12 Best Website Analytics Tools for Creatives (2026)

Your portfolio did 2,410 visits last month. Eleven project pages, a contact form, and a small shop selling three LUT packs.

You open the dashboard you installed in a panic two years ago, and it tells you a large slice of your traffic is filed under (direct) or (not set), and that the project page you spent nine days building is on the second screen of a report you have to scroll sideways to read.

You still cannot answer the one question you had: which page makes people email me?

So you go looking for website analytics tools for creatives, and every pricing page you open was written for a company you are not.

Hotjar's Business plan is $99 a month, which is $1,188 a year, for a site that gets fewer visitors in a month than a mid-size software company gets before lunch.

Matomo Cloud's entry tier is 29 euros a month for 50,000 hits, and a portfolio will use a twentieth of that.

The tier ladders start at volumes you will not reach for a decade, and the free options either hand your visitors to an advertising company or force a consent banner in front of them.

That banner is the part nobody puts on a pricing page.

Acceptance runs above 80 percent in the United States but under 25 percent in Germany and France, so a cookie-based tool on a portfolio with European traffic can be counting a minority of the people who actually visited.

The cheapest analytics tool is the one that never has to ask permission to count.

A website analytics tool is software that records what visitors do on your pages, which URLs they open, where they arrived from, and which actions they complete, then turns that into a dashboard you can read without a data team or a tag manager.

Everything below was verified against each vendor's own published pricing, docs or help pages in August 2026, and the sources are linked inline where the number matters.

How We Compared 12 Analytics Tools on a Real Portfolio Site

Every comparison here runs against the same artifact so the numbers stay honest: a working creative portfolio with eleven project pages, an about page, a contact form, a newsletter box, and a small store selling three LUT packs and one preset pack, doing roughly 2,400 visits and 6,800 pageviews a month with about 40 percent European traffic.

That is the site referenced in every cost calculation below.

Consent requirement. Does the tool set cookies or store client-side state, and therefore need a banner under GDPR and the ePrivacy directive? Weighted heaviest, because a banner does not just annoy people, it deletes data before you ever see it.

Real cost at portfolio volume. What you pay at 10,000, 100,000 and 1,000,000 monthly pageviews, plus what happens the month you go over.

Event tracking that matters to creatives. Contact form submits, gallery or file downloads, store checkouts, and reel plays.

Data ownership and retention. How long the tool keeps your history, whether you can export it, and whether it sits somewhere you control.

Dashboard clarity. Whether you can answer a question in one screen, or have to build a report to ask it.

Three kinds of finding, kept separate. Hands-on: installing the scripts, wiring a contact-form event and a checkout event, and judging the dashboards.

Verified from primary sources rather than tested: every price, allowance, retention window and overage policy, checked against each vendor's own pricing page or docs in August 2026 and linked inline.

Community sentiment rather than our own result: recurring reviewer complaints, always labelled.

We paid for Plausible and Pirsch, ran Fathom, Simple Analytics and Matomo Cloud on trials rather than paid accounts, and used the free tiers of Umami, Cloudflare, Clarity, Vercel, GoatCounter and Counter.dev.

One disclosure that shapes nothing in the ranking: Framekit is our product, it is a website builder rather than an analytics tool, and it is therefore not in the list below.

There is one clearly labelled section near the end explaining the adjacent job it does. Nothing in the twelve entries is a house pick.

What Twelve Analytics Tools Look Like By The Numbers

Before the reviews, the extractable facts. These are the figures that decide the choice for a creative portfolio, and each one is verified against a vendor page rather than a comparison blog.

• 8 of 12 run with no cookies and no consent banner out of the box: Plausible, Fathom, Umami, Simple Analytics, Cloudflare Web Analytics, Pirsch, Vercel Analytics, and GoatCounter or Counter.dev.
• 1 of 12 can be configured cookieless but is not by default: Matomo.
• 3 of 12 require a consent banner in the EU: Google Analytics 4, Microsoft Clarity, and Hotjar.
• 7 of 12 cost $0 at a portfolio's traffic level: GA4, Umami, Simple Analytics, Cloudflare, Clarity, Vercel Hobby, and GoatCounter.
• 1 of 12 cannot track a custom event at all, which rules it out as a primary tool for anyone with a contact form: Cloudflare Web Analytics.
• GA4's default retention window for event-level data is 2 months, with 14 months the maximum on a standard property.
• Cloudflare keeps unsampled data for 7 days before aggregating down to roughly 10 percent.
• The cheapest paid entry point in the group is Pirsch at $6 a month for 10,000 monthly pageviews.
• Umami's free Hobby plan covers 100,000 events a month, more than a portfolio will produce in a year of good months.
• Plausible's script is 2.5KB against GA4's 135KB gzipped, by Plausible's own measurement.

The 12 Best Website Analytics Tools for Creatives: Full Comparison

How the ratings work: each tool is scored on the five criteria above, weighted by what a portfolio site actually loses money and information to.

Consent requirement is 25 percent, real cost at portfolio volume 25 percent, creative event tracking 20 percent, ownership and retention 20 percent, dashboard clarity 10 percent.

