Presence analytics vs engagement analytics

Choosing between presence and engagement analytics depends on your specific goals: use presence for overall footfall volumes or engagement for marketing. Purple supports both CCPA/CPRA standards on networks like Cisco Meraki across 80,000+ venues, helping estates combine location data with opted-in guest profiles to drive operations.

Presence analytics vs engagement analytics: which should you choose?

The choice follows the question you are trying to answer.

• "How many people, where and when?" That is a presence question. It covers visit volumes, peak hours, dwell time and which sites perform best and worst.
• "Who are they, will they come back, and can we reach them?" That is an engagement question. It covers known visitors, repeat behavior and opted-in contacts you can market to.

Presence analytics sees everyone with a WiFi-enabled device, but it cannot tell you who they are. Engagement analytics tells you exactly who signed in, but only for the share of visitors who choose to connect. Neither covers the other's blind spot. That is why most multi-site operators run both from the same network.

Purple runs both from one WiFi network. It runs as a hardware-agnostic cloud overlay on Cisco Meraki, HPE Aruba, Ruckus, Juniper Mist, Ubiquiti UniFi, Cambium, Extreme and Fortinet. Purple operates across 80,000+ live venues globally and recorded 440 million logins in 2024 (Purple's own data).

Where do presence and engagement analytics genuinely differ?

How presence analytics works

A smartphone with WiFi switched on regularly sends probe requests. These are short radio frames that ask nearby networks to identify themselves. Your access points hear these frames and record the received signal strength indicator (RSSI), a measure of how strong the signal is. Strong, sustained signals suggest the device is inside the venue. Weak or brief ones suggest a passer-by.

Purple's Presence report turns this into a set of operational measures:

• Conversions: the total number of visits where signal strength indicated the device was inside the venue.
• Average dwell: how long authenticated visitors stayed connected per visit.
• Visits by day and hour: a heatmap of visit volume across the week.
• Most and least popular venues: your top 10 and bottom 10 sites by visits. Venues with zero visits are excluded.

The full field definitions are in the WiFi Analytics: Presence support article.

How engagement analytics works

A captive portal is the web page a visitor sees before they get internet access. On that page, the visitor signs in with an email address, a form or a social login. They also make a conscious-choice opt-in to marketing if they want one. Every visit after that is tied to a known profile with first-party data the visitor gave you directly.

That profile changes what you can measure. You can count new versus returning visitors and segment by the attributes they shared. You can then reach the opted-in group through your CRM. Purple's Zone Analytics can filter zone-to-zone movement by gender or age. That filter relies on sign-in data, so it is an engagement capability layered onto location data. See the Zone Analytics support article for the report definitions.

Side-by-side comparison

Axis	Presence analytics	Engagement analytics
Data source	Probe requests and signal strength from nearby devices	Captive portal sign-in and connected sessions
Who is counted	Every device with WiFi switched on in range	Only visitors who sign in
Identity	Device identifier, no name or contact details	Name, email and attributes the visitor provides
Typical metrics	Visits, conversions, busy hours, venue ranking	Repeat visits, opt-in rate, segment behavior, campaign response
Effect of MAC randomization	High: inflates unique counts and breaks repeat-visit matching	Low: identity sits in the sign-in record
Usual CCPA/CPRA lawful basis	Legitimate interests	Consent, for marketing
Visitor action needed	None	Sign in and choose whether to opt in
Hardware needed	Existing access points with a location feed	Existing access points plus a captive portal
Best suited to	Operations, estates, staffing and site comparison	Marketing, loyalty, CRM and personalization
Main blind spot	Cannot tell you who anyone is	Cannot see visitors who never connect

How MAC randomization changes presence data

Every WiFi device has a MAC address, a hardware identifier broadcast in its radio frames. Early presence systems treated each MAC address as one person. That assumption no longer holds.

Apple devices have used random MAC addresses in probe requests while unassociated since iOS 8. Since iOS 14 and Android 10, cell phones also use a random address per network by default when they connect. The practical effects on presence data are threefold:

• Unique-device counts rise. One cell phone can appear as several devices over a day or a week.
• Repeat-visit matching fails. A cell phone that returns next week may use a new address, so it looks like a first-time visitor.
• Hourly and daily volumes stay usable. Within a short window, the count of strong, sustained signals remains a sound indicator of how busy a venue is.

