A complete tour of the admin panel. The AI Concierge admin lives under HomeRunner → AI Concierge in your WordPress dashboard.
Tab map #
The admin panel is organized into tabs. This guide walks through each one in the order you’re likely to use it.
| Tab | What you do here |
|---|---|
| General | Master enable/disable, debug mode, voice-input language |
| Performance | Cost dashboard, booking attribution, response times |
| License | License activation, deactivation, trial status |
| Search Prompt | Tune which property attributes the AI prioritizes |
| Concierge | Brand voice + recovery messaging |
| Chatbot | On-page chat drawer + property page experience |
| Property KB | Bulk per-property knowledge base editor |
| Self-Assessment | Quality loop where the AI grades its own responses |
| Photo Analysis | One-time vision pass over property photos |
| Test Search | Operator sandbox for dry-running queries |
| Chat Sessions | Review every chat conversation |
| Search History | Analytics over guest search queries |
General #
The master switch and basic operating preferences.
- Enable Enhancer: site-wide kill switch. Off = the Concierge does nothing, your HomeRunner Core search behavior is unchanged.
- Debug Mode: adds verbose logging to your PHP error log. Leave off for production.
- Voice Input Language: language preference for the voice-input feature on the chat drawer. Default is “auto” (let the browser decide).
Save settings before leaving the tab.
Performance #
A live dashboard showing the business value AI Concierge is delivering. This is what you’d show your business partner at renewal time.
Metrics displayed:
- Token usage and dollar cost per call: rolled up daily, weekly, monthly
- Average response time: median and 95th percentile
- AI-assisted bookings: count and revenue attribution for bookings where the guest interacted with the Concierge in the same session
- Query volume: searches and chat messages per period
- Top intents: what guests are actually asking for
Use this tab to:
- Justify the subscription at renewal
- Spot months where cost grew faster than volume (often a sign of a misconfigured prompt or runaway KB)
- See which query types convert best so you can lean into them in your marketing copy
License #
License activation, deactivation, and trial status.
- Activate: paste your key here on first use
- Check License: manually re-validate (useful if something seems off; the plugin re-validates automatically twice a day)
- Deactivate: releases the activation slot for this site; useful when migrating to a new domain
Trial sites show the trial end date prominently. The plugin gives a 30-day grace period after trial end before it disables, so you have time to subscribe without an interruption.
Search Prompt #
This tab controls how the AI interprets search queries. Two halves:
Top half: Search Behavior toggles Site-wide rules like “always require at least one bedroom” or “respect explicit date constraints.” Defaults are sensible and most operators don’t need to change these.
Bottom half: Prompt Fields table For each property meta field (bedrooms, max_guests, pet_friendly, location, amenities, etc.) you can:
- Toggle whether the AI considers it
- Mark it as a “hard filter” (must match exactly) vs. “soft preference” (the AI uses it as a tiebreaker)
There’s also a Prompt Preview section at the bottom that shows the actual system prompt the AI sees, with placeholders for runtime data. Useful for verifying changes do what you expect.
Concierge #
The most important configuration tab for response quality.
- Brand Voice: a free-text field where you describe how the Concierge should sound. See Getting Started for examples and guidance. Specific input produces specific output. “Warm but professional, drop coffee shop references” beats “friendly tone” every time.
- Concierge Lessons Panel: over time the Self-Assessment loop surfaces patterns where the AI is off-target. This panel shows those lessons and lets you accept them into the prompt.
Chatbot #
Configure the on-page chat drawer guests interact with.
Drawer settings #
- Position: left or right side of the screen
- Greeting: the opening message guests see when they open the drawer
- Tooltip: optional small nudge that appears next to the chat tab after N seconds on the page
Recovery & Outreach #
When a search returns zero results or too many results, the Concierge offers proactive recovery:
- Alternative-date chips: “Your dates aren’t available, but the same property is open the week before”
- Lead-capture prompt: for queries that genuinely exceed the catalog (“we couldn’t find that, want to tell us more so a human can follow up?”)
Property Page Experience (new in v1.49) #
A site-wide toggle for the property-aware chat behavior:
- Enable on property pages: when ON, the chat drawer opens on individual listings with a property-specific greeting, three contextual chips, and the property’s knowledge base injected into the prompt
- PMS Knowledge Base custom field: dropdown of all property meta keys. The Concierge reads this field per property and includes it in the prompt. Default:
homerunner_concierge_kb. If your PMS doesn’t sync a knowledge field, leave this and use the Property KB tab to author content manually.
