Review a design portfolio as a busy hiring reviewer would. Then say what to fix. Back every finding with evidence from UXfolio's State of UX Portfolios 2026 (a survey of people who hire for UX and product design, plus eight interviews; source: https://uxfol.io/state-of-ux-portfolios-2026/report).
1. Get the link
- If the request has no portfolio URL, ask for it and stop. Do not guess the URL. Do not search for the person. Do not review from memory.
- In the same message, ask once for two optional things: the target role and any password. If they skip these, carry on and say what you assumed.
- Review one link at a time. If they send several, ask which one to do first.
2. Open it like a reviewer
- Fetch the URL. If that fails, try a browser. If it still fails, report that as the first finding (see section 3, "Access"). Stop there until they send a working link.
- Limit the first look to about 30 seconds. Note only what you can see in that time: who this is, what they do, which projects there are and where to click. Write this down before you read further.
- Open the home page, the About page and the case studies. Read at least the first listed project and the one closest to the target role. If there are more than four, read the first three and skim the rest.
- If you can, also check a narrow viewport (resize the browser or use a mobile user agent).
- If the site links to a CV or LinkedIn, open it and compare roles and dates. Do not search for them elsewhere.
- Record facts as you go: URL, format, load problems, number of projects, card text, headings, any numbers claimed, AI mentions, NDA notes. When you criticise something, quote the portfolio's own words.
3. What to check, and the evidence for each
Use these figures exactly. Cite them as "UXfolio 2026, ch. X.Y". Do not round, combine or invent numbers. If a point has no figure, say it is a judgement call.
First 30 seconds (ch. 1)
- 78% rely on the portfolio and case studies to judge fit, ahead of industry experience (46%) and CV (40%). (1.1)
- 65% spend under four minutes on a first look: 10% under a minute, 55% 1 to 4 minutes. (1.2)
- Quote: "When I have 100 portfolios to go through, I may have only 30 seconds for each one." (1.2)
- The scan is a gate. Shortlisted case studies get a closer read later. Check the portfolio works for both: easy to skim in minutes, and solid under a close read. (1.3)
- 71% say their organisation doesn't screen portfolios with AI. The reader is a person, so do not judge it on keywords. (1.4)
- About section: 53% rate it important or extremely important (35% + 18%). Career path and tone of voice matter; personal detail less so. (1.5)
- The story must match the CV and LinkedIn. Inconsistent roles or dates are early red flags. The report gives no percentage. (1.6)
Format and access (ch. 2)
- 72% prefer a website. Figma prototype 30%, PDF 27%, slides 26%, Behance or Dribbble 14%. (2.1)
- Quote: "If a portfolio is shared as a Google Drive folder, I tend to move it to the back of the pile." (2.1)
- What stops a review immediately: broken links 75%, missing password 61%, requires download 40%, distracting animation 35%. (2.2, 2.3)
- Quote: "Failing to check that a portfolio is accessible suggests that they may not double-check their work." (2.2)
- Templates: 76% spot one when the same design appears across candidates, 51% notice placeholder copy, 49% notice a platform subdomain. A template does not count against them on its own. Flag leftover placeholder text. Do not flag the template. (2.4)
Finding the right project (ch. 3)
- What decides which case study gets opened first: relevance 73%, clear outcome in the preview 40%, visual appeal 39%, title 33%, first listed 28%. (3.1)
- Most valued elements: relevant case studies 89%, clear navigation 83%, mobile optimised 45%, visually ambitious design 40%, testimonials 11%. (3.2)
- Number of projects: 71% prefer three or four. 10% want 1 to 2, 7% want 5 or more. (3.3)
- The first slot reads as their best work. The report shows 31% for this but does not state the question behind it. Present it as "about a third, per UXfolio's graphic". If you need a firm number, use the 28% "first listed" figure from 3.1. (3.4)
- Project cards: 27% flag cards without enough context. A card should say what the project is, who it was for and what kind of product it is. (3.5, Bonus #5)
Inside a case study (ch. 4)
- What makes one strong: explanation of decisions 64%, business impact 46%, scannable layout 45%, constraints 31%. Visuals only 20%. (4.1)
- Biggest weaknesses: no explanation of thinking 49%, overly long or dense 44%, not showing outcome 43%, feels exaggerated 38%, no problem statement 38%, unclear ownership 33%, artefacts without context 30%. (4.2)
- Outcomes: the problem is missing outcomes. Documenting the process is rarely the issue (14% name "documenting all the steps"). If the product has not launched, they should say how success will be measured. 4.3 repeats the 4.2 question, so cite them as one finding. (4.3)
- Ownership: 33% flag an unclear role. Look for a plain line saying what the designer did and what the team did. (4.4)
- Quote: "It's mostly an easy-to-catch lie that you have 'done it all' on your own. Nobody can." (Bonus 6.4)
NDA work (4.5)
- 84% expect NDA work to appear in some form; only 16% say leave it out. Preferred: brief description and discuss in interview 35%, anonymised visuals 31%, general terms without visuals 18%.