The rule that follows: use presence data for volume, timing and site comparison. Do not use it for loyalty or frequency analysis. Engagement data is far less exposed to randomization because the identity sits in what the visitor submitted, not in the radio address.

When is presence analytics the right call?

Choose presence analytics when you need coverage of everyone, not depth on a few. It is the right call when you need to:

• Compare sites fairly. Ranking venues by visits shows which locations underperform. Purple ranks your top 10 and bottom 10.
• Plan staffing. The day-and-hour heatmap shows when demand peaks, so you can schedule against it.
• Measure conversion inside a site. Divide transactions by visits to get a visit-to-sale rate for each store.
• Understand space. Zone Analytics shows average dwell per zone, average daily visits per zone and an hourly heatmap of people within zones.
• Avoid friction. Nobody has to sign in, so you see the large share of visitors who never connect.

In Retail, presence is usually the first layer you deploy. It answers estate and operations questions that marketing data cannot. For a deeper look at one vendor's location feed, read HPE Aruba Central presence analytics: setup, exports and limits.

Worked scenario: a 30-store fashion chain (illustrative figures)

Situation. A fashion retailer with 30 stores wants to know why two similar city-center stores trade differently. Both have similar floor space and similar register volumes.

What was done. The retailer compared a week of presence data with register transactions. Store A recorded 12,000 conversions and 2,400 transactions, a visit-to-sale rate of 20%. Store B recorded 9,000 conversions and 2,700 transactions, a rate of 30%. The day-and-hour heatmap showed Store A peaked on Saturday between 2:00 PM and 4:00 PM. During that window, only three staff members were on the sales floor.

Measurable outcome. The analysis isolated a 10-point conversion gap. It traced that gap to a staffing shortfall at a known peak, not to a lack of visitors. The retailer moved two staff members onto the floor for that window. It then set the visit-to-sale rate as the measure to review four weeks later.

Note what the retailer did not do. It did not use presence data to claim Store A had more loyal shoppers. MAC randomization makes repeat-visit counts from presence data unreliable.

When is engagement analytics the right call?

Choose engagement analytics when you need to know, and act on, who your visitors are. It is the right call when you need to:

• Grow an owned audience. Every conscious-choice opt-in adds a contact you can reach without paying a third-party platform.
• Measure loyalty. Repeat visits are tied to a known profile, so the new-versus-returning split is reliable.
• Prove campaign impact. You can see whether a contacted visitor came back, not just whether they opened an email.
• Segment your audience. Attributes from sign-in, such as age band, let you compare how different groups use your venue.
• Feed your CRM. Purple integrates with CRM and marketing platforms, so first-party data moves into the systems your team already uses.In Hotels, engagement analytics turns guest WiFi sign-ins into a direct relationship. That relationship matters most with guests who booked through a third party.

Worked scenario: a 200-room hotel with a restaurant and bar (illustrative figures)

Situation. A 200-room hotel books most rooms through online travel agents. As a result, it holds contact details for few of its guests. Its restaurant and bar sit half-empty on weeknights.

What was done. The hotel set up a captive portal that asked for an email address. Marketing consent was a separate, unchecked opt-in. Over a month, 1,500 guests signed in and 600 opted in, an opt-in rate of 40%. Zone Analytics showed a short average dwell in the bar zone compared with the lobby. Guests passed through the bar but did not stay. The hotel sent opted-in guests a midweek dinner offer.

Measurable outcome. The hotel built 600 contactable, consented guest profiles it did not hold before. It now measures two numbers each month: offer redemptions from that group and the change in bar-zone dwell. It also has a baseline opt-in rate to improve by changing what the portal offers in return.

Do you need both presence and engagement analytics?

For most estates, yes. The two measures fit together as a denominator and a numerator.

• Presence gives you the total. It shows how many people were in the venue.
• Engagement gives you the known share. It shows how many of them signed in and how many opted in.

Divide one by the other and you get a connection rate for each site. This is a practical measure of how well your guest WiFi offer is working. A store with high presence and low engagement has visitors but a weak reason to connect. Fix the portal or the incentive, not the location.

The two layers also correct each other's weaknesses.