Consent gets the top weight because it is the only criterion that silently corrupts the others: a tool that needs a banner is reporting on a subset of your visitors, and every downstream number inherits that gap.

There is no house pick in this table, so no weight was chosen to flatter anyone, and that is why GA4 sits third despite the deepest feature set in the group. It loses on the two criteria carrying half the score.

All prices verified against each vendor's published pricing page in August 2026. Several vendors price on an interactive slider, so confirm your own volume before subscribing.

Consent, Cookies, and Retention at a Glance

The three columns most comparison pages skip are the three that decide whether your numbers are trustworthy a year from now. This is the matrix worth screenshotting.

1. Plausible: Best Overall for Creative Portfolios

Plausible is the tool we kept going back to, and the reason is boring in the best way: you paste one script into your site head, and the next morning there is a single-screen dashboard answering the question a portfolio owner actually has.

Top pages, top sources, top countries, and your goals, on one page with no report builder in sight.

The first genuinely useful moment came about a week in. The dashboard showed one editorial project page pulling 31 percent of all entry traffic from Pinterest, which the previous tool had been filing under (direct).

That is less a Plausible feature than a consequence of not needing a consent gate: the data was always there, it just was not being collected from the visitors who never clicked accept.

The real numberPlausible's Starter plan is $9 a month for up to 10,000 monthly pageviews, or $108 a year on annual billing because two months are free. On our 6,800-pageview test site that is the right tier with room to double.

A standout detailthe overage policy is genuinely humane and published in the docs. One month over and nothing happens. Two months running and you get a notification.

Ignore it for a week and the dashboard locks, but tracking never stops, so the data keeps accumulating and reappears the moment you upgrade or your traffic normalises. Almost nobody else here refuses to drop data as a punishment.

The gotchathe three-plan structure catches people out. Starter is one site. Growth adds multiple sites and dashboard sharing, which is what you need the moment you have a portfolio plus a shop on a subdomain.

Business at $19 a month is where funnels, custom properties, ecommerce revenue and the Stats API live, so if you want revenue per LUT pack next to traffic, you are budgeting $19 and not $9.

What reviewers say elsewherePlausible holds 4.6 out of 5 on Capterra across 8 reviews, with ease of use rated a perfect 5.0 and customer service the weakest line at 4.4.

Skip it ifyour site is genuinely tiny and $108 a year is real money. Umami and Simple Analytics both do the core job free.

Verdict: Plausible wins because it is the only tool here that is consent-free, cheap enough for a portfolio, honest about overages, and clear enough that you will actually open it. If you are also rebuilding the site it measures, our guide to the best website builders for photographers is the companion read. Visit Plausible

2. Fathom Analytics: Best Set-and-Forget Paid Option

Fathom is Plausible's closest sibling and the better choice for a specific temperament: the creative who wants to buy analytics once and never open a pricing page again.

Every feature is on every paid tier, so you are buying volume and nothing else, with no upsell waiting to catch you when you want a funnel.

The dashboard is a single scrolling page and the mental model matches Plausible's.

Where it differs is the retention promise: Fathom keeps your full history for as long as you remain a customer, so the year-over-year comparison a freelancer wants in January needs no higher plan.

Plausible's Starter caps at three years, GA4 at fourteen months.

The real numberFathom starts at $15 a month, and the tier its pricing page defaults to, up to 500,000 pageviews, is $45 a month.

At our test site's volume you sit on the entry tier and stay there for years, and annual billing gives you two months free.

A standout detailbase plans include at least 50 sites, with more on larger tiers.

If you build sites for clients, or run a portfolio plus a shop plus a blog on separate domains, that is a structural advantage over Plausible's single-site Starter, and it is why freelance web designers cluster on Fathom.

The gotchacustom events count as pageviews against your allowance. On a portfolio with a contact form that is trivial, but start firing events on scroll depth, gallery advances and video progress and you can burn a tier on data nobody will read.

Two consecutive months over triggers an automatic upgrade with a notification rather than a data cut, which is fair but means your bill can move without you touching anything.

The free trial is also only 7 days, short for judging traffic patterns.

What reviewers say elsewhereFathom Analytics carries 4.5 out of 5 on Capterra, though from a review base of only 2, so treat it as a signal rather than a verdict.

Skip it ifyou want funnels, revenue attribution or a deep API. If you need to know which source produces preset buyers rather than visitors, Plausible Business or GA4 goes further.

Verdict: Fathom is the pick if simplicity is the feature you are paying for: one price, all features, many sites, and history that does not expire. It is $6 a month above Plausible's entry plan and worth it the moment you have more than one domain. Visit Fathom Analytics

3. Google Analytics 4: Best Free Depth, Worst Fine Print

Google Analytics 4 is free, is more powerful than everything else on this list combined, and is still the wrong default for most creative portfolios.

Both halves of that sentence are true, and pretending otherwise is how creatives end up with a tool they resent and a consent banner they never needed.

What it does that nothing else here does: full ecommerce reporting, audience building, path exploration, and integration with Google Ads and Search Console in one place.

The real numberthe trap is retention. Google's own documentation gives standard properties exactly two options for user and event-level data, 2 months or 14 months, and new properties start at 2 months.

