Back to the building3Level 3 · Neza
Level 3 · Proposal · new site concept

Birrieria Neza

A first website for a San Jose birria truck that people love and can't find online.

Role
Design + build, pitching it
Year
2026
Stack
WordPress, a custom theme
Status
Concept Not live
Proposed Birrieria Neza homepage: ratings, hours and a neon birria sign
The proposed homepage on a phone
386Google reviews, and no website
3listings, 2 names, 2 sets of hours
99Lighthouse score on the demo build
Not livea proposal, built before asking
I

The brief

Birrieria Neza is a black food truck on East Santa Clara Street in San Jose. It has 386 Google reviews and 13,000 Instagram followers, and no website. Google's listing still says "Add website."

Nobody asked me for this. It was the top lead on my list of local trucks, and the gap was obvious: the prices only exist on the board at the truck, and it isn't on DoorDash or Uber Eats. So I built the site first and I'm bringing it to the owner as a proposal.

How much is the pizza birria?A question on Yelp that nobody answered. A second asked about tacos and consomé.
II

A night at the truck

The truck is open until 11pm. Late at night, someone looking it up has three questions, and today the answers are scattered or missing.

10:40

Are they still open?

Today
Instagram says weekdays open at "12am" (it means noon). Yelp says noon.
On the new site
One set of hours with a live "open now" badge, written once and used everywhere.
10:45

Where is it, and what's it called?

Today
"Birrieria Neza Food Truck" on Google, "Neza Birrieria" on Yelp and Instagram.
On the new site
One name, the address with directions, and a map that loads when you tap it.
10:50

What do I get, and how much?

Today
Only on the menu board at the window. The Yelp price questions sit unanswered.
On the new site
The full board online, plus a tap-to-call button for questions.
III

The site, phone first

People look up a food truck on their phone, often while nearby. So the phone layout came first.

Phone view: name, hours, ratings and the call button above the fold
  1. Open now, until 11pmThe first screen answers the late-night question: hours, address and the Google and Yelp ratings (4.6 and 4.4 stars).
  2. See the pricesOne button to the menu, built from the truck's own board, with each meat and the extras listed once.
  3. Call or get directionsTap to call and tap for directions, since it isn't on delivery apps.
  4. Know what to orderA "Don't know what to order?" list of fan favorites, and a "What is consomé?" answer for first-timers.

Neza's menu

  • Quesabirria $5
  • Quesa tacos $5
  • Consomé $3
  • Pizza birria $30
Read from the truck's menu board (photos of the board, Feb 2026). Prices to confirm with the owner.
Desktop menu page: Neza's menu in red neon with an Open late sign, meats and extras, then the birria section with prices
The desktop menu page. The title borrows the truck's red neon, with a small "Open late" sign hanging off the last letter.
IV

From avatar to truck

all I had was a 150px Instagram logo
The original logo: a small, blurry Instagram profile picturethe only logo online
The logo redrawn clean: red circle, black bull head, Birrieria Neza letteringredrawn clean, brand red
Cartoon of the Neza food truck with the clean logo on its sideon the truck, in perspective

1.Stud, my local AI on a Mac Studio, redrew the logo sharp. Snapped to the truck's red.

2.AI garbles lettering, so the truck is generated blank and the real logo is placed on the side.

3.Every image is generated, not a photo of the truck or food. Real food photos come from the owner.

V

The hard parts

The decisions that took the most thought.

01

Custom for each business

Margaritas and Neza start from the same code, but each one gets its own design.

The business details live in two data files, one for the business and one for the menu. When Neza needed fewer pages, I added an optional page list, and Margaritas stayed unchanged. If the owner signs on, the theme becomes a one-off built for this truck.

02

A new layout, not a recolor

My first round was Margaritas in Neza's colors. I rejected it.

The second round got its own layout: a neon hero, a find-the-truck card with hours and map, a meats ticker. "New client means new layout" is now a rule in my build checklist.

03

Hours that match everywhere

Instagram's hours don't match Yelp, and Google uses another name.

The site holds one set of hours and one name in its data file, and the FAQ, open-now badge and search data all read from it. The proposal asks the owner which name they prefer, so the listings can be fixed to match.

04

Art without a photo shoot

No photos I could use, and Canva ran out of credits after four images and returned small previews.

I moved the illustrations to Stud, which made eight full-size images in about a minute. The neon sign is split into its own layer so the red glow on the brick wall is CSS.

VI

The stack

WordPressThe owner can change a price or the hours without calling me.
Custom themeDesigned for this truck, so the hours and menu stay easy for the owner to update.
StudMy local AI image model, running on my Mac Studio. Free, full size, and fast enough to iterate.
CloudflareA private demo link that can expire, so the pitch can be shared before anything is public.
Map on tapThe Google map loads only when tapped, which keeps the first screen fast on a phone.
VII

Expected impact

It isn't live, so there are no customer numbers yet. Here is what it should change, and how fast the demo is.

  • The price questions on Yelp finally have an answer, one search away.
  • "Add website" on Google becomes a real link, with hours that match.
  • One tap to call or get directions at 11pm.
  • The owner edits the menu when prices change.
99Lighthouse performance
1.8 smain content loaded
445 KBhome page weight

Measured on the demo build after the speed pass (first run: 92, 3.2 s, 1.1 MB). Accessibility and best practices both scored 100.

VIII

Looking back

What worked

Building the site before asking. A finished demo with the truck's real menu says more than a pitch deck. Research took minutes, since every fact went into one file I reused for the site and the proposal.

Options for Alec
  1. The shared starting code: the second site took design time, not build time.
  2. Starting from customer questions (the unanswered Yelp posts) instead of a design idea.
  3. Stud for art: local, free, full size, after Canva hit its limit.

What I'd change

I'd get the layout right in round one. My first pass recolored an earlier client and I threw it away. Most of the design time after that went to small refinements that could have been batched into fewer rounds.

Options for Alec
  1. Visit the truck before designing, for real photos and the owner's voice.
  2. Check the business's own story sources (Facebook, local news) before writing copy.
  3. Write the Spanish outreach alongside the site, not after.

What's next

Show it to the owner: a message on Instagram in Spanish, then a visit with the proposal. If the answer is yes, confirm prices, publish, and point the Google listing at the new site.

Options for Alec
  1. A weekly check that hours, ratings and the menu still match everywhere.
  2. A Spanish version of the site.
  3. The next truck on the list, with its own custom design.