Question	Presence alone	Engagement alone	Both together
How busy was Saturday?	Answered for all visitors	Answered only for those who signed in	Total volume plus the known share
Are visitors coming back?	Unreliable after MAC randomization	Answered for signed-in visitors	Repeat rate for known visitors, scaled against total volume
Which zone holds attention?	Dwell per zone, no segment detail	No zone view without location data	Zone dwell filtered by age or gender
Did the campaign work?	Shows a change in volume only	Shows response from contacted visitors	Response plus the effect on total visits

Worked scenario: a downtown rail station (illustrative figures)

Situation. A station operator wants to improve its concourse retail leasing. It also wants to communicate disruption information to passengers.

What was done. Presence data showed concourse visit volumes by hour. The data also confirmed the morning peak between 7:00 AM and 9:00 AM carried around twice the volume of the evening peak. Engagement data from passengers who signed in to station WiFi gave the operator an opted-in group for service updates.

Measurable outcome. The operator could show prospective tenants hourly visit counts, using numbers from its own network rather than estimates. It also gained a consented contact list it could measure by message open and reconnection rates. Purple works across rail estates on Trains as well as in stations.

Is presence analytics CCPA/CPRA compliant?

Presence analytics can be CCPA/CPRA compliant, but compliance depends on how you deploy it, not on the technology. The same applies to engagement analytics. The two simply rest on different lawful bases.

What CCPA/CPRA says about device identifiers

Under the CCPA/CPRA, personal information includes information that identifies a person through an "unique identifier." This includes identifiers provided by devices, applications, tools, and protocols. A MAC address can single out a device, and so its owner, over time. Treat it as personal data unless you can show otherwise.

The FTC and state attorneys general have published guidance on location analytics using WiFi signals. They treat MAC addresses as personal data where they can single out an individual. They also recommend clear signage, early anonymization, and short retention.

Hashing a MAC address does not on its own take the data outside the CCPA/CPRA. Pseudonymized data, which can be re-linked to a person with additional information, is still personal data. Only truly anonymous data falls outside the regulation.

The lawful basis for presence analytics

Presence data is collected without visitors taking any action, so consent is rarely practical. Most operators rely on legitimate business purposes. That basis requires three things:

• A business necessity assessment. Document the purpose, why presence data is needed, and how you balance it against visitors' expectations.
• Transparency. Tell visitors what you collect, why, and for how long. Signage at entrances and a linked privacy notice are the usual route.
• A way to opt out. Explain how visitors can opt out, for example by switching WiFi off.

A shopping center, stadium, or station running presence analytics is likely to require a thorough assessment. Complete the privacy impact assessment before you switch presence on.

The lawful basis for engagement analytics

Engagement data comes from visitors who act, so consent is available and expected for marketing. Under Article 7, consent must be freely given, specific, informed and unambiguous. It must also be as easy to withdraw as to give. Pre-checked boxes do not count.

In the US, the TCPA and CAN-SPAM, Regulation 22, add a separate rule for email marketing. You need specific consent for it, so keep the marketing opt-in apart from the terms of WiFi access. Purple's conscious-choice opt-ins are built on this separation. Purple is ISO 27001 certified and GDPR compliant, and also holds CCPA, Cyber Essentials and B Corp certifications.

Controls to put in place for both

• Minimize. Collect only the fields you will use.
• Set retention periods. Delete or aggregate raw presence records on a fixed schedule.
• Separate purposes. Do not reuse operational presence data for marketing without a fresh lawful basis.
• Restrict access. Limit raw identifiers to the people who need them.
• Publish your notice. Place it at entrances and on the captive portal.

The same principles apply in Healthcare. Patients and visitors there often have heightened privacy expectations, so make your signage and privacy notice especially clear.

What does each one cost to run?

Neither layer usually needs new access points. Purple runs as a cloud overlay on the hardware you already own. The real costs sit in configuration, governance and the staff time to act on the data.

Cost line	Presence analytics	Engagement analytics
Access point hardware	Existing access points from the supported vendor list	Existing access points from the supported vendor list
Extra configuration	Enable the controller's location feed and set venue boundaries	Design the captive portal, sign-in fields and opt-in wording
Floor plan work	Draw zones per site if you want Zone Analytics	Not required for sign-in data alone
Legal and governance	Legitimate interests assessment, DPIA, signage and retention schedule	Consent wording, privacy notice and the TCPA and CAN-SPAM review for email
Ongoing effort	Review busy hours and venue ranking against staffing and sales	Run campaigns, manage the CRM feed and handle opt-outs
Data quality upkeep	Recalibrate expectations as MAC randomization evolves	Keep sign-in friction low to protect the connection rate
Main hidden cost	Overreading unique and repeat counts	A weak incentive to connect, which shrinks the known audience

The biggest avoidable cost is a bad decision made on the wrong data. Two common examples are a loyalty figure taken from presence data and a busy-hours chart taken from signed-in visitors only.

