The short version: you break into an entry level data analyst job by building three real portfolio projects with tools like Power BI, learning SQL and Excel to an interview-ready standard, and applying to the unglamorous job titles such as MI Analyst or Reporting Analyst that get far fewer applicants than “Data Analyst.” Most people land their first role through an internal move or a small company willing to train someone, not a cold application backed by a certificate. Give it three to nine months of consistent work and you’ll have an offer.
What “entry level” means (and why the job ads mislead you)
I want to get one thing out of the way because it wastes people months of confidence. If you’ve searched LinkedIn and found “entry level data analyst” postings asking for two to three years of experience, SQL, Python, Tableau, and a stats degree, you’re not misreading it. That’s a real, widely known problem in this job market. Recruiters and hiring managers write “entry level” to mean “cheap to hire” rather than “no experience needed.” It’s frustrating and it’s also a filter you can use to your advantage once you know it exists, because half the people reading that ad will talk themselves out of applying, which thins the field for you.
Real entry level pay in the UK for a first data analyst role sits between £24,000 and £30,000 outside London, and £28,000 to £34,000 in and around London. In the US it’s roughly $50,000 to $62,000 for a genuine first job (junior reporting or insights, not a senior title dressed down). These numbers matter because if you see a listing offering £22,000 with a demand for three years of Python and machine learning experience, that’s not entry level, that’s a company hoping someone desperate applies. Skip it.
The five skills hiring managers test for
Forget long course syllabuses for a second. When I’ve sat in on hiring for analyst roles, five things get checked, in this order:
- Excel: pivot tables, VLOOKUP or XLOOKUP, and being able to spot a broken formula in someone else’s spreadsheet. This is tested more often than SQL for true entry level roles.
- SQL: SELECT, JOIN, GROUP BY, and basic window functions. You don’t need to know stored procedures. You need to write a clean query under time pressure without panicking.
- One BI tool, used: Power BI or Tableau, built to the point where you can explain why you chose a chart type, not just that it looks tidy.
- Basic statistics: mean versus median, what a standard deviation tells you, and the difference between correlation and causation. Interviewers ask this constantly because so many candidates get it wrong.
- Communication: turning a chart into one paragraph a non-technical manager can act on. This is the skill that separates hired candidates from rejected ones, and almost nobody trains for it.
Notice Python isn’t on that list. It helps later. For a genuine first job it’s rarely the deciding factor, and I’ve seen candidates spend four months learning Python who would have been hired three months sooner with solid SQL and a working knowledge of statistics instead.
Build a portfolio that doesn’t look like a tutorial clone
Here’s the problem with most beginner portfolios: they’re the same three projects. Titanic survival prediction, a COVID dashboard, and a Netflix dataset analysis. Hiring managers have seen these hundreds of times because they’re the default projects in every free course. Copying them tells an interviewer you followed instructions, not that you can think.
Use datasets that fewer people touch: UK ONS population and employment data, Companies House filings, Premier League match statistics, Spotify streaming charts, or your own city council’s open data portal (most UK councils publish spending and planning data). Pick a question a real business would ask, not “what patterns exist in this data.” For example: “Which UK regions had the fastest small business growth in 2025, and does that correlate with local unemployment figures?” Answer it with a short written summary covering the business question, your method, the finding, and one recommendation. Three projects like that, each with a two-paragraph write-up, beat ten half-finished dashboards.
A few years ago I brought on a marketing assistant, Priya, who wanted to move into analytics. She had no data background, just enthusiasm and a spreadsheet habit. She did the Google Data Analytics Certificate on Coursera (about £39 a month, most people finish it in three to five months studying part time), then built a Power BI dashboard tracking our own client campaign performance, which is exactly the kind of internal project that makes a portfolio credible because it wasn’t hypothetical. Fourteen months after she started, she was hired as a junior reporting analyst at a fintech company. Her offer came from a recruiter who found her LinkedIn post about that dashboard, not from a job board application. That detail matters more than people admit.
Why your first offer probably won’t come from a job board
This is the part most guides skip because it’s less encouraging than “just apply to 50 jobs.” Cold applications to external data analyst postings, from someone with zero paid data experience, have a low success rate, often under 2 percent by most recruiters’ own estimates, because you’re competing against people who already have “analyst” somewhere in their job title. Certificates alone rarely change that math. Hiring managers now see so many identical Google Certificate portfolios that the certificate itself has stopped being a differentiator.
The routes that work more often are less glamorous. An internal move within a company you already work for, moving from customer service, admin, or finance into a reporting or MI role, happens far more than people expect, because your manager already trusts you and knows you show up. It’s the same pattern I’ve written about for people trying to move out of call centre work into remote roles, and for people trying to break into HR from an unrelated background: the fastest door is usually the one inside the building you’re already in, not the one on LinkedIn.