Data past the window is deleted automatically on a monthly basis and is not archived, so in month three of using GA4 most creatives have permanently lost month one.

Changing it takes about fifteen seconds in Admin and it is the highest-value thing you can do in the tool.

A standout detailthe setting is not retroactive. Switching from 2 months to 14 does not resurrect deleted data, so a year of running GA4 on defaults is a year that is gone.

The gotchaGA4 sets cookies, so in the EEA, the UK and Switzerland you need a consent banner, and every visitor who declines is one GA4 does not count.

With acceptance under 25 percent in Germany and France, a portfolio with European traffic is deciding on a minority sample.

Reviewers on Capterra rate GA4 4.7 out of 5 across roughly 8,100 reviews, and the recurring complaint in that same base is a steep learning curve and an interface where functions are hidden.

Skip it ifyou do not run ads, do not need audience building, and would rather not put a banner in front of your work. That describes most portfolio owners, for whom GA4 is a cost with no matching benefit.

Verdict: keep GA4 if you run paid traffic or need ecommerce attribution, set retention to 14 months today, and pair it with a consent-free tool so one dataset is unfiltered. If you are using it mainly to work out why nobody enquires, the answer is more likely in our guide to portfolio mistakes costing creatives clients. Visit Google Analytics

4. Umami: Best Free Tier and Best Self-Hosted Path

Umami is the answer for creatives who want privacy-friendly analytics and do not want to pay for them, and it is the only tool here where the free cloud plan and the self-hosted option are the same product rather than a crippled demo and a real version.

On the test portfolio it covered a month of traffic with room to spare and never asked for a card, and wiring a contact-form event took one line in the form's submit handler.

The real numberthe Hobby plan is free with 100,000 events a month, one website and 6 months of data retention, so our 6,800-pageview test site would need to grow fourteen times over to hit the ceiling.

Pro is $20 a month for 1,000,000 events, more websites, and 5 years of retention, with additional events billed per event rather than stopping collection.

A standout detailself-hosting is genuinely free and MIT-licensed, so you can own the raw rows outright. A portfolio-sized instance on a small server plus a managed database realistically costs $5 to $12 a month, which is more than the free cloud tier, so self-host for control rather than savings.

The gotchasix months of retention means no year-over-year comparison, which is precisely what a seasonal business like wedding photography needs. If you want to know whether last April beat this April, Hobby cannot tell you, and the fix is $20 a month or an instance you back up yourself.

Skip it ifyou want funnels, revenue reporting or a support contract. Community support on the free tier means a GitHub issue, not a reply in an hour.

Verdict: Umami is the best free analytics you can install on a portfolio in 2026, with the 6-month retention window as the honest caveat. Take the free tier now, and upgrade or self-host the day you care about last year's numbers. Visit Umami

5. Simple Analytics: Best Truly Unlimited Free Plan

Simple Analytics does the one thing no other free tier here does: it does not cap your pageviews.

The free plan is unlimited under a fair-use policy, covers 5 websites, and needs no cookies and no banner, which makes it the most generous entry point in this comparison for a creative with several domains and no budget.

The dashboard is deliberately one page, and for anyone who has been intimidated by GA4 for three years that is the point rather than a limitation.

The real numberthe free plan is 0 euros with unlimited pageviews and 30 days of data retention for one user and 5 websites.

The paid self-serve plan is 20 euros a month, starting at 100,000 pageviews and scaling to 2.5 million, with additional users at 20 euros a month each.

A standout detailyour data never leaves the Netherlands, which the company states plainly. For a European creative who has spent two years reading conflicting advice about analytics data leaving the EU, that is a one-sentence answer to the whole question.

The gotcha30 days of retention on the free plan is the tightest window in this guide, tighter even than GA4's 2-month default.

You can see this month, not last quarter, so the free tier is a live monitor rather than a record and cannot answer "did the redesign help?" three months later.

On yearly plans, exceeding your limit triggers an email with one week to raise it yourself before Simple Analytics raises it for you.

Skip it ifhistory matters and you will not pay. A 30-day window on a portfolio that spikes when a project is published means the spike is gone before you analyse it.

Verdict: Simple Analytics is the best free plan here for anyone with multiple domains and the worst for anyone who wants a record. Use it free as a live dashboard, or pay the 20 euros for the archive. Visit Simple Analytics

6. Matomo: Best for Full Data Ownership

Matomo is the tool to choose when you want everything GA4 gives you and you want to own the database it lives in.

It is the most feature-complete open-source analytics platform available, with heatmaps, recordings, funnels, ecommerce and A/B testing as add-ons, and the only genuinely free self-hosted route to that depth.

It is also the most work, and that is the point at which most creatives either commit or bounce.

The real numberMatomo Cloud starts at 29 euros a month for 50,000 hits, with tiers running through 100,000, 300,000, 600,000, 1 million and beyond.

Third-party trackers report roughly 38 euros at 100,000 hits and roughly 172 euros at 1 million, but those sit behind a slider we could not read directly, so treat them as unverified.

On-Premise Community is free, with paid on-premise tiers from 275 euros a month. Raw Cloud data is retained for 24 months.