How to decide for your estate

Start from the question the business needs answered this quarter. Then pick the layer that answers it.

Your situation	Recommendation	Why
Multi-site retailer comparing store performance	Presence first, engagement second	Venue ranking and conversion need total volume. CRM growth comes next.
Hotel group building direct guest relationships	Engagement first, presence second	Opted-in guest profiles reduce dependence on third-party bookings.
Shopping center or station selling space to tenants	Presence first	Tenants want hourly visit counts across everyone.
Stadium or event venue with ticketed attendees	Engagement first	Attendees already expect to sign in, and opt-ins feed future sales.
Venue with low WiFi connection rates	Presence first, then improve the portal	You need the total visitor count to measure the connection rate.
Organization in a sensitive sector, such as healthcare	Presence with a DPIA, engagement only with clear consent	Privacy expectations are higher and need documented governance.

A practical sequence for most estates:

1. Write down the three questions your operations and marketing leads most need answered.
2. Map each question to presence or engagement using the tables above.
3. Complete the DPIA and legitimate interests assessment before enabling presence at any site.
4. Deploy presence across the estate to establish baseline visit volumes and busy hours.
5. Launch the captive portal with separate, unchecked marketing opt-ins.
6. Report the connection rate per site and treat low performers as a portal problem to fix.
7. Review quarterly and retire any metric that MAC randomization has made unreliable.

Frequently asked questions

What's the difference between presence and engagement analytics?

Presence analytics counts devices from WiFi signals without the visitor connecting, so it measures total visits, dwell time and busy hours for everyone nearby. Engagement analytics comes from visitors who sign in through a captive portal and opt in, so it adds identity, repeat visits and marketing reach. Presence gives you breadth across all visitors. Engagement gives you depth on the known share. Most operators use presence for operations and engagement for marketing.

Which is better for retail, presence or engagement analytics?

For retail, start with presence analytics, then add engagement. Presence gives you visit counts, visit-to-sale conversion and venue ranking across every shopper, connected or not. Those are the measures that drive staffing and estate decisions. Engagement then turns the shoppers who sign in into opted-in contacts and reliable repeat-visit data. Using either one alone leaves a gap, either in identity or in coverage.

Do I need both presence and engagement analytics?

Yes, for most multi-site estates. Presence measures the total number of visitors, while engagement measures the share you can identify and contact. Together they give you a connection rate per site, which shows where your guest WiFi offer is underperforming. Presence also cannot measure loyalty reliably after MAC randomization, and engagement cannot see visitors who never connect. Running both closes each blind spot.

Is presence analytics CCPA/CPRA compliant?

Presence analytics can be CCPA/CPRA compliant when you deploy it with the right controls. Treat MAC addresses as personal data under privacy laws. Most operators rely on documented business necessity and clear disclosures. You also need clear signage, a privacy notice, and short retention. Large public venues usually need a privacy impact assessment before switching presence on.

Do I need new hardware to run presence analytics?

No, in most cases you use the access points you already own. Purple is hardware-agnostic and runs as a cloud overlay on Cisco Meraki, HPE Aruba, Ruckus, Juniper Mist, Ubiquiti UniFi, Cambium, Extreme and Fortinet. Presence needs the controller's location feed enabled and venue boundaries configured. Zone Analytics also needs zones drawn on a floorplan for each site you want to analyze.

How does MAC randomization affect my visit data?

MAC randomization mainly damages unique-device and repeat-visit counts from presence data. Since iOS 14 and Android 10, cell phones use a random MAC address per network by default, and many randomize probe requests too. One cell phone can appear as several devices, and returning visitors can look new. Hourly and daily visit volumes remain usable. Use engagement data, which is anchored to a sign-in, for loyalty and frequency.

Can I use WiFi sign-in data for email marketing?

Yes, but only with a separate, specific marketing consent. Signing in for WiFi access is not consent to receive marketing. Under the TCPA and CAN-SPAM, consent must be freely given and unambiguous. Use an unchecked opt-in that sits apart from the access terms. Purple's conscious-choice opt-ins follow this model, so your list holds only visitors who chose to join.