Website & CRO
Coffee Shop Website CRO: Fix These 5 Things Before You Ever A/B Test
Part of the guide: Coffee Shop Google Maps & Local SEO
Here is the number that should change how a coffee shop thinks about "testing". Evan Miller's online sample size calculator, with its default example of a 10.2% baseline rate and a lift to 13.2%, asks for 2,545 visitors per variation. That is about 5,090 visitors to compare two versions of one button. If your shop's website gets 1,500 visits a month, you would wait months for one verdict. So for most cafes the smart conversion work is not split testing. It is fixing the five basics that stop a phone user from reaching you at all.
This guide is for independent coffee shop owners in the US, UK, Australia and Canada. It replaces the usual pile of unsourced lift percentages with what can be checked: Google's published speed thresholds, Google's own Business Profile metrics, Nielsen Norman Group's usability findings on menus in PDFs, and a simple way to count real conversions. Where a number is my illustration, it is labelled as an assumption.
First, Define a Conversion for a Coffee Shop
Most cafe customers never "convert" on your website in the e-commerce sense. They tap one of five things on a phone, then walk in:
- Directions (to the door)
- Call (to ask about hours, orders or a booking)
- Menu view (to decide if you have what they want)
- Order link (if you offer pickup or delivery)
- Loyalty or email sign-up (to come back)
Google's Business Profile performance help reports several of these for your listing: directions, calls (taps on the call button), website clicks and menu clicks. That is your free baseline. Your website's job is to finish what the listing started. If you have not tuned the listing itself, start with our Google Business Profile guide for coffee shops.
Fix 1: Make the Mobile Page Fast Enough
Google's Core Web Vitals guidance on web.dev sets three "good" thresholds, assessed at the 75th percentile of page loads: Largest Contentful Paint within 2.5 seconds, Interaction to Next Paint of 200 milliseconds or less, and Cumulative Layout Shift of 0.1 or less.
2.5 s→
Largest Contentful Paint
Good threshold, web.dev
200 ms→
Interaction to Next Paint
Good threshold, web.dev
0.1→
Cumulative Layout Shift
Good threshold, web.dev
Estimates for illustration; actual numbers vary.
In plain English: the main thing on the screen should appear within about two and a half seconds, taps should respond quickly, and the page should not jump around as it loads. Test your homepage on your own phone over mobile data (not office wifi), then run it through Google's free PageSpeed Insights. The usual cafe culprits are a huge hero photo straight from a camera, a looping background video, and a stack of embedded widgets. Resize photos to the width they display at, drop the video, and cut widgets you cannot justify. Our guide on how fast a local business site should load goes deeper.
Fix 2: Put the Menu in the Page, Not in a PDF
This is the most common cafe-site mistake. Nielsen Norman Group's article PDF: Still Unfit for Human Consumption reports usability findings that read like a description of the average cafe site. In one study, a cafeteria menu stored as a PDF on an intranet took almost 3 minutes to download, and the user gave up, saying they did not care what was for lunch. NN/g also found PDFs are sized for paper, which means tiny fonts and pinch-and-zoom on a phone, and that participants often skipped PDFs or abandoned them quickly. Its direction is clear: do not use PDFs for content that could and should be a web page.
What to do instead:
- Build the menu as plain text on a page, with headings (Espresso, Filter, Food, Milk options) and prices. Text also lets people find you when they search for a specific item.
- Mark dietary and milk-alternative information in the text, not only in a photo.
- Keep a printable PDF only as an optional download, linked from the menu page, as NN/g's "gateway page" advice suggests.
- Add the menu to your Google Business Profile as well. The performance report counts menu clicks as one of the actions it tracks, so you can see whether people use it.
Fix 3: Put the Three Actions in the First Screen
Open your homepage on a phone and do not scroll. Can you see, without hunting, whether you are open right now, where you are, and how to get there or call? If not, rearrange. There is no peer-reviewed rule for the right layout, so treat this as a practical checklist, not a statistic:
- Today's opening hours as text ("Open today until 4pm"), kept current.
- A Directions button that opens the maps app, and a Call button using a tap-to-call link.
- One primary button that matches your business: Order pickup, See the menu, or Join the club. Pick one, not four.
- Address as text, not baked into an image, so it can be selected and copied.
Holiday hours are where cafes lose people. Update hours on the site and on Google at the same time; mismatches are discussed in our trust signals guide, alongside which other credibility cues are worth your time.
Fix 4: Count the Taps Properly
You cannot improve what you do not count. Google's key events help for Analytics 4 defines a key event as an event that measures an action particularly important to your business, and says any event you collect can become one. The practical steps:
- In Google Analytics 4, mark the actions you care about as key events: order-button clicks, newsletter sign-ups, and a custom event for taps on the phone number.
- Check what is collected by default. Google's enhanced measurement page lists page views, scrolls, outbound clicks (links leading to another domain), site search, video engagement, file downloads and form interactions. It does not say that tel: (phone) links are tracked, so test your own site, and if needed add a custom event, for example through Google Tag Manager.
- Use a separate record for directions and calls from Google itself, from the Business Profile performance report above.
- Write down the weekly count of each action in a spreadsheet next to what you changed that week.
Our walkthroughs on Google Analytics 4 for small business and call tracking show the setup if you want screenshots.
Fix 5: Change One Thing at a Time, and Do Not Pretend to A/B Test
An older version of this article recommended Google Optimize as a free A/B testing tool. That advice is out of date: Google's help page states that Optimize and Optimize 360 stopped being available on September 30, 2023, and that Google is pointing users to third-party testing integrations such as AB Tasty, Optimizely and VWO.
