Mesures d'analyse WiFi qui comptent vraiment pour le commerce de détail

WiFi Analytics Metrics That Actually Matter for Retail A Purple Intelligence Briefing — approximately 10 minutes --- INTRODUCTION & CONTEXT (approx. 1 minute) --- Welcome to the Purple Intelligence Briefing. I'm your host, and today we're cutting straight to the point on a topic that comes up in almost every conversation I have with retail operations directors and IT teams: WiFi analytics metrics. Specifically — which ones actually matter, and which ones are just noise. Most platforms will hand you a dashboard full of numbers. Total connections. Bandwidth consumed. Peak concurrent users. And while those figures have their place in a network capacity conversation, they tell you almost nothing about what's happening on your shop floor, how long customers are staying, or whether they're coming back. So in the next ten minutes, we're going to walk through the metrics that genuinely correlate with retail revenue, dwell time, and customer loyalty. We'll look at how to translate raw WiFi data into business intelligence, and I'll give you a practical framework for aligning your IT team and your marketing team on a single, shared dashboard. Let's get into it. --- TECHNICAL DEEP-DIVE (approx. 5 minutes) --- Let's start with the most fundamental metric in retail WiFi analytics: footfall. Footfall, in a WiFi context, is the count of unique devices detected within your venue over a given time period. Now, this is distinct from the number of WiFi connections. A platform like Purple's WiFi Analytics uses passive probe detection — meaning it can detect devices that haven't connected to the network at all. That's a critical distinction. If you're only counting connected users, you're potentially missing sixty to seventy percent of the people actually in your store. The two sub-metrics that matter most within footfall are new versus returning visitors. A new visitor is a device seen for the first time. A returning visitor is a device that has been detected previously. That split immediately tells you something about your marketing effectiveness. If your new visitor rate is consistently above eighty percent, you're not retaining customers — you're running a leaky bucket. If your returning rate is above forty percent, you have a loyalty story to tell. Now, footfall alone is a vanity metric unless you pair it with dwell time. Dwell time is the duration a device — and by proxy, a customer — spends within your venue or within a specific zone. This is where WiFi analytics starts to earn its keep. The research is consistent across retail environments: customers who spend more than eight minutes in a store spend, on average, two to three times more than those who spend under five minutes. That's not a small effect. That's a fundamental driver of basket size. The key dwell time thresholds to benchmark against are these. Under three minutes is a bounce — the customer came in, didn't engage, and left. Three to eight minutes is a browse. Eight to fifteen minutes is an engaged visit. Over fifteen minutes typically indicates either a high-value customer or a friction point — like a queue — and you need to know which one it is. Zone-level dwell time is where this gets really powerful. If you've deployed access points across distinct areas of your store — entrance, apparel, electronics, café, checkout — you can measure dwell time per zone independently. A high dwell time at checkout with no corresponding increase in transaction value is a queue problem. A high dwell time in your premium product zone is a conversion opportunity. These are operationally very different situations, and without zone-level data, you can't tell them apart. The third tier of metrics is what I'd call engagement rate — the percentage of detected devices that actually connect to your guest WiFi network. This is your data capture funnel. A well-designed captive portal with a frictionless login flow — social login, email, or a one-tap option — should convert somewhere between twenty-five and forty percent of detected devices into identified profiles. If you're below fifteen percent, your portal experience needs attention. If you're above fifty percent, you're likely in a venue with a captive audience — a transport hub, a stadium, or a food court — where WiFi is a genuine utility. The fourth metric tier is the one most retail teams underinvest in: cohort-based repeat visit analysis. A cohort, in this context, is a group of visitors who first appeared in your venue during a specific time window — say, January 2025. Cohort analysis then tracks what percentage of that group returned within seven days, thirty days, and ninety days. This is the retail equivalent of a customer lifetime value calculation, but derived entirely from WiFi signal data — no loyalty card required, no app install needed. A healthy retail cohort typically shows a seven-day return rate of around thirty to forty-five percent for convenience or food-and-beverage retail, dropping to fifteen to twenty-five percent for fashion or general merchandise. If your ninety-day cohort retention is below ten percent, you have a loyalty problem that no amount of footfall growth will fix. The fifth and final metric tier is revenue correlation — and this is where IT and marketing finally speak the same language. The formula is straightforward: multiply your daily footfall by your average dwell time, then apply your known conversion rate and average transaction value. What you get is a revenue proxy that you can track over time. When footfall increases but revenue doesn't, your conversion rate or basket size is the problem. When dwell time drops, you can expect revenue to follow within two to three weeks — it's a leading indicator. Purple's analytics platform surfaces all five of these tiers in a unified dashboard, allowing operations directors to correlate network data with POS data without requiring a custom data engineering project. --- IMPLEMENTATION RECOMMENDATIONS & PITFALLS (approx. 