diff --git a/README.md b/README.md
index 95cc720..343de5c 100644
--- a/README.md
+++ b/README.md
@@ -16,6 +16,7 @@ Each couple gets their own set of interest checkboxes ("His Picks" / "Her Picks"
 - **Interest filter** — narrow the list to His Picks, Her Picks, Both Picked, or Either Picked
 - **Random picker** — "Pick for Us!" draws a random winner from the currently filtered pool, with an option to re-roll
 - **Course browse** — dedicated First Course, Second Course, and Third Course buttons open a modal listing every dish for that course across all restaurants; click any dish to jump straight to that restaurant's detail view
+- **Year selector** — dropdown on the course browse bar switches between available event years (2025, 2026); restaurant data, filters, and interest picks are all scoped to the selected year
 - **Clear Filters** — one-click button to reset all search and filter fields back to their defaults
 - **Reset Selected** — clears all saved picks (with a confirmation prompt) so you can start fresh
 - **Mobile-friendly** — responsive single-panel layout on small screens; the list and detail views swap in place, with a Back button to return to the list
@@ -23,7 +24,7 @@ Each couple gets their own set of interest checkboxes ("His Picks" / "Her Picks"
 
 ## Usage
 
-1. Serve the folder from a web server and open `restaurant-picker.html` in a browser. (The app fetches `2026-restaurants.json` at runtime, so it cannot be opened directly as a local `file://` URL — a server is required. A simple option is the VS Code Live Server extension, or `npx serve .` in the project folder.)
+1. Serve the folder from a web server and open `restaurant-picker.html` in a browser. (The app fetches a `YYYY-restaurants.json` file at runtime, so it cannot be opened directly as a local `file://` URL — a server is required. A simple option is the VS Code Live Server extension, or `npx serve .` in the project folder.)
 2. Browse the restaurant list or use the toolbar filters to narrow choices.
 3. Use the **First Course**, **Second Course**, or **Third Course** buttons to browse all dishes for a given course and jump to any restaurant that interests you.
 4. Check the **You** and **Wife** boxes on any restaurant you're each interested in.
@@ -45,12 +46,25 @@ Each couple gets their own set of interest checkboxes ("His Picks" / "Her Picks"
 
 ```
 restaurant-picker.html   # App shell — HTML, CSS, and JS logic
-2026-restaurants.json    # Restaurant data for the 2026 event (loaded at runtime via fetch)
+2026-restaurants.json    # Restaurant + event date data for the 2026 event
+2025-restaurants.json    # Restaurant + event date data for the 2025 event
 LICENSE
 README.md
 ```
 
-Each year, create a new `YYYY-restaurants.json` and update the filename in the `fetch()` call near the bottom of `restaurant-picker.html`.
+### Adding a New Year
+
+1. Create a new `YYYY-restaurants.json` file following the schema described below (see [Data Entry Guide](#data-entry-guide)).
+2. In `restaurant-picker.html`, add a new `<option>` to the year selector dropdown:
+   ```html
+   <select id="yearSelect" onchange="loadYear(this.value)">
+     <option value="2027">2027</option>  <!-- add new year at top -->
+     <option value="2026">2026</option>
+     <option value="2025">2025</option>
+   </select>
+   ```
+3. Update the hardcoded default in the `loadYear('2026')` init call at the bottom of the `<script>` block to the new year.
+4. Update the static fallback date in the `<span id="eventDates">` header element — it will be overwritten at runtime, but the static value is shown briefly on first load.
 
 ---
 
@@ -58,6 +72,20 @@ Each year, create a new `YYYY-restaurants.json` and update the filename in the `
 
 Each year's restaurant data is stored as a JSON file (`YYYY-restaurants.json`) loaded by `restaurant-picker.html` at runtime. The source data comes from the [IRW restaurant listing pages](https://inlanderrestaurantweek.com/restaurants/). This section covers the data schema and common pitfalls to avoid when entering or updating restaurant records.
 
+### File-Level Schema
+
+Each `YYYY-restaurants.json` is an **object** (not a bare array) with two top-level keys:
+
+```json
+{
+  "eventDates": "FEBRUARY 26 – MARCH 7, 2026",
+  "restaurants": [ ... ]
+}
+```
+
+- `eventDates` — the display string shown in the header when that year is selected; use an em dash (`–`) between the start and end dates
+- `restaurants` — array of restaurant objects (see schema below)
+
 ### Restaurant Object Schema
 
 ```json
@@ -179,7 +207,9 @@ Run the following Python script against the JSON file to identify any restaurant
 import json
 
 with open('2026-restaurants.json', encoding='utf-8') as f:
-    restaurants = json.load(f)
+    data = json.load(f)
+
+restaurants = data['restaurants']
 
 for r in restaurants:
     name = r.get('name', 'Unknown')