More important is the arithmetic from the hook. A proper split test needs thousands of visitors per variation, and the sample you need grows as the lift you want to detect shrinks. A small cafe site cannot do that for a button colour. A better method at your size:
- Make one change, then compare the same number of weeks before and after. Compare like with like: do not judge a December change against November.
- Write the change and date in your spreadsheet. Two changes in the same week means you will not know which one worked.
- Look at big moves, not tiny ones. Moving from 20 call taps to 28 may be noise; keep the change anyway if it is free and sensible, but do not announce a win.
- Prefer changes with a known direction, such as a faster page and a readable menu, over cosmetic experiments.
In the Shop: The Same Logic Applies
The same "remove friction" thinking works at the counter. Put the loyalty sign-up QR code where people wait, not where they pay. Train the person at the till to mention it with every order. Our loyalty ideas that work without an app shows low-tech versions and the math to check they pay. I am deliberately not quoting uplift percentages for these tactics, because the figures usually passed around for them have no traceable source.
The Math
What is a fixed website worth? Every number below is an assumption for illustration. Replace them with your own counts from GA4 and the Business Profile report.
| Before | After | |
|---|---|---|
| Monthly mobile website visitors (assumption) | 1,200 | 1,200 |
| Visitors who tap directions, call or order (assumption) | 6% | 8% |
| Taps | 72 | 96 |
| Taps that become a visit (assumption) | 40% | 40% |
| Extra or retained visits | 28.8 | 38.4 |
| Average ticket (assumption) | $8.00 | $8.00 |
| Revenue from website visits | $230 | $307 |
| Extra revenue per month | about $77 |
The gain is small on one month's figures, and that is the honest picture for a single cafe: the website will not transform the business by itself. What it does is stop wasting the people your Google listing and ads already send you. If you pay for clicks, the same lift lowers your cost per visit; see our Google Ads for coffee shops guide for that side. A visit is also worth more than one ticket if the person joins your loyalty club and returns.
A Slow Menu Is a Closed Door
If someone standing on the pavement cannot read your menu in a few seconds, they will walk to the next cafe on the map. Test your own site on a phone, on mobile data, while standing outside your shop.
Do This Today
- Open your homepage on your phone over mobile data. Note how long the main content takes to appear, and run the URL through PageSpeed Insights.
- Replace any PDF menu with a text menu page, and keep the PDF only as an optional download.
- Move today's hours, a Directions button, a Call button and one primary action into the first screen.
- In GA4, mark order clicks and sign-ups as key events, and add a custom event for phone-number taps. Open Business Profile Performance and record last month's directions, calls and menu clicks.
- Start a change log in a spreadsheet: one change per week, with the weekly counts beside it.
FAQ
Q: How long does it take to see results from website changes at a coffee shop?
A: Speed and menu fixes can show in your counts within a few weeks, but at a small site's traffic you are comparing totals, not running a statistically solid test. Evan Miller's calculator example needs 2,545 visitors per variation to detect a move from 10.2% to 13.2%, so judge small changes cautiously.
Q: Is there a free A/B testing tool for a cafe website?
A: Google Optimize, the old free option, ended on September 30, 2023. Google points to third-party tools, which are paid. For most cafes, a before-and-after comparison in a spreadsheet is more honest than a split test with too little traffic.
Q: Should I use a PDF or a web page for my menu?
A: A web page. Nielsen Norman Group's research found users skip or abandon PDFs and that they are poorly suited to phones. Offer a PDF only as an optional printable download.
Q: How do I know if my problem is online or in the shop?
A: Compare three counts: Google Business Profile directions and calls, website visits, and people who say they found you online when you ask at the till. If listing views are strong but taps are low, the page is leaking; if taps are healthy but sales are flat, look at the shop experience.
Q: Does my cafe even need a website?
A: A profile on Google plus a simple, fast page with hours, menu and map covers most needs. Our local business website checklist lists what the page needs.
Sources
- Evan Miller: Sample Size Calculator for A/B tests (default example: 10.2% baseline, 2,545 per variation)
- web.dev: Web Vitals
- Nielsen Norman Group: PDF, Still Unfit for Human Consumption, 20 Years Later
- Google Business Profile Help: Understand your Business Profile performance
- Google Analytics Help: About key events
- Google Analytics Help: Enhanced measurement events
- Google Analytics Help: Google Optimize sunset
Want your site set up to count calls, directions and orders? See our website service, the coffee shop marketing page, or book a free audit.
Free for local businesses
Want this applied to your business?
I'll review your Google presence, local SEO, and ad accounts — and send you a specific action plan within 48 hours. No pitch, no pressure.
Want hands-on help?
See how DataLatte handles Website & Landing Pages for local businesses.
☕ Industry Guide
Coffee Shop Marketing Guide
Nataliia
Local marketing strategist with 10+ years at global agencies — OMD, Dentsu, GroupM, and BBDO. Now helping small businesses get the same data-driven edge. Based in Europe, working with clients in the US, UK, Australia, and beyond.
About NataliiaRelated articles
Website & CRO
How to Measure the ROI of Your Local Business Website
12 min read
Website & CRO
Should Your Local Business Website Show Prices? (Plus How to Do It Right)
12 min read
Website & CRO
Brewing a Personalized Experience with Website Personalization for Coffee Shops
12 min read
Website & CRO
Hero Section Examples for Local Business Websites That Convert
14 min readWant this applied to your business?
Let's review your current marketing setup together — free, no obligations.
Get Your Free Marketing Audit