Guest WiFi Analytics: From Raw Data to Revenue (2026)

The average restaurant generates more data from its WiFi network in a single Saturday night than most marketing agencies collect from a month of paid ad campaigns. The problem isn't data scarcity. It's that most of that data — device connections, session durations, visit frequencies, movement patterns — hits the access point logs and dies there. Nobody reads AP logs. Nobody builds campaigns from RADIUS accounting tables.

WiFi analytics is the discipline of turning that dead data into business intelligence. Not just pretty dashboards (though those matter for client retention). Actual, revenue-generating intelligence: which guests are at risk of churning, when foot traffic peaks and dips, which marketing campaigns drive repeat visits, and what the real capture rate is across a venue's WiFi network.

This guide covers the full analytics stack — from the raw signals your access points emit to the client-facing dashboards and reports that justify your monthly management fee.

---

The Data Taxonomy: What WiFi Networks Actually Capture

Before you can analyze WiFi data, you need to understand what data exists. WiFi networks generate four distinct categories of data, each with different collection mechanisms, accuracy levels, and privacy implications.

Category 1: Authentication Data (First-Party, Consented)

This is the data guests explicitly provide when they log into the captive portal. It's the highest-quality data because it's identity-verified and consent-backed.

Fields captured at authentication:

• •Email address (typed or via social login)
• •Phone number (SMS OTP or WhatsApp OTP — verified)
• •Name (typed or via social profile)
• •Social profile URL and profile photo (social login only)
• •Gender, age range, locale (social login, where API provides)
• •Device type, OS, manufacturer, model
• •MAC address (device identifier)
• •Authentication timestamp
• •Portal language selection
• •Consent record (checkbox state, timestamp, consent text version)

Authentication data is the gold standard. It links a real human identity to a device, enabling personalized marketing, segmentation, and cross-visit tracking. Every other data category is anonymous or semi-anonymous by comparison.

Category 2: Session Data (RADIUS Accounting)

Every authenticated WiFi session generates RADIUS accounting records. These are the same records that ISPs use for billing — detailed logs of when sessions start, stop, and how much data is transferred.

Fields in RADIUS accounting:

• •Session start time and end time (dwell time calculation)
• •Session duration in seconds
• •Data uploaded and downloaded (bytes)
• •Session termination reason (user logout, timeout, AP disconnect)
• •NAS (Network Access Server) identifier (which AP the guest connected to)
• •Framed IP address
• •Called-Station-Id (BSSID — identifies the specific AP radio)

Session data tells you how long guests stay, how much bandwidth they consume, and which AP they're connected to. For multi-AP venues (hotels, malls, airports), the AP identifier enables zone-level tracking — you can tell which floor, wing, or area the guest spent time in.

Category 3: Presence Data (Probe Requests)

This is where analytics gets interesting — and legally complicated.

Every WiFi-enabled device, whether connected to the network or not, periodically broadcasts probe requests. These are radio frames that essentially ask "Is there a WiFi network here?" Every access point in range receives these probes and can log the sending device's MAC address and signal strength (RSSI).

What presence data captures:

• •MAC address of passing devices (note: modern devices randomize MACs)
• •Signal strength (RSSI) — used to estimate distance from AP
• •Timestamp of each probe
• •Frequency band and channel

What you can derive:

• •Total foot traffic (all devices in range, connected or not)
• •Passersby vs. visitors vs. connected guests (layered by engagement level)
• •Dwell time for non-connected devices (how long they're in range)
• •Repeat visit patterns for non-connected devices
• •Zone-level occupancy (with multiple APs triangulating)
• •Peak hour analysis across the full visitor population

The MAC randomization problem: Starting with iOS 14 and Android 10, most devices randomize their MAC address for probe requests. This means a single iPhone might appear as dozens of different devices across different probe sessions. The practical impact: raw probe request counts overestimate unique visitors by 2–5x depending on the device mix.

Platforms that account for MAC randomization use statistical models — adjusting raw counts based on known randomization rates per OS and device type — to produce more accurate estimates. The estimates aren't perfect, but they're far more useful than raw counts.