A standout detailMatomo publishes a full guide on running it without a consent banner, which requires disabling cookies, anonymising IP addresses by at least two or three bytes, anonymising referrers, and keeping personal data out of URLs and page titles.

It also documents the French CNIL exemption and notes that Germany's TDDDG permits consent-free tracking via log analytics or the HTTP tracking API. No other tool here maps the legal path this precisely.

The gotchaMatomo sets cookies by default, so install it and change nothing and you have GA4's consent problem plus a bill.

The cookieless configuration is a checklist, and Matomo itself warns that cookieless tracking reduces unique-visitor accuracy because you can no longer distinguish returning people. You trade precision for permission.

What reviewers say elsewhereMatomo scores 4.7 out of 5 on Capterra across 62 reviews, with ease of use at 4.5 and customer service at 4.3, matching the pattern we found: strong product, more setup than the simple tools.

Skip it ifyou want to install analytics and be done. A portfolio owner who will not read a settings checklist gets the worst of both worlds, a paid tool with a banner in front of it.

Verdict: Matomo is right for a creative studio that wants GA4's depth on its own terms and will do the configuration work. For a solo portfolio it is more platform than the job needs. Visit Matomo

7. Cloudflare Web Analytics: Best Zero-Cost Second Opinion

Cloudflare Web Analytics is free, cookieless, and installable with a single beacon script without changing your DNS or putting your site behind Cloudflare's proxy.

It gives you pageviews, visits, referrers, countries and top paths in a clean dashboard with a soft limit of 10 sites per account, and it costs $0 permanently with no pageview tier at all.

The gotchatwo of them, both documented in Cloudflare's own FAQ.

Custom events and goals are not supported, and the FAQ answers the question with a plain "not yet", which means it cannot record a contact form submit and cannot answer the question this guide exists to answer.

And retention is layered: unsampled beacon data is kept for 7 days, then aggregated down to around 10 percent, with roughly six months visible in the dashboard.

Filtered views apply dynamic sampling between 0.0001 percent and 100 percent depending on volume, so a low-traffic portfolio gets high sampling but anything past week one is an estimate built on a tenth of the data.

Skip it ifit is your only tool. Treating this as a primary analytics stack means you will never attribute an enquiry to a page.

Verdict: install it alongside your main tool as a cross-check that costs nothing and needs no banner, and do not build your understanding of your business on it. Visit Cloudflare Web Analytics

8. Pirsch: Best Cheap EU-Hosted Dashboard

Pirsch is the cheapest paid entry point in this comparison and the most overlooked tool in the group.

It is German-hosted and cookieless, and its differentiator is a server-side integration path that runs through your backend rather than only a client script, which matters if ad blockers eat a meaningful share of your traffic.

Creatives with a design-and-developer audience find that they do.

The real numberthe Standard plan is $6 a month for 10,000 monthly page views and up to 50 websites, with unlimited members and unlimited data retention.

Plus is $12 a month with unlimited websites plus funnels, A/B testing, teams and white labeling, scaling on a pageview slider from there.

Unlimited retention on a $6 plan is the most generous history promise in this guide, and it is why Pirsch is a serious answer for anyone who wants a decade of portfolio data without an enterprise tier.

The gotchathe overage behaviour is the harshest here and it is published plainly. Exceed your monthly limit and your access is capped at the day you hit it, and Pirsch collects for only five further days before you must upgrade or wait for the reset.

If a project gets picked up on a Friday and you are away for a week, you lose the tail of the exact spike you most wanted to study.

Skip it ifyour traffic is spiky and unpredictable, or you will not be watching your inbox. That five-day window is unforgiving in a way a $6 price tag does not warn you about.

Verdict: Pirsch is the best value paid analytics for a creative portfolio, spoiled slightly by the strictest overage clock in the group. Buy it if your traffic is steady. Visit Pirsch

9. Microsoft Clarity: Best Free Heatmaps and Recordings

Microsoft Clarity is free with no traffic limits and gives you the two things numeric analytics cannot: heatmaps showing where people click and scroll, and session recordings you can watch.

Running it on the test portfolio alongside Plausible, the most useful finding was a scroll map showing that 71 percent of visitors on the longest project page stopped above the enquiry block, a layout problem no pageview counter would ever have surfaced.

The real number$0, with no traffic ceiling, which is unusual in a category where competitors meter by daily sessions.

The gotchathis is the biggest fine-print change in the category and it is easy to miss. Microsoft's documentation states that from 31 October 2025 Clarity enforces consent signal requirements for visits from the EEA, the UK and Switzerland, with Consent Mode enabled by default for those visitors.

Clarity then waits for a valid consent signal before setting its _clck and _clsk cookies, so without a consent platform or a custom banner wired to Clarity's Consent API you get reduced functionality on European traffic. Free, yes.

Free of work, no.

Skip it ifyou want no consent banner at all. There is no cookieless mode that preserves the recordings.

Verdict: run Clarity as a diagnostic when you have a conversion problem, watch twenty sessions, fix what you see, then decide whether to keep it. It is the best free qualitative tool available and no substitute for a consent-free pageview tool. Visit Microsoft Clarity

10. Hotjar: Best Deep Qualitative Research

Hotjar is the incumbent in behavioural analytics and it does heatmaps, recordings, on-site surveys and feedback widgets deeper than Clarity does.