Small companies are the other realistic door. A 20-person business that needs someone to “sort out our spreadsheets and build a monthly report” isn’t going to demand two years of experience, because they can’t afford a senior hire and they need the problem solved now. These roles rarely say “data analyst” in the title. They say office manager, operations coordinator, or marketing executive with data responsibilities. Take one of those, build your skills on the job, and move sideways into a proper analyst title after twelve to eighteen months. It’s slower on paper and faster in practice.
Specialist data recruiters (Harnham and Understanding Recruitment are two of the bigger UK ones) also place junior candidates that job boards never see, because companies use them precisely to skip the flood of applications you’d be competing against on LinkedIn.
Widen the job titles you’re searching for
“Data Analyst” gets thousands of applicants per posting in most UK cities. These titles, doing near-identical work, get a fraction of that:
- MI Analyst (Management Information)
- Reporting Analyst
- Insights Analyst
- Junior Business Analyst (data-heavy version, check the job description, not just the title)
- Marketing Analyst or Campaign Analyst
- Data Coordinator or Data Officer (common in charities and local government)
If you’re still building basic comfort with spreadsheets and structured data, roles listed under part time data input jobs are a legitimate, underrated stepping stone, not a dead end. They teach you to spot data quality problems from the inside, which is exactly what analysts get paid to catch. The same goes for other entry level remote roles worth considering first if you need income while you build skills on evenings and weekends.
It’s also worth saying plainly: the same automation that’s reshaping customer service teams right now is starting to do the same to basic reporting work. Tools like Power BI’s Copilot and ChatGPT-generated SQL can now produce a passable first-draft chart or query in seconds. That doesn’t kill entry level analyst jobs, but it does shift what companies want from junior hires: less “can you write a query” and more “can you sanity-check what the AI produced and explain what it means.” Build that judgement into your portfolio write-ups now and you’ll be ahead of most candidates who are still just showing charts.
What gets asked in the interview
Expect three stages for most junior roles:
- A short screening call about your background and why data.
- A technical test: a live or take-home SQL exercise (practise on DataLemur or StrataScratch, both free tiers), often paired with an Excel task.
- A presentation or case study: you’re given a dataset, asked to find something interesting, and present it to a small panel in five to ten minutes.
The case study is where people lose it, and not for the reason they think. Panels aren’t grading the elegance of your chart. They’re watching whether you can say “I don’t know” about a limitation in the data, whether you draw one clear conclusion instead of six vague ones, and whether you can answer a follow-up question without freezing. Practise saying your finding out loud to a friend who has no data background. If they understand it in one pass, you’re ready.
Your first 90 days once you’re hired
Nobody warns you about this bit, so I will. Your first weeks won’t be exciting analysis. They’ll be “can you pull last month’s numbers into this template,” fixing a broken spreadsheet formula someone left behind three years ago, and learning where the company keeps its actual source data, which is rarely where the documentation says it is. That’s normal, and it’s not a sign you’ve taken the wrong job. The people who move up fastest are the ones who do the boring pulls without complaint for the first three months while quietly noting where the real problems are, then bring one useful improvement to their manager around week ten or twelve. Don’t try to overhaul the reporting system in your first fortnight. Earn trust with reliability first, then bring insight.
Frequently asked questions
Do I need a degree to become a data analyst?
No. A degree in maths, economics, or a related subject helps for some graduate schemes, but most junior analyst roles, especially at smaller companies, care more about a working portfolio and solid SQL and Excel skills than your degree title.
How long does it take to become job-ready with no experience?
Realistically three to nine months of consistent part time study and building, roughly ten hours a week: two to three months on Excel and SQL fundamentals, two to three months building portfolio projects, and the rest applying while you keep learning. People who treat it as a two-week crash course rarely last through the interview stage.
Is the Google Data Analytics Certificate worth doing?
It’s a solid, structured starting point that covers SQL, Excel, and Tableau basics for about £39 a month on Coursera, but it won’t get you hired on its own because tens of thousands of other candidates have the identical certificate and identical sample projects. Use it as a foundation, then build original portfolio pieces on top of it.
Can I become a data analyst without learning to code?
Yes, up to a point. Many entry level MI and reporting roles run entirely on Excel and a BI tool like Power BI with no coding required. Basic SQL still opens far more doors and takes most people six to eight weeks of steady practice to become interview-ready in, so it’s worth the investment even if you skip Python entir
Related reading: Best Entry Level Online English Teaching Jobs and What a Remote Data Analyst Job Involves for Beginners (No Fluff).