Category 4: Behavioral Data (Derived)

Behavioral data is calculated by combining authentication data, session data, and presence data over time. It doesn't come from a single event — it emerges from patterns.

Derived metrics:

• •Visit frequency per guest (weekly, monthly, quarterly)
• •Average dwell time per guest
• •Day-of-week and time-of-day visit patterns
• •Recency-frequency-monetary (RFM) scoring (adapted for visit behavior)
• •Churn risk (declining visit frequency over a rolling window)
• •First visit to second visit conversion rate
• •Campaign response correlation (did the guest visit after receiving an email/SMS?)

Behavioral data is what your clients care about. Raw data answers "what happened." Behavioral data answers "what does it mean, and what should we do about it?"

---

The Analytics Pipeline: From Signal to Insight

Understanding the data pipeline helps you diagnose issues, explain results to clients, and configure analytics correctly.

Stage 1: Collection

Data enters the system from three sources simultaneously:

[Access Points] ──→ Probe requests (presence)
                ──→ RADIUS accounting (sessions)
[Captive Portal] ──→ Authentication events (identity)

Collection happens in real time. Presence data is typically batched at 1–5 minute intervals (depending on the AP vendor's API). Session data flows as RADIUS packets. Authentication data hits the platform instantly when the guest completes login.

Stage 2: Ingestion and Normalization

Raw data from different hardware vendors arrives in different formats. A Meraki CMX Scanning API response looks nothing like a UniFi controller's client statistics export. The platform normalizes all incoming data into a common schema:

• •Standardized device identifiers (handling MAC randomization)
• •Unified timestamps (timezone-normalized)
• •Consistent field names across vendors
• •Deduplication (the same guest appearing on multiple APs simultaneously)

Stage 3: Identity Resolution

This is the critical step that turns anonymous signals into actionable records. When a guest authenticates through the captive portal, the platform links their identity (email, phone, name) to their device's MAC address. From that point forward, every session and presence event from that MAC address is attributed to a known identity.

For returning guests with "Welcome Back" auto-reconnect enabled, the MAC-to-identity link persists across visits. The guest doesn't re-authenticate, but their visit is still logged under their profile.

The identity gap: Not every person in the venue connects to WiFi, and not every WiFi user completes authentication. The ratio of identified guests to total visitors is your identity resolution rate. Typical rates:

Hotels and coworking spaces have high identity rates because WiFi is essential. Shopping malls and airports have low identity rates because most visitors use mobile data. The identity gap is normal — presence analytics fills the gap by providing aggregate insights for the non-identified population.

Stage 4: Aggregation and Analysis

Individual events are aggregated into time-series metrics, cohort analyses, and segmented views. This is where raw data becomes analytics:

• •Hourly/daily/weekly/monthly rollups of visitors, sessions, dwell time
• •Cohort tracking (guests who first visited in January — how many returned in February?)
• •Segment performance (email-captured guests vs. SMS-captured guests — who has higher visit frequency?)
• •Campaign attribution (guests who received Campaign X — did their visit rate change?)

Stage 5: Visualization and Reporting

The final stage renders analytics into dashboards, reports, and exports that you and your clients consume:

• •Real-time dashboard: Live visitor counts, current occupancy, active sessions
• •Historical reports: Daily/weekly/monthly trends, year-over-year comparisons
• •Heatmaps: Visual representation of foot traffic density by zone (requires multiple APs and presence analytics)
• •Scheduled reports: Automated PDF/CSV reports emailed to clients on a configured cadence
• •Data export: CSV, PDF, and JSON webhook for integration with external BI tools

What Most Resellers Get Wrong About the Pipeline

The most common analytics failure is treating the pipeline as a one-way flow: data in, dashboards out. The pipeline should be a feedback loop.

The feedback loop: Analytics → Insight → Action → Measurement → Analytics

Example: The analytics show that a restaurant's Tuesday foot traffic is 40% below the weekly average (insight). The reseller launches a "Taco Tuesday" email campaign targeting guests who've visited on other days but never on Tuesday (action). Two weeks later, the analytics measure whether Tuesday traffic increased among campaign recipients (measurement). The result feeds back into the next campaign optimization cycle.