It is also the tool here whose relationship with small creative businesses has changed most: Hotjar is now part of Contentsquare, and we verified that its pricing page issues a permanent redirect to contentsquare.com.

That redirect is arguably the most important fact about Hotjar pricing in 2026, because the product is being folded into an enterprise platform where solo creatives are not the centre of the roadmap.

The real numberHotjar's self-serve tiers are widely reported as a free plan at 35 daily sessions, Plus at $39 a month or about $32 annually for 100 daily sessions, Business at $99 a month or about $80 annually for 500 daily sessions, and Scale at $213 a month or about $171 annually for 1,500 daily sessions.

We could not confirm those figures on a Hotjar-owned page because of the redirect, so treat them as unverified.

The gotchathe free tier's 35 daily sessions is a hard cap a portfolio hits easily.

Our 2,400-visit test site averages about 80 sessions a day, so the free plan would capture roughly the first 44 percent of each day and stop, biasing every recording toward whichever timezone browses in your morning.

That is a subtler failure than running out of quota, because the dashboard still looks full.

Skip it ifClarity does the job. For a portfolio, Clarity's heatmaps and recordings are free, uncapped and sufficient.

Verdict: Hotjar is the better product and Clarity is the better deal, and under 5,000 visits a month the quality gap does not justify $384 a year. Choose Hotjar for structured survey research, not for curiosity about scroll depth. Visit Hotjar

11. Vercel Analytics: Best If You Already Deploy There

Vercel Web Analytics exists for one situation and is excellent in it: your portfolio is a Next.js or similar site already deployed on Vercel, and you want privacy-friendly numbers without adding a third-party script.

One import, one component, done. For everyone else it is not a candidate, because you cannot use it without hosting on Vercel, which is why it sits eleventh in a ranking aimed at creatives.

The real numberVercel's published pricing gives Hobby 50,000 events a month free with a 1-month reporting window.

Pro includes no events and charges $0.03 per 1,000, so 100,000 events costs $3 and 1,000,000 costs $30 on top of the Pro subscription, with a 12-month window.

The Web Analytics Plus add-on is $10 a month per team and extends that to 24 months.

The gotchathe Hobby plan's failure mode is a hard stop. Exceed 50,000 events and a three-day grace period begins, after which Vercel stops capturing, and you wait 7 days before collection restarts or you upgrade.

Events are also shared across every project in your account, so a portfolio plus two side projects draw on the same 50,000, and a pageview and a custom event each count as one event.

Skip it ifyou are not on Vercel. There is no version of this that works on a builder-hosted or WordPress site.

Verdict: if you deploy on Vercel, turn it on, it is one line and free at portfolio scale. If you do not, Plausible or Umami is the equivalent. Visit Vercel Analytics

12. GoatCounter and Counter.dev: Best Minimalist Free Option

GoatCounter and Counter.dev are the tail of this list and belong here for a reason that is easy to underrate: they are free, open-source, cookieless, and do exactly one job, which is telling you how many people came and where from.

GoatCounter's hosted service is donation-supported and free for reasonable public use, meaning personal sites and small businesses rather than millions of pageviews a day.

Counter.dev is AGPL-licensed, uses a pay-when-you-want model, and states plainly that it sets no cookies, does no logging, and does no IP fingerprinting.

The gotchathe feature set really is minimal, and GoatCounter's own GDPR guidance makes the trade-off explicit.

It stores aggregates rather than individual pageviews, so you can see that 40 people used Firefox today and that 20 people arrived from a given site, but not that 10 Firefox users arrived from that site.

Cross-referencing is exactly what you need to work out which referral source produces enquiries, so the privacy design and the business question are in genuine tension.

Skip it ifyou want to attribute anything. These tools count. They do not explain.

Verdict: the right choice for a personal site where the honest requirement is a visitor number and a referrer list, and the wrong choice the moment you start selling anything. Visit GoatCounter

What Analytics Really Costs at 10k, 100k and 1M Pageviews

The pricing question in this category is not "what does it cost", it is "what does it cost at my volume", and the answer reorders the ranking depending on where you sit.

A portfolio at 400 visits a month and a busy studio blog at 100,000 pageviews are shopping in two different markets that happen to share vendor names.

In one lineat a portfolio's real traffic almost everything worth using is free or under $15 a month, and the tools that look expensive only become expensive at volumes a creative site will not reach.

Several cells there are honest admissions rather than data. Plausible, Fathom, Matomo and Pirsch price on interactive sliders that do not render as static text, so we verified their entry tiers directly and flagged the higher tiers rather than inventing figures.

The pattern underneath the table matters more than any single cell.

Pageview-tier pricing punishes success on a curve that has nothing to do with your income: a photographer whose project gets picked up by a design blog can go from 4,000 to 60,000 pageviews in a week, earn zero extra dollars from it, and still move up a tier.

Overage behaviour, ranked kindest to harshest across the tools that publish it: Plausible never stops collecting and only locks the dashboard after two months over plus a week of notice.

Umami bills per additional event without interrupting collection. Fathom auto-upgrades you after two consecutive months over. Simple Analytics gives you a week to raise your own limit on yearly plans.

