GUIDE

Analytics or data science: how to choose

They sound similar and lead to different jobs. A plain comparison of the work, the tools, the routes in, and who each one suits.

Ashwood Education · 4 min read · Sept 2026

A laptop displaying a dashboard of charts and figures

Data analytics and data science overlap enough to be confusing and differ enough that picking the wrong one can cost you months. Job titles do not help: one company’s “data scientist” is another company’s “senior analyst”. So rather than compare titles, it is more useful to compare the work.

Analytics answers questions people already have

An analyst is usually handed a question by someone in the business. Why did sign-ups drop last month? Which stores are underperforming? Did the new pricing work? The job is to find the right data, check it, analyse it, and come back with an answer a busy person can act on.

The day-to-day work looks like this:

  • Writing SQL to pull and combine data from the company’s databases
  • Cleaning that data, which takes longer than anyone admits
  • Building dashboards and reports in tools such as Power BI, Tableau or Looker
  • Spreadsheets, still, for quick analysis and for colleagues who live in Excel
  • Explaining what the numbers mean, in meetings and in writing

The output is a decision. A good analyst is judged on whether their work changed what someone did.

Data science builds things that make predictions

A data scientist is more often building a model or system that keeps running: a forecast of next quarter’s demand, a model that flags likely fraud, a recommendation engine, or, increasingly, a system built around a large language model. The question is less “what happened?” and more “what will happen, and can we act on it automatically?”

The day-to-day work looks like this:

  • Python, and sometimes R, for data preparation, modelling and evaluation
  • Statistics and machine learning: choosing a method, testing it honestly, and knowing its limits
  • Designing experiments, such as A/B tests
  • Working with engineers to put a model into production and monitor it
  • Still a lot of SQL, and still a lot of cleaning

The output is a working system. A good data scientist is judged on whether the model performs in the real world, not just in a notebook.

A laptop showing a bar chart and a pie chart, next to a tablet

The overlap is bigger than the difference

Both jobs need SQL. Both need you to understand the business well enough to ask the right question. Both involve explaining your work to people who do not share your vocabulary. Many data scientists started as analysts, and many analysts now use Python and simple models in their work. Choosing one route does not lock you out of the other.

How the routes in compare

Analytics is usually the more accessible first step. There are more junior analyst roles than junior data science roles, and the core skills (SQL, a visualisation tool and clear communication) can be learnt to a hireable standard in months rather than years. People coming from operations, finance, retail management or any job that involves reports often bring useful knowledge of how a business runs.

Data science asks for more. Employers expect comfort with programming, statistics and machine learning, and junior roles often attract applicants with quantitative degrees. It is very achievable as a career change, but it usually takes longer and benefits from a portfolio that shows real modelling work rather than tutorials.

On pay, data science roles generally advertise higher salaries than analyst roles at a similar level of experience. Pay varies a lot by sector and region, so look at live job adverts for the roles you would actually apply for, in the places you would actually work, rather than relying on national averages.

Which suits you?

Analytics probably suits you if:

  • You like answering concrete questions and seeing the answer used quickly
  • You enjoy explaining things to colleagues who are not technical
  • You want to be job-ready sooner, and build from there

Data science probably suits you if:

  • You enjoy programming and maths, rather than just tolerating them
  • You are drawn to building systems more than producing answers
  • You are prepared for a longer run-up and a more technical interview process

A practical test

Before you commit to either, spend a weekend on each. For analytics, take a public dataset, ask three business questions of it in SQL, and build one dashboard. For data science, take the same dataset, build a simple prediction model in Python, and write down honestly how well it works. Notice which one you would happily do again the following weekend. That is usually your answer.

Where Ashwood fits

Our Data Analytics course follows the analyst route, from Excel and SQL to Power BI, Tableau and Python. The Data Analytics + Generative AI course adds machine learning and generative AI to the same foundations. The Data Science + Agentic AI course covers the data science route in depth, from Python and statistics to deploying models. If you are still unsure, book a call and talk it through with an adviser.

Want to work at Ashwood?

We hire practitioners to teach and mentor each cohort.

See open roles