Resellers who build this feedback loop into their service delivery generate visibly better results for clients — which translates to higher retention and more referrals. Resellers who just install the system and send monthly reports are data collectors, not marketers.

---

RFM Analysis: Borrowing from E-Commerce

RFM (Recency, Frequency, Monetary) analysis is a segmentation technique borrowed from e-commerce and direct marketing. In WiFi marketing, we adapt it as RFD: Recency, Frequency, Dwell.

How RFD Segmentation Works

Recency: When was the guest's last visit? Score 1–5 (5 = visited in the last 7 days, 1 = haven't visited in 90+ days)

Frequency: How often does the guest visit? Score 1–5 (5 = weekly visitor, 1 = single visit ever)

Dwell: How long does the guest typically stay? Score 1–5 (5 = 2+ hours, 1 = under 15 minutes)

Each guest gets a 3-digit RFD score. A guest scoring 5-5-4 is your best customer: they visited recently, visit frequently, and stay for extended periods. A guest scoring 1-1-1 is a one-time visitor who left quickly and hasn't returned.

Segments and Actions

RFD segmentation turns a flat guest list into an actionable customer portfolio. Each segment gets different messaging, different offers, and different campaign timing. The "Champions" segment gets loyalty recognition. The "At Risk" segment gets a win-back offer with urgency.

Practical implementation: Most WiFi marketing platforms don't have built-in RFM/RFD scoring. You calculate it by exporting guest data (last visit date, visit count, average session duration) to a spreadsheet and applying the scoring formula. For larger databases (5,000+ records), use a simple script or Zapier workflow to calculate scores automatically.

---

Data Quality: The Silent Analytics Killer

Analytics are only as good as the data feeding them. Bad data in, misleading analytics out. Here are the data quality issues specific to WiFi analytics and how to address them.

Issue 1: Fake Email Addresses

10–15% of email-form captures produce fake or disposable email addresses. Guests type test@test.com, asdf@gmail.com, or use throwaway services like Guerrilla Mail. These records pollute your database and inflate your capture count.

Fix: Email validation at the point of capture (reject obviously fake formats), periodic list hygiene (bounce-based cleanup), and — most effectively — using verified authentication methods (SMS OTP, WhatsApp OTP) that inherently produce real, validated contact data.

Issue 2: MAC Address Randomization Overcounting

Presence analytics that don't account for MAC randomization will overccount unique visitors by 2–5x. A device that randomizes its MAC address every 15 minutes appears as 4 "new" devices per hour.

Fix: The platform's presence analytics engine should apply a randomization adjustment model. If it doesn't, raw presence counts are directionally useful (trends over time) but not accurate in absolute terms.

Issue 3: Duplicate Guest Records

The same person connecting from their phone and their laptop creates two guest records. A guest who re-authenticates after a factory reset (new MAC address) creates a duplicate. Over time, duplicates accumulate and inflate the database size.

Fix: De-duplication by email address or phone number. If the same email appears in two records (different MACs), merge the records. Most platforms handle this automatically, but verify by searching for known duplicates periodically.

Issue 4: Session Duration Artifacts

A guest who falls asleep with their phone connected to WiFi generates a 6-hour session. A guest whose device maintains a background WiFi connection while they leave the building (common if the WiFi signal reaches the parking lot) generates a phantom extended session.

Fix: Apply reasonable session caps (e.g., cap dwell time at 4 hours for restaurants, 24 hours for hotels) and filter outliers from dwell time calculations. Median dwell time is more robust than mean dwell time for this reason.

---

The KPIs That Matter: What to Put on the Dashboard

Not all metrics belong on a client-facing dashboard. Overwhelming clients with data is worse than showing them nothing. The art is selecting the KPIs that directly connect to business outcomes.

The Core Six KPIs

These six metrics should appear on every client dashboard, regardless of vertical:

1. Total Guest Captures (Period) How many new guest records were captured this week/month. This is the top-line growth metric. Trend arrow (up/down vs. previous period) provides immediate context.