2 minutes) --- Right, let's talk about how you actually deploy this in practice — and where teams typically go wrong. The most common mistake I see is deploying WiFi analytics as a network tool rather than a business intelligence tool. The IT team installs the access points, configures the SSID, and hands over a login to the dashboard. Marketing then looks at it once, doesn't know what to do with it, and it becomes shelfware. The fix is to define your KPI framework before deployment, not after. Agree with your marketing and operations stakeholders on the five or six metrics that will appear on the shared dashboard. Everything else is secondary. The second pitfall is poor access point placement. For accurate zone-level dwell time measurement, your access points need to be positioned to create distinct detection zones — not just to provide coverage. This often means deploying more APs than a pure coverage calculation would suggest, particularly in large-format stores. Work with your network architect to overlay the coverage plan against the store's zone map before installation. Third: GDPR and data minimisation. Under GDPR Article 5, you must collect only the data necessary for your stated purpose. For WiFi analytics, that means your captive portal data capture must be tied to a clear, specific consent statement. MAC address randomisation — which is now default on iOS 14 and above and Android 10 and above — means that passive probe data is less reliable for individual tracking than it was three years ago. Your platform needs to handle this gracefully, either through authenticated session data or through statistical normalisation. Purple's platform accounts for randomised MAC addresses in its footfall calculations, which is something to verify with any vendor you're evaluating. Finally, on the integration side: the real ROI from WiFi analytics comes when you connect it to your other data sources. A CRM integration allows you to match WiFi profiles to known customers. A POS integration allows you to close the loop between dwell time and actual spend. Neither of these is technically complex — both Purple and most enterprise WiFi platforms offer standard API connectors — but they require a data governance conversation upfront. Define your data ownership, your retention periods, and your consent chain before you start joining datasets. --- RAPID-FIRE Q&A (approx. 1 minute) --- Let me run through a few questions that come up regularly. "How many access points do I need for accurate analytics?" — For a standard retail unit of up to five hundred square metres, three to four APs positioned to create overlapping but distinct detection zones is a reasonable starting point. Larger formats need a proper RF survey. "Can I use WiFi analytics without a captive portal?" — Yes. Passive probe detection works without any user interaction. But you lose the ability to build identified profiles, which limits your cohort analysis and CRM integration. The captive portal is what turns anonymous signal data into actionable customer intelligence. "What's a realistic timeline to see ROI?" — Most retail deployments see meaningful data within the first thirty days. Cohort analysis becomes statistically significant after ninety days. Full revenue correlation modelling typically takes one quarter of clean, integrated data. "Does WiFi analytics replace footfall counters?" — It complements them. Traditional door counters give you entry events. WiFi analytics gives you dwell time, zone behaviour, and repeat visit data. Use both where budget allows; prioritise WiFi analytics if you have to choose one. --- SUMMARY & NEXT STEPS (approx. 1 minute) --- To wrap up: the five WiFi analytics metrics that actually matter for retail are footfall — specifically new versus returning split — dwell time at both venue and zone level, engagement rate through your captive portal, cohort-based repeat visit analysis, and revenue correlation as a composite leading indicator. The implementation principles are: define your KPI framework before deployment, position APs for zone detection not just coverage, handle MAC randomisation correctly, and integrate with POS and CRM to close the revenue loop. If you're evaluating platforms, the questions to ask are: how does the platform handle randomised MAC addresses, does it support zone-level dwell time natively, and what does the cohort analysis output look like out of the box? Purple's WiFi Analytics platform is built specifically around these retail use cases — footfall, dwell time, and cohort repeat-visit data are core to the product, not bolt-ons. For the full technical reference guide, including worked examples, KPI benchmarks, and a decision framework for aligning IT and marketing on a shared dashboard, visit purple.ai. Thanks for listening. Until next time. --- END OF SCRIPT ---