essential. The challenge isn’t just cleaning data once. It’s designing a process that works reliably as new CRM data arrives. Poor data quality directly impacts: customer counts segmentation accuracy campaign performance metrics downstream modelling and forecasting If cleaning logic lives only in ad hoc steps or manual fixes, errors reappear quietly over time. A Python based approach allows analysts to formalise assumptions, document decisions, and reproduce results consistently. In CRM analytics, this reliability is foundational. Intermediate technical explanation: how to think about CRM data cleaning Before...

 Introduction CRM data is rarely clean by default. Records are entered by different teams, updated at different times, and stored across multiple tables that were never designed for analytics. As a result, analysts often struggle with inconsistent metrics, confusing filters, and dashboards that break as soon as requirements change. In most cases, the root issue isn’t the visuals or the calculations. It’s the data model underneath . Why this problem matters A poorly designed data model leads to: Double counted customers KPIs that change unexpectedly when filters are applied Complex DAX written just to “fix” modelling issues Dashboards that are hard to maintain or scale In CRM analytics, where insights often drive engagement strategy, segmentation, and forecasting, unreliable numbers quickly erode trust. A clean data model acts as the foundation that keeps analytics consistent, explainable, and reusable. Modelling concept Fact vs Dimension thinking A reliable...

  Introduction Despite the rise of modern analytics tools, Excel remains one of the most widely used tools in data analysis. What separates an average Excel user from a highly effective data analyst is not the number of functions they know, but how they structure data, solve problems, and support decision-making. In this post, I share a practical approach to building advanced Excel skills that reflect real-world analytics work rather than exam-style knowledge. Why Excel Still Matters for Data Analysts Excel is often the first tool used to: explore unfamiliar datasets validate assumptions perform quick analyses communicate insights to non-technical stakeholders Understanding how to use Excel well improves analytical thinking, regardless of the tools used later. Thinking in Tables, Not Worksheets Experienced analysts treat Excel as a structured data tool, not a canvas. Key habits include: using Excel Tables consistently keeping raw data separate from analy...