2. New vs. Returning Guest Ratio Percentage of connected guests who are first-time vs. returning. A restaurant seeing 70% new guests has an acquisition success but a retention problem. A restaurant seeing 70% returning guests has loyalty but may be stagnating on new customer acquisition. The healthy range depends on the venue, but 40–60% returning is a common target for food and beverage.

3. Average Dwell Time How long guests stay connected, averaged across all sessions. Longer dwell time correlates with higher spend in retail and hospitality. Track changes over time and after campaign launches to measure impact.

4. Capture Rate Percentage of WiFi-connecting devices that complete authentication. This measures the captive portal's effectiveness. Below 25% means the portal needs optimization (see the Captive Portal Guide). Above 40% is strong.

5. Visit Frequency (30-day) Average number of visits per identified guest over a rolling 30-day window. This is the retention metric. For restaurants, 2–3 visits/month indicates a loyal customer. For retail, even 1.5 visits/month is strong.

6. Campaign Engagement Rate Open rate and click rate for email/SMS campaigns sent through the platform. Compare against baseline (industry average email open rate: 20–25%; WiFi-captured lists typically perform 30–40% because the relationship is venue-specific).

Vertical-Specific KPIs

Beyond the core six, add metrics relevant to the client's industry:

Restaurants: Peak hour distribution (helps with staffing), post-campaign visit attribution (did guests who received the Tuesday offer visit on Tuesday?), birthday redemption rate

Hotels: Check-in WiFi adoption rate, pre-checkout review prompt conversion, loyalty program enrollment via WiFi

Retail: Foot traffic vs. transaction correlation (requires POS integration or manual cross-referencing), fitting room vs. checkout area dwell time (multi-AP zones), weekend vs. weekday traffic patterns

Shopping Malls: Tenant-level foot traffic distribution, floor-by-floor occupancy, average visit duration by day of week, cross-shopping patterns (visitors who pass multiple tenants)

Events/Conferences: Session-level attendance tracking, exhibitor booth dwell time, post-event contact list size, networking density metrics

---

Presence Analytics: The Full Picture

Presence analytics extends beyond authenticated guests to capture the entire visitor population. It answers the question: "How many people are in or near this venue, whether they're using our WiFi or not?"

How Presence Analytics Works

1. •

   Probe request capture. APs listen for probe requests from all nearby devices (within radio range, typically 30–100 meters depending on AP power and environmental factors).

2. •

   Device classification. Each detected device is classified:

   • •Passerby — Detected briefly (under 5 minutes). Likely walking past the venue, not entering.
   • •Visitor — Detected for an extended period (5+ minutes). Likely inside or immediately adjacent.
   • •Connected — Authenticated on the network. Positively identified.
3. •

   Deduplication and MAC randomization adjustment. Statistical models reduce overcounting from randomized MAC addresses.

4. •

   Aggregation. Hourly, daily, and weekly tallies of passersby, visitors, and connected guests.

The Engagement Funnel

Presence analytics creates a conversion funnel for physical spaces:

Passersby (100%)
    ↓
Visitors (30-50% of passersby enter the venue)
    ↓
WiFi Connectors (40-70% of visitors connect to WiFi)
    ↓
Authenticated Guests (30-60% of connectors complete login)
    ↓
Marketing-Engaged Guests (15-25% of authenticated click/open campaigns)
    ↓
Repeat Visitors (20-40% of authenticated return within 30 days)

Each stage of the funnel is optimizable. Low visitor-to-connector ratio? Maybe the WiFi network name isn't visible enough, or the connection is unreliable. Low connector-to-authenticated ratio? The captive portal needs work. Low repeat visitor rate? Marketing automation needs attention.

What most resellers get wrong: they report only on authenticated guests — the bottom of the funnel. Presence analytics lets you report on the entire funnel, which demonstrates a much larger opportunity to the client. "You had 5,000 visitors this month. We captured 800 of them. Here's our plan to get that to 1,500."

Heatmaps

For venues with multiple access points, presence analytics can generate heatmaps — visual representations of foot traffic density across the physical space.

How it works: each AP has a known physical location. Device signal strength (RSSI) at each AP allows triangulation of the device's approximate position. Aggregate thousands of position estimates over time, and you get a density map.