- Quote: "I'd rather see anonymised work from a real client than a concept redesign of a famous app with no real constraints."
AI use and disclosure (ch. 5)
- 100% of reviewers have noticed AI in a portfolio. (5.1)
- How they spot it: buzzwords without detail 68%, no personal voice 62%, visuals that don't match the process 41%, generic process with no artefacts 39%, generic headers 33%. (5.2)
- Low visual quality is a red flag for 55%; visual ambition matters to 40%. Both figures come from other charts (Bonus list and 3.2), so cite each once. Clean beats impressive. (5.3)
- Obvious AI overuse is a red flag for 43% (same Bonus figure). Not using AI at all can read as falling behind, especially for senior designers. (5.4)
- Good disclosure: name the tool and the task, show the prompt or first draft and what changed, separate their own decisions from the output, and don't claim to be an "AI expert". (5.4)
The ranked red flags (Bonus): check each one and mark it present, absent or unclear.
- Very low overall visual quality, 55%
- Hard to navigate or confusing structure, 48%
- Obvious overuse of AI, 43%
- Overly long or dense, 42%
- Case study cards without enough context, 27%
- Irrelevant case studies, 20%
- Distracting animations, 20%
- Slow loading or technical issues, 18%
Note: "overly long or dense" is 42% here but 44% in 4.2, and "distracting animation" is 20% here but 35% in 2.2. They come from different questions. Cite the one that matches your point. Do not mix them.
Reviewer advice (Bonus 6.1 to 6.5): use these to frame the fixes.
- The portfolio is a product and the hiring manager is the user.
- One relevant case study beats five generic ones.
- Show impact, but don't inflate it: "having nothing is better than putting something that is not real."
- Explain why you did something, as well as what you did.
- Keep it short and in your own voice.
4. Score it
Score out of 100. Keep the maths visible.
| Area | Points |
|---|---|
| Access and format (opens, loads, no download, website) | 15 |
| First scan (clear who, what, where to click) | 15 |
| Project choice and cards (relevance, 3 or 4, card context) | 15 |
| Thinking and decisions | 20 |
| Outcomes and ownership | 15 |
| Voice, AI use, visual quality | 10 |
| About, consistency, NDA handling | 10 |
The weights are this skill's own, informed by the report's rankings. They are not from UXfolio. Say so in the report.
Overrides: if the link is broken, or locked with no password given, score 0 and stop. Take 3 points off the area total for each red flag marked present.
Verdict bands: 80+ ready to send, 60 to 79 send after fixes, 40 to 59 needs work first, under 40 rebuild.
5. Write the report
Write in plain British English, in short sentences. Address the designer as "you". Do not use em dashes, hype, filler openers or closers, or "Overall" summaries that repeat the list. Every criticism must quote or point to something real on their site. It must also cite a UXfolio figure or say it is a judgement call. Start with what most affects the first four minutes. Give five fixes at most, ranked.
Use this template. Keep the section names as they are. Do not add sections.
# Portfolio review: {name or domain}
{URL} · reviewed {date} · target role: {role, or "not given, assumed {x}"}
**Score: {n}/100. {verdict band}.**
{Two sentences: what a reviewer would think after 30 seconds, and the single biggest thing holding it back.}
## The first 30 seconds
{What you could tell, what you couldn't. Format and access in one or two lines.}
## What's working
- {Strength, with where on the site.} UXfolio 2026, ch. {x.y}: {figure or quote}.
- ...
## What's costing you
- {Problem, quoting their page.} UXfolio 2026, ch. {x.y}: {figure or quote}.
- ...
## Red flag check
| # | Red flag | % of reviewers | Here? |
|---|---|---|---|
| 1 | Very low overall visual quality | 55% | {yes / no / unclear: one-line note} |
| ... all eight ... |
## Other notes
{Observations that aren't strengths or mistakes: CV and LinkedIn match, NDA handling, AI disclosure, template use, mobile. Mark judgement calls as such.}
## Fix these first
1. {Action}. Why: {one line plus figure}. Effort: {small / medium / large}.
2. ...
## Scoring
{Area table with points given, red flag deductions, total. One line saying the weights are this review's own, not UXfolio's.}
Source: UXfolio, State of UX Portfolios 2026 (survey of hiring reviewers, 8 interviews).
6. Limits
- Say which pages you couldn't open and which you only skimmed.
- Judge the portfolio only. Do not judge the person's skill or hireability.
- Do not rewrite their case studies unless asked. Offer to do it.