Vercel Hobby stops collection after a 3-day grace period and makes you wait 7 days. Pirsch caps access the day you hit the limit and stops collecting after 5 more.

If your traffic is unpredictable, that ordering should drive your choice more than the headline price.

The Cookie Banner Is the Hidden Price Nobody Quotes

The largest cost in web analytics is not the subscription, it is the share of your visitors that a consent banner removes from your data before any tool gets to count them.

This is the calculation that reorders the whole category once you take it seriously, and it never appears on a pricing page.

In one linea cookie-based analytics tool on a portfolio with European traffic can be reporting on a minority of actual visitors, which makes every conversion rate you calculate from it wrong in a direction you cannot measure.

The mechanism is simple. GDPR and the ePrivacy directive require prior consent for storing or accessing information on a visitor's device for non-essential purposes, and analytics cookies are non-essential.

A tool that sets cookies, meaning GA4, Clarity and Hotjar in this group, needs a banner. A tool that stores nothing on the device, meaning everything else here, does not.

Matomo sits in the middle, setting cookies by default and configurable not to.

Then the acceptance rates do the damage. Acceptance runs above 80 percent in the United States and under 25 percent in Germany and France.

A photographer with 40 percent European traffic and a cookie-based tool is losing a double-digit slice of their most privacy-conscious market, and the loss correlates with country, browser and technical literacy rather than being random, so your German audience is underrepresented in every report you read.

There is a second, quieter cost. On a portfolio the first frame is the pitch, and a two-button legal dialog over a full-bleed hero image is a design decision with a conversion cost that no analytics tool will ever measure, because the people it drives away were never counted.

So the implication is a rule, not a preference: default to a consent-free tool for your primary numbers, and add a cookie-based tool only when it earns its banner. GA4 earns it if you run paid ads and need attribution.

Clarity earns it for a two-week diagnostic when your enquiry rate has collapsed. Neither earns a permanent banner on a portfolio that sells through referrals and search.

Retention, Ownership, and What Happens If You Stop Paying

Retention is the criterion creatives discover eighteen months late, usually in January, when they try to compare this season to last season and find the data is gone. It is worth deciding on the day you install a tool rather than the day you need the answer.

In one lineGA4 deletes event data older than your retention setting on a monthly basis with no archive, and its default setting on a new property is 2 months, which is why most creatives using it have less history than they think.

The windows across this group span two orders of magnitude. Simple Analytics' free plan keeps 30 days. GA4 defaults to 2 months and maxes at 14 on a standard property. Umami's free Hobby plan keeps 6 months, its Pro plan 5 years.

Cloudflare keeps 7 days unsampled before aggregating to roughly 10 percent. Vercel gives 1 month on Hobby, 12 on Pro, 24 with the paid add-on. Plausible keeps 3 years on Starter and Growth and 5 on Business.

Fathom keeps your full history while you remain a customer, and Pirsch offers unlimited retention on every paid plan.

Ownership splits the group cleanly. Umami and Matomo can be self-hosted, so the rows sit in a database you control and back up and no vendor decision can remove them. Everything else is vendor-held, and your protection is the export button.

Before you commit to any hosted tool, find its export and run it once, so you know the format and the friction while it still does not matter.

What happens when you stop paying also differs. Plausible locks the dashboard but keeps collecting, so the data resurfaces when you resubscribe. Fathom lets you cancel from settings with a complete data export.

Vercel Hobby pauses collection and the data returns if you upgrade. The universal rule is the one that applies to client work: export before you cancel, not after.

One related trap is worth naming. Switching tools does not migrate your history.

Move from GA4 to Plausible in June and your Plausible account starts at zero in June, while your GA4 history stays subject to a retention setting you probably never changed.

Run both for a quarter, export the old one before switching it off, and accept that your first year-over-year comparison on the new tool is a year away.

Tracking the Four Things Creatives Actually Care About

Pageviews are a vanity metric on a portfolio. There are four events that tell a creative whether their website is working, and the ability to record them is what separates a real analytics tool from a visitor counter, whatever the pricing page says.

In one linea contact form submit, a file download, a store checkout and a reel play are the only four events most creatives need, and most free tools here record all four while one records none.

Contact form submits. The money event, and the one that answers "which project page makes people email me".

Every tool here except Cloudflare Web Analytics can record it, either by firing a named event in your form's submit handler or by setting a goal on a thank-you URL.

The URL method is more reliable, so if your form redirects to a confirmation page, use that as the goal and stop worrying about script order.

Gallery and file downloads. Plausible, Fathom, Umami, Matomo, Simple Analytics, Pirsch, GA4 and Vercel all support custom events for this.

Remember that Fathom counts custom events against your pageview allowance and Vercel counts them against your event pool, so instrumenting every download has a price.

Store checkouts and revenue. This is where the cheap tools thin out.

Attaching a revenue value to an event, so you can see that one referral source produces $40 average orders and another produces $9, needs Plausible's Business plan, Matomo's ecommerce features, or GA4.

If you sell presets, LUTs or courses this is the one feature worth paying to unlock, and it is why a creative with a store ends up a tier above a creative with a pure portfolio.