Heatmap accuracy: WiFi-based heatmaps are zone-level accurate, not meter-level accurate. You can tell that the west wing of a mall has 3x more foot traffic than the east wing. You can't tell that aisle 7 has more traffic than aisle 8 within a single store. For meter-level accuracy, you need BLE beacons or camera-based systems — WiFi is a zone-level tool.

Client value of heatmaps: Mall operators use them for tenant lease negotiations (high-traffic zones command higher rent). Retail stores use them for merchandise placement. Event venues use them for sponsor booth positioning. Airports use them for terminal congestion management.

For a deep dive into heatmap implementation, see our WiFi Heatmap Analytics guide.

---

API Integration: Getting Data Into External Systems

The WiFi analytics platform is one piece of a larger data ecosystem. For many clients — especially enterprise-tier accounts — the value isn't in the platform's built-in dashboards. It's in the data flowing out to their existing systems.

Integration Methods

Webhooks (Real-Time) The platform sends JSON payloads to a URL you configure whenever specified events occur: new guest authentication, session start/end, campaign engagement. Webhooks are event-driven and near-real-time (sub-second delivery).

Use case: A hotel wants new guest WiFi signups to appear in their CRM within seconds of authentication. A webhook fires on the "guest authenticated" event, posting the guest's name, email, and room association to the CRM's API.

REST API (On-Demand) The platform exposes API endpoints for querying guest records, analytics summaries, campaign data, and configuration. API calls are authenticated (API key or OAuth) and rate-limited.

Use case: A data engineering team pulls daily analytics into their data warehouse for cross-referencing with POS data, loyalty program data, and ad campaign data.

Zapier (No-Code) For clients without development teams, Zapier connects the WiFi platform to 1,000+ apps without writing code. Triggers: new guest, campaign sent, etc. Actions: create HubSpot contact, add Mailchimp subscriber, post Slack notification, create Google Sheets row.

Use case: An agency wants every new WiFi signup to appear in their client's Mailchimp list automatically. A Zapier workflow connects "New Guest" trigger to "Add Subscriber" action.

CSV/PDF Export (Batch) Scheduled data exports for manual analysis or import into systems that don't support APIs. Reports can be configured to auto-email to clients on a daily, weekly, or monthly schedule.

Data Architecture for Scale

For resellers managing 50+ locations with API integrations, the data architecture matters:

[WiFi Platform API]
        ↓ (webhooks + scheduled pulls)
[Data Warehouse / ETL]
        ↓
[BI Dashboard (Tableau, Looker, Power BI)]
        ↓
[Client-Facing Reports]

This architecture decouples the WiFi platform from the reporting layer. You can cross-reference WiFi data with POS data, loyalty data, weather data, and event calendars to produce insights that no single system could generate alone.

---

Analytics by Vertical: What Each Industry Cares About

Not every metric matters to every client. The art of WiFi analytics for resellers is selecting the right metrics for the right vertical and presenting them in a way that connects to business outcomes the client already cares about.

Restaurants: The Repeat Visit Engine

Restaurants care about one thing above all others: getting guests to come back. Every other metric is in service of that goal.

Primary dashboard metrics:

• •Visit frequency distribution (what percentage of guests visit 1x, 2x, 3x+ per month)
• •Capture rate (is the portal working?)
• •Campaign-to-visit attribution (did the Tuesday email drive Tuesday traffic?)
• •Review volume trend (are review prompts generating results?)

The insight that sells: "Your top 20% of guests — the ones who visit 3+ times per month — account for 55% of your WiFi connections. Your bottom 40% — one-time visitors — account for only 12% of connections. The re-engagement campaigns we're running target that bottom 40% to move them into the repeat-visitor segment."

A 3-location taco chain in the southwest tracked this metric religiously. After 6 months of WiFi marketing with re-engagement campaigns, their one-time visitor rate dropped from 62% to 48% — meaning 14% of previously one-time guests became repeat visitors. At $28 average ticket, the attributable revenue was approximately $4,200/month across 3 locations.