Notification dot and 3-pulse animation draw guest attention to the chat tab on property pages without being intrusive.
Property KB #
New in v1.49.2. A bulk admin browser for per-property local knowledge.
The table shows every published property with:
- Status: Has KB or Empty
- Character count: how much KB content is stored
- Preview: first 18 words of the entry
- Actions: Edit (inline), Clear (with confirmation)
Toolbar:
- Filter by title: type-ahead search
- All / With KB / Empty: quick filter by status
- Count summary showing how many properties have KB authored
Inline edit row uses the same UX as the per-post metabox:
- 4,000-character hard cap
- Soft warning at 3,000 characters
- Live character counter
When to use the Property KB tab vs. the per-post metabox:
- Property KB tab: when you want to see all properties at once, batch-fill empty ones, or quickly review preview snippets
- Per-post metabox (on the Property edit screen): when you’re already editing a specific property’s other content
Both write to the same underlying field; they’re two views of one data store. Edits made in one show up in the other.
Content guidance:
- Cover things not in the listing description: house rules, lock codes, owner favorites, area tips
- Don’t repeat the listing description; that’s already in the prompt
- Keep entries under 1,500 characters when possible (long entries slow responses)
- Use bullet points or short paragraphs, not walls of text
Self-Assessment #
A quality loop where the Concierge grades its own responses.
Periodically (you control the schedule) the AI re-reads a sample of recent chat responses against operator-defined criteria. For example: “Did the response stay on-brand?”, “Did it recommend a property that actually matches the constraints?”, “Did it avoid claims about features the property doesn’t have?”
Outputs:
- Quality dashboard: pass/fail scores over time
- Concierge Lessons: when patterns of failure are detected, the loop drafts a correction (e.g. “When guests ask about ‘family-friendly,’ prioritize properties with sleep_count >= 6 even if not explicitly tagged family-friendly”). You review and accept these lessons into the prompt.
This tab is the foundation of how AI Concierge gets better at your specific catalog over time. Worth checking every 1–2 weeks once you’re settled in.
Photo Analysis #
A one-time vision pass over every property photo that extracts amenities, vibe descriptors, and feature callouts the AI can use during search.
Once you click Analyze All Properties, the job runs in the background:
- 26–50 listings → 5–10 minutes
- 51–200 listings → 10–25 minutes
- 200+ listings → up to 60 minutes
You can navigate away. New properties added later are picked up automatically. The dashboard at the bottom of the tab shows progress and lets you re-run for individual properties if needed.
Output is invisible to guests; it’s enrichment that quietly makes search smarter.
Test Search #
An operator sandbox. Type any query as if you were a guest; see the full Concierge response including which properties it would have surfaced.
Use this to:
- Verify changes to brand voice, KB, or search-prompt fields before they affect real guests
- Compare two phrasings of the same query to understand how the AI reasons
- Show prospects what the Concierge can do during a sales call
- Triage support questions when a guest claims “your AI told me X”
Test Search does not log to Chat Sessions or Search History; it’s a sandbox, not a production channel.
Chat Sessions #
A browsable log of every real chat conversation guests have had with the Concierge.
Each session shows:
- Timestamp and session duration
- Every message in the conversation (guest and Concierge)
- The properties that were surfaced (if any)
- Whether the session led to a booking (when attribution data is available)
Use Chat Sessions to:
- Pull a transcript when a guest references something the Concierge said
- Spot underserved query types (“lots of guests ask about EV charging, should we surface that more?”)
- Train your eye for prompt improvements
Search History #
The historical record of every AI-routed search query and how it resolved.
Columns:
- Timestamp
- Original query (verbatim)
- AI’s interpretation (the parsed parameters it applied)
- Result count
- Whether the search led to a property view or booking
Search History is your single best source of marketing intelligence: exactly what guests are looking for, in their own words. Use it for SEO, ad-copy, blog topics, and product decisions about which destinations to expand.
Next steps #
- Looking for setup guidance? → Getting Started
- Questions about data handling? → Privacy, Data, and Cost Controls
- Something not working? → FAQ & Troubleshooting
Last updated: 2026-05-17 · AI Concierge v1.49.4