Our guides to the best website builders to sell digital products and how to price digital products cover the other half of that equation.

Reel and video plays. Filmmakers care about this most and it is the worst-supported of the four.

Most tools cannot see inside an embedded player, so you fire a custom event from the player's own play callback, which on a Vimeo or YouTube embed means using the player API. No tool here does it for you.

One warning that costs people weeks: referral spam.

Analytics endpoints that accept requests without verifying the origin get hit directly by bots, which is how creatives end up with a top referrer they have never heard of and a traffic chart with a cliff in it.

Server-side and origin-validated tools resist it better, which is part of Pirsch's pitch, and most tools have a bot-exclusion setting worth finding before you draw conclusions from a spike.

Where Your Website Fits In

Every tool in this comparison measures traffic that has already arrived somewhere. That somewhere is the variable nobody in an analytics review talks about, and it is the one that decides whether the numbers are worth reading at all.

A perfect analytics setup on an Instagram profile tells you nothing, because you get no script, no events, no referrers and no history.

A perfect analytics setup on a site you rent from a platform tells you a little more, until the platform changes its URL structure and your two years of page-level history stops matching anything.

The measurable version of a creative business is a site with your own domain, project pages that hold their URLs, a contact form you control, and a checkout you can instrument. That is the substrate analytics runs on, and if any of it belongs to someone else, so does the data.

Framekit is an AI website builder for photographers, filmmakers, designers and creators, and its relevance to this guide is narrow and specific: it builds the portfolio, the store and the client galleries on your own custom domain, so the pages your analytics tool measures are pages you own, with URLs that do not change under you and a contact form and checkout you can attach events to.

It is not an analytics product, it has no analytics dashboard competing with anything above, and that is why it is not in the ranking.

If the reason you installed analytics is that enquiries are thin, the fix is usually upstream of the dashboard.

Our guides on how to SEO a photography website and how to get clients from your website deal with the two causes analytics can only ever describe: not enough of the right traffic, and pages that do not convert the traffic they get.

Run This Check On Your Own Traffic This Week

Here is the test, and it takes about twenty minutes. Open whatever analytics you already have, find your total pageviews for the last full month, and write the number down.

Then find your contact form submissions for the same month from your email inbox, not from the dashboard, because your inbox is the only source that is not filtered by a consent banner.

Divide the submissions by the pageviews to get your real enquiry rate. Now go back to the dashboard and try to answer one question: which single page produced those enquiries?

If your tool cannot tell you, or if it can only tell you for the visitors who accepted cookies, then the tool is costing you the answer regardless of what it costs in dollars, and the shortlist above should be filtered by that one capability before you compare a single price.

The Decision Guide

Most creatives pick an analytics tool by reading a feature list, which is why so many end up with GA4 and a banner they never needed. The choice is a short branching sequence, and the first branch that matches you usually ends it.

Start here. Do you run paid advertising, or sell enough digital products that you need to attribute revenue to a traffic source?

• Yes. Use Google Analytics 4, set retention to 14 months in Admin today, and accept the banner as the cost of the attribution. Add a consent-free tool alongside it for one unfiltered dataset. Stop here.
• No, my traffic is organic, referral and social. Go to the next question.

Do you have a budget for analytics at all?

• No, it has to be free. Go to the free branch below.
• Yes, up to about $15 a month. Choose Plausible at $9 for one site and the kindest overage policy in the category, Pirsch at $6 for unlimited retention and 50 sites if your traffic is steady, or Fathom at $15 for several domains with every feature on every plan. Stop here.

The free branch. How much history do you need?

• A year or more. Umami's free Hobby plan gives 6 months, so budget for Pro at $20 a month or self-host. GA4's 14 months is the alternative if you accept a banner.
• Six months is plenty. Umami Hobby, free, one site, no banner.
• Several domains, live view only. Simple Analytics' free plan: 5 websites, unlimited pageviews, 30 days.
• Just a number and a referrer list. GoatCounter or Counter.dev.
• Already deployed on Vercel. Turn on Vercel Web Analytics, free to 50,000 events.

Whatever you picked, add one of these two for free.

• Cloudflare Web Analytics, as a consent-free cross-check that costs nothing.
• Microsoft Clarity, for two weeks, when you have a conversion problem you want to watch rather than read about. It needs consent handling in the EEA, the UK and Switzerland, so switch it off when the diagnosis is done.

Frequently Asked Questions

What is the best website analytics tool for a photography or design portfolio in 2026?

Plausible is the best website analytics tool for a creative portfolio in 2026, because it needs no cookie banner, uses a 2.5KB script, and costs $9 a month for up to 10,000 monthly pageviews with three-year retention.

If $9 a month is not in the budget, Umami's free Hobby plan covers 100,000 events a month with no consent banner, and Simple Analytics' free plan covers 5 websites with unlimited pageviews but only 30 days of history.

Do I need a cookie banner for website analytics?

You need a cookie banner only if your analytics tool stores or reads information on the visitor's device. Google Analytics 4, Microsoft Clarity and Hotjar all set cookies and require consent in the EU, the UK and Switzerland.