Hotels: The Review and Rebooking Machine

Hotels measure success through average daily rate (ADR), occupancy, and online reputation (star ratings and review volume). WiFi analytics serves all three.

Primary dashboard metrics:

• •WiFi adoption rate (percentage of guests who connect — should be 85%+)
• •Post-stay review completion rate
• •Review sentiment trend (average rating of WiFi-prompted reviews)
• •Rebooking campaign conversion rate

The insight that sells: "Since deploying WiFi marketing, your TripAdvisor review volume increased from 12/month to 45/month. Your average rating moved from 4.1 to 4.3. According to Cornell Hospitality Research, a 1-point increase in review score allows a hotel to raise ADR by 11.2% without losing occupancy. Even a 0.2-point improvement at your $180 ADR and 72% occupancy represents significant incremental revenue."

Shopping Malls: The Tenant Intelligence Platform

Mall operators don't care about individual guest profiles. They care about aggregate foot traffic intelligence that helps them manage tenant leases, optimize common area programming, and measure marketing campaign effectiveness.

Primary dashboard metrics:

• •Total foot traffic by zone/floor (presence analytics)
• •Tenant area dwell time distribution
• •Peak hour patterns (daily, weekly, seasonal)
• •Shopper capture rate (what percentage of foot traffic becomes an identified profile)

The insight that sells: "Zone A (anchor tenant wing) saw 34% more foot traffic than Zone B (specialty retail) this quarter. But Zone B's average dwell time is 40% longer, suggesting higher engagement per visitor. This data supports repositioning the Zone B lease rates based on engagement quality rather than raw traffic volume."

Events and Conferences: The Attendee Intelligence Report

Event organizers need post-event analytics that justify sponsor spend and inform next year's programming.

Primary dashboard metrics:

• •Total unique WiFi connections (attendee count proxy)
• •Peak concurrent connections (capacity planning)
• •Session/zone dwell time (which sessions/exhibits drew the most engagement)
• •Contact capture volume (database size for post-event marketing)

The insight that sells: "The keynote session drew 1,847 concurrent WiFi connections — 92% of registered attendees. The breakout sessions averaged 340 connections. Exhibit Hall Zone 3 had 2.1x the dwell time of Zone 1, suggesting exhibitor placement in Zone 3 should command premium booth pricing."

---

Privacy Compliance: The Analytics Dimension

Analytics creates privacy obligations beyond what the captive portal consent covers. Specifically, presence analytics — which captures data from devices that haven't opted in — sits in a regulatory gray area.

Presence Analytics and GDPR

Under GDPR, even randomized MAC addresses can be considered personal data if there's any possibility of re-identification. The European Data Protection Board (EDPB) has guidance on WiFi tracking that resellers operating in Europe need to understand:

1. •

   Legitimate interest basis. Aggregated, anonymized foot traffic counts (total visitors per hour) are generally defensible under legitimate interest. No individual is identified or identifiable.

2. •

   Purpose limitation. Presence data collected for aggregate analytics (foot traffic, occupancy) cannot be repurposed for individual tracking without consent.

3. •

   Data minimization. Capture the minimum data needed. Hashing or immediately discarding raw MAC addresses after aggregation reduces privacy risk.

4. •

   Transparency. Even for aggregate analytics, venues should display signage informing visitors that WiFi analytics are in operation. Some DPAs require this explicitly.

CCPA and Presence Data

California law is less prescriptive about presence analytics than GDPR, but the principle holds: if you're collecting data that could identify an individual, you need to disclose the collection and provide an opt-out mechanism.

Best Practices for Privacy-Compliant Analytics

1. •

   Separate authenticated analytics from presence analytics. Authenticated data has consent. Presence data does not. Keep the analysis separate and never attempt to correlate unconsented presence data with identified guest records.

2. •

   Aggregate, don't individualize. Report "250 visitors between 2–3 PM" not "Device AA:BB:CC was in the east wing for 12 minutes." Aggregate data is almost always compliant. Individual device tracking without consent is almost always problematic.

3. •

   Respect MAC randomization. Don't attempt to de-randomize MAC addresses. The randomization exists specifically to prevent tracking. Working around it violates the spirit (and potentially the letter) of privacy laws.

