Job search · 7 min read

Portfolio when you are not a designer: proving your work as analyst, marketer, PM or ops

A designer has a portfolio of pixels. A product manager, analyst, marketer, operations person or support specialist does not. Your work is often invisible to the outside world: it lives in spreadsheets, Slack conversations, process improvements or decisions made behind closed doors. When you are applying for a job, an employer wants proof that you can do the work. But what does a credible artefact look like when your output is not a design file or a deployed application? This post covers what to show, how to show it without breaching confidentiality, and where to host it.

What a portfolio needs to do

A portfolio has two jobs. First, it is proof that you have done the work before. Second, it demonstrates your approach and thinking, not just your output. For a designer, the portfolio shows the visual result. For everyone else, the portfolio needs to show the work that led to the result: the research, the trade-offs, the decision-making.

A hiring manager reading a portfolio for a marketer is asking: 'Did you actually run that campaign or did someone else? Can you articulate why you made the decisions you made? Do you know what happened after—did it work?' They are not asking for beautiful presentation. They are asking for evidence.

The portfolio also needs to be honest. If you took a tiny part in a large project, do not present it as your own. If you built a model that someone else implemented, be clear about your boundary. Exaggeration will be found out in reference calls or when you start the job and you cannot do what you claimed.

What an artefact looks like for different roles

For a data analyst, a portfolio should include a worked example: a business question, the data you sourced or used, your analysis and your conclusion. This could be a PDF report showing a trend over time and what you did to investigate it. It could be a one-page memo with a chart and your recommendation. The point is someone can see: you found data, you understood it, you drew a conclusion and you communicated it to someone else. If the original work is confidential, you can anonymise the data and redo the analysis with public datasets or artificial data, so long as it still demonstrates your method.

For a marketer, include a campaign you ran or contributed significantly to. The portfolio piece should show what the objective was, what you tested or tried, what the results were and what you learned. If you ran a social media campaign, show the content (text and images), show the metrics (engagement, clicks, conversions), and say what you would do differently next time. If you do not have metrics, that is fine—say 'I did not have tracking set up at the time' and move on. Honesty is better than invented numbers.

For a project or operations manager, the artefact is usually a written description of a process improvement or a project. What problem were you solving? What was the scope? What approach did you take? What was the outcome? How did you measure success? You do not need a 20-page deck. A two-page memo with clear headings is better. Imagine a hiring manager reading it in three minutes. They should understand what you did and why it mattered.

For a support specialist, include examples of how you handled customer interactions or improved support processes. This could be a one-page case study: a difficult customer question, how you diagnosed the issue, how you resolved it and what you improved afterwards. If you built a self-service resource or an FAQ, that is portfolio-worthy. If you reduced response time, show the before and after. Quantify where you can.

For a financial analyst or controller, portfolios are trickier because much financial work is sensitive. Instead of pulling actual numbers, you can describe the work: 'I built a three-year forecast for department budgets, including scenario analysis for 20% and 40% revenue changes. I then tracked actual spend against forecast and flagged variances over 5% to management.' That is less concrete than a real spreadsheet but it shows you know what you are doing. You could also offer to walk through a sanitised example in an interview.

The confidentiality constraint is everything

You cannot, under any circumstances, publish confidential company information, financial data, client lists, strategy documents or anything else marked proprietary. This is not caution. This is a legal and ethical line. If you are tempted to think 'Just a little bit of anonymisation will make it okay', stop. Do not do it. Breaching confidentiality will destroy your career. A future employer will think: if you will do this to your current employer, you will do it to us. References will be sour or missing. You may face legal action.

This means most of your real work cannot go in a portfolio. That is fine. You have options instead.

Building a portfolio from public data

If your field relies on proprietary data, use public data instead. This works particularly well for analysts and marketers. Take a public dataset—anything from Kaggle, government statistics, census data, sports statistics, public financial reports—and redo work you did at your employer. If you built a cohort analysis at work with proprietary customer data, build one with public e-commerce data or sports player data. If you ran A/B tests on your company's website, design and run an experiment with public data or build a write-up of how you would test a hypothesis with an available dataset.

The point is not that the public version is better than your real work. It is that the public version proves you can think the way your real work requires. A hiring manager will see: this person knows how to structure a question, find data, analyse it and draw conclusions. That is what matters.

For marketers, you can build a case study of a campaign you would run for a fictional brand or a real brand you do not work for. Use real channels—social media, email, display advertising—to show what you would do. Walk through your research, your creative choices, your channel decisions and how you would measure it. You will not actually run it (you cannot spend a real company's money), but you can show your thinking.

For PMs and operations people, build a process document or a one-page product requirement for something you would build or improve in a space that interests you. If you work in healthcare, describe how you would change patient onboarding. If you work in fintech, describe how you would improve the account opening flow. You will not have real users or real constraints, but you can show you understand the problem and have thought through a solution.

Where to host it

You do not need a complex website. A simple WordPress blog or Google Site is fine. On Medium or LinkedIn, you can publish case studies directly. Some people use GitHub for non-code projects—it is designed for code but it works for markdown documentation too. For spreadsheet-based work, Google Sheets is often better than PDF because it allows a reader to poke around and verify the work.

Put each piece on its own page with a clear title and a one-paragraph description of what it is. Link from your LinkedIn profile and from your CV. When you send a CV to a recruiter, include a note: 'I have some case studies and examples of my work at [URL]. Have a look if you get a chance.' Do not assume they will find it on their own.

Your portfolio does not need to be visually stunning. It needs to be clear and honest. Someone reading it should understand your approach and be able to imagine you solving a real problem at their company.

Quantifying and proving results

If you can attach numbers to your work, do. 'I reduced processing time from 40 minutes to 12 minutes' is better than 'I improved efficiency.' 'I increased response rate by 23% by testing subject lines' is better than 'I ran email campaigns.' But be careful: invented numbers are worse than no numbers at all. If you do not have precise metrics, say so clearly: 'I did not have tracking in place at the time, but based on feedback from the team, the process became significantly faster.' Be specific about what 'significantly' meant: 'People went from spending most of a morning on the task to spending about an hour.'

The specificity itself is credible. Vague claims like 'improved results' or 'better outcomes' sound hollow. A genuine description of what changed—even if approximate—sounds like someone who actually did the work.

If you are early in your career and most of your work is small projects or you contributed as part of a team, still build a portfolio. Pick the strongest piece you can credibly claim and describe it fairly. 'I led the analysis' is different from 'I ran a query'. 'I contributed to this campaign' is different from 'I built this from strategy to execution'. Be precise. A hiring manager will respect someone who is clear about their contribution more than someone who oversells it.

You can also include work from university, volunteer roles or personal projects. If you built a data model for a non-profit, that is legitimate portfolio work. If you ran a small business, that is real experience. If you did a capstone project, you can sanitise it and include it. Start where you are.

The short version

Your portfolio should prove you have done the work and that you understand the reasoning behind it. It does not need to show real company data—that is illegal and will wreck your career. Instead, use public datasets, fictional examples or fully anonymised descriptions. Pick artefacts that demonstrate your full process: the problem, your approach, the result and what you learned. A two-page memo or a single worked example is better than ten pieces of shallow work. Host it simply on Medium, Google Docs or a basic website and link from your CV.

Related guides

Try JobStraight free →