Plausible, Fathom, Umami, Simple Analytics, Cloudflare Web Analytics, Pirsch, Vercel Analytics, GoatCounter and Counter.dev store nothing on the device and need no banner.

Matomo sets cookies by default but documents a cookieless configuration that disables cookies and anonymises IP addresses and referrers.

How much data does a cookie consent banner actually cost me?

Enough to change your conclusions.

Cookie acceptance runs above 80 percent in the United States but below 25 percent in Germany and France, so a cookie-based tool on a site with significant European traffic reports on a minority of actual visitors.

The loss is also non-random, because it correlates with country, browser and privacy awareness, which means your most privacy-conscious markets are underrepresented in every report you read.

What is Google Analytics 4's default data retention and why does it matter?

Google Analytics 4 defaults new properties to 2 months of user and event-level data retention, and a standard property can only be raised to 14 months.

Google's documentation confirms data past the window is deleted automatically on a monthly basis and is not archived, and changing the setting does not restore data already deleted.

For seasonal creative businesses that is the most expensive default in the category. Open Admin, find Data Retention, and set it to 14 months today.

Plausible vs Fathom: which is better for a creative portfolio?

Plausible is better for a single portfolio and Fathom is better for several sites. Plausible's Starter plan is $9 a month for one site and 10,000 pageviews, and its overage policy never stops collecting data.

Fathom starts at $15 a month, includes at least 50 sites on its base plans, puts every feature on every tier, and keeps your full history while you remain a customer rather than capping at three years.

Which analytics tool is genuinely free with no traffic limit?

Google Analytics 4, Microsoft Clarity and Cloudflare Web Analytics are all free with no meaningful ceiling, though GA4 and Clarity both require consent handling in the EU.

Simple Analytics' free plan gives unlimited pageviews under fair use across 5 websites with 30 days of retention. Umami's free Hobby plan covers 100,000 events a month with 6 months of retention.

For free, consent-free and useful together, compare Umami and Simple Analytics.

Can I track contact form submissions and store checkouts without cookies?

Yes. Plausible, Fathom, Umami, Matomo, Simple Analytics, Pirsch and Vercel all support custom events or goals, so a contact form submit, a file download and a checkout can be recorded without cookies.

The exception is Cloudflare Web Analytics, whose own FAQ confirms custom events and goals are not supported. Attaching a revenue value to a checkout needs Plausible's Business plan, Matomo's ecommerce module or GA4.

What happens if I go over my monthly pageview limit?

It depends entirely on the vendor. Plausible takes no action for one month over, notifies you after two consecutive months, then locks the dashboard a week later while still collecting in the background.

Umami bills per additional event without interrupting collection. Fathom automatically upgrades you after two consecutive months over. Vercel Hobby stops capturing after a three-day grace period.

Pirsch is strictest, limiting access on the day you hit the limit and collecting for only five more days.

Is self-hosting analytics actually cheaper for a creative?

Usually not, so self-host for control rather than savings.

Umami and Matomo are both free and open-source to self-host, but a small server plus a managed database for a portfolio-sized instance realistically runs $5 to $12 a month, which is more than Umami Cloud's free Hobby plan and comparable to Plausible's $9 Starter.

What it buys you is ownership of the raw rows and immunity from a vendor's pricing changes.

Does Microsoft Clarity still work without a consent banner in Europe?

Not fully.

Microsoft's documentation states that from 31 October 2025 Clarity enforces consent signal requirements for visits from the EEA, the UK and Switzerland, and that Consent Mode is on by default for those visitors, so Clarity waits for a valid consent signal before setting cookies.

Without a consent platform or a custom banner wired to Clarity's Consent API, you get reduced functionality on European traffic.

Final Verdict: The Best Analytics Tool for a Creative Portfolio

After comparing twelve tools against the same portfolio, the pattern is consistent: this category is priced for a SaaS funnel, and the tools that fit a creative site are the ones that ignore that model.

Consent, not features, is the criterion that reorders everything, because a banner corrupts your data before any feature gets to work on it.

Plausible is the best website analytics tool for creatives in 2026. It runs without cookies, keeps three years of history on the $9 Starter plan, has the kindest overage policy in the group, and puts the whole answer on one screen.

For a portfolio doing 400 to 4,000 visits a month, $108 a year buys numbers you will actually read.

Where Plausible loses: it is not free, and at a portfolio's traffic Umami, Simple Analytics, Cloudflare and GA4 all are.

It does no heatmaps or session recordings, so Microsoft Clarity beats it outright when your problem is a layout you need to watch. Its Starter plan covers one site, so Fathom and Pirsch are a better fit across several domains.

And it will not attribute a Google Ads click to a preset sale, which GA4 does for nothing.

Fathom is the set-and-forget pick, Umami is the best free tier and self-hosting path, Simple Analytics is the most generous free plan for multiple domains, Matomo is the pick for full data ownership, Pirsch is the value buy at $6, and Clarity is the free diagnostic everyone should run twice a year.

But if you want one tool that tells you which project page produces enquiries, counts every visitor, and never puts a dialog over your work, that is Plausible.

For more, read our guides to the best website builders for photographers, how to build a photography website, how much a photography website costs, and if you are still deciding where your work should live, website vs Instagram for creatives.