4. •

   Post signage. A simple notice — "This venue uses WiFi analytics to measure foot traffic. No personal data is collected from non-connected devices." — addresses transparency requirements.

5. •

   Establish retention limits. Presence data should have a shorter retention period than authenticated data. 90 days for presence, 12–24 months for authenticated records is a common configuration.

---

Building the Client Report

The analytics dashboard is the platform. The client report is the product. The report is what your client sees once a month, and it's what justifies their continued payment.

Report Structure

Page 1: Executive Summary

• •Total guests captured this period
• •New vs. returning ratio
• •Highlight metric (biggest win: "Your email list grew 12% this month")
• •One action item for next month

Page 2: Traffic Overview

• •Daily visitor chart (line graph, 30-day view)
• •Peak days and hours
• •Foot traffic trend (up/down vs. previous period)
• •Presence analytics summary (total visitors vs. captured)

Page 3: Guest Profile

• •Top devices (iOS vs. Android split)
• •Authentication method breakdown
• •New guest acquisition by week
• •Visit frequency distribution

Page 4: Campaign Performance

• •Campaigns sent this period
• •Open rates, click rates
• •Best performing campaign (by engagement)
• •Post-campaign visit attribution (if trackable)

Page 5: Recommendations

• •Portal optimization suggestion (based on capture rate trends)
• •Campaign suggestion (based on inactive guest count)
• •Expansion opportunity (if applicable)

Report Automation

Manual report creation for 20+ clients is unsustainable. The platform should generate reports automatically on a configured schedule (weekly or monthly) and email them directly to clients. Your only task is reviewing the reports before they send (or configuring auto-send if you trust the format).

Automated reports are one of the highest-ROI features for resellers. They demonstrate ongoing value without ongoing effort.

---

Advanced Analytics: What's Next

Predictive Analytics

Machine learning models trained on historical visit patterns can predict future behavior:

• •Churn prediction: Guest hasn't visited in 10 days but typically visits weekly → flag for re-engagement campaign
• •Peak forecasting: Based on 6 months of historical data, predict next Tuesday's foot traffic within 15% accuracy
• •Campaign timing optimization: Send campaigns at the time each guest is most likely to open (based on historical engagement patterns)

Predictive analytics is available as an add-on on Agency and MSP tiers, and it's where the platform starts to feel less like a reporting tool and more like an AI-powered marketing engine.

AI Agents and MCP

The Model Context Protocol (MCP) enables AI agents to query WiFi data directly. Instead of a human reviewing a dashboard and deciding what to do, an AI agent can:

• •"What's the predicted guest count at Location A next Saturday?"
• •"Which guests haven't visited in 21+ days but visited 3+ times before?"
• •"Generate a re-engagement campaign for guests who last visited in February."

MCP server access is available on the MSP tier. It's early-stage, but it represents the future of WiFi analytics: AI agents that turn data into action without human intervention.

---

Further Reading

• •WiFi Marketing: The Definitive Guide — The complete WiFi marketing overview
• •Captive Portal Guide — Optimizing the data capture layer
• •WiFi Marketing ROI Guide — Turning analytics into revenue proof
• •WiFi Presence Analytics Explained — Deep dive into presence technology
• •WiFi Heatmap Analytics Implementation — Heatmap setup and use cases
• •WiFi Analytics API Integration — Technical API guide
• •Guest WiFi Data Pipeline Architecture — Data engineering for WiFi analytics
• •RADIUS Analytics Deep Dive — RADIUS accounting for session analytics
• •Real-Time Venue Analytics Guide — Live dashboard configuration
• •Guest WiFi Analytics ROI Guide — Proving analytics value to clients
• •GDPR WiFi Data Compliance — Privacy compliance for analytics
• •WiFi Hardware Guide — Hardware capabilities affect analytics depth
• •Location-Based Marketing Guide — WiFi analytics in the broader location intelligence landscape
• •AI-Powered Guest Segmentation — Machine learning applied to WiFi data
• •WiFi Data Capture: State of 2026 — Industry trends in data collection