BoostBI: Footfall Analytics Software & Shopper Analytics Platform

What visitor analytics software does

Visitor analytics software turns anonymous movement through a physical space into numbers a team can act on. In V-Count’s system the sensors do the measuring and BoostBI does the thinking: Nano AI, Nano Outdoor and Nano Prime produce the counts at the door and inside the space, and BoostBI stores, aggregates and reports them as a single shopper analytics platform.

If you are comparing visitor analytics solutions, this is the checklist most buyers work through, and how each item behaves in BoostBI.

• Footfall. Entries and exits per entrance, per hour and per day, with separate lines for each door so a mall entrance and a car-park entrance are never blended. This is the base layer of any people counting deployment and the reason footfall analytics software exists.
• Conversion. Visits measured against transactions imported from your POS, so a quiet day with strong selling and a busy day with weak selling stop looking the same in the report. See retail store analytics for how the sales side is mapped.
• Dwell time. How long visitors stay overall, and how long they linger in a specific area, which separates browsing from genuine interest.
• Zone analytics. Counts and heat patterns for defined areas — entrance, promotional table, fitting rooms, a specific aisle — so layout decisions are made on observed movement. This is covered by heatmap and zone analytics.
• Queue measurement. Waiting-line length and waiting time at tills and service desks, which is the pairing between counting hardware and queue management.
• Occupancy. A live headcount of how many people are inside right now, derived from entries minus exits, for capacity rules, safety limits and space planning.
• Demographics. Anonymous estimated age band and gender distribution of visitors, produced by gender and age recognition, used to check whether the people in the store match the people the range was bought for.
• Alerts. Threshold and schedule-based notifications — occupancy above a limit, a queue past an agreed wait, a sensor that stopped reporting — delivered without anyone watching a dashboard.
• Benchmarks. Store against store, region against region, this period against the same period last year, so a manager can see whether a result is a local problem or a chain-wide pattern.
• Multi-site roll-ups. One estate-level view that totals every location, with the ability to drill from group to region to a single door.
• API and exports. Scheduled exports and API access so the same visitor counts can sit inside the BI tool, data warehouse or workforce system your business already runs on.

The practical value of visitor analytics software is not the individual number but the comparison. Footfall on its own says how busy you were; footfall next to transactions, dwell, zone traffic and staffing is what produces usable store insights — which hour is under-staffed, which display is ignored, which entrance actually brings buyers rather than passers-by.

AI visitor analytics: what is actually automated

“AI visitor analytics” is used loosely, so it is worth separating the two places automation genuinely sits. The first is measurement: V-Count’s Nano AI sensor runs computer-vision models on the device itself to decide what is a person and what is a shopping trolley, a reflection or a shadow.

That is the AI doing the counting, and it happens before any data reaches the platform. The second is interpretation — what a BI layer can do with a long, clean history of counts.

The capabilities below are the ones worth expecting from a modern shopper analytics platform; treat them as what the category can do rather than as a promise of a specific number.

Anomaly alerts

Once a location has enough history, a platform can learn what a normal Tuesday afternoon looks like for that specific door and flag the days that do not fit — a sudden drop that usually means a sensor fault or a blocked entrance, or a spike that means a local event nobody told head office about. The value is that the exception finds you, instead of waiting for someone to open a report.

Forecasts and expected ranges

Historic footfall is seasonal and highly repetitive, which makes it well suited to forecasting. A visitor analytics platform can project an expected range for the coming days or weeks from the same store’s past pattern, then show actuals against that range. Forecasts can be wrong — weather, promotions and roadworks are not in the data — so they are best used as a planning baseline, not a commitment.

Staffing suggestions

If the system holds hourly traffic and hourly transactions, it can point at the hours where conversion falls while traffic stays high, which is the usual signature of too few people on the floor. BoostBI’s AI Sales Coach is the version of this on this platform: a weekly summary that turns the week’s numbers into a short set of suggested actions for a store manager, rather than another chart to interpret.

What is not automated

No visitor analytics solution decides your range, your pricing or your rota for you. It narrows where to look and removes the argument about what happened; the judgement about why, and what to change, stays with the retail team. Any vendor describing more than that is describing a decision-support tool in ambitious language.

Visitor analytics solutions by venue type

The sensors are largely the same across venues. What changes is which zones you bother to count, which KPI the site is actually judged on, and who reads the report.

Retail stores

A store is judged on conversion, so the pairing that matters is traffic against transactions, then dwell and zone data to explain the gap. Counting the entrance is the minimum; counting the fitting rooms, the promotional area and the till line is what makes the number diagnostic.

Chains typically add a multi-site roll-up so underperforming branches surface without anyone opening twenty dashboards. The detail for this venue type sits on the retail store analytics page.

Shopping malls

Malls measure for two audiences at once: their own operations team and their tenants. That means entrance counts per door, common-area and corridor traffic, capture rate between the mall and individual units, and occupancy for safety and event planning. Leasing conversations change when a landlord can show measured passing traffic for a unit rather than an estimate. See people counting for shopping malls.

Events and exhibitions

An event is a short, dense measurement window with no second chance, so the setup has to be right on day one. Organisers usually want total attendance, per-hall and per-stand traffic, dwell at individual stands and peak occupancy against the venue limit — numbers that become exhibitor evidence and next year’s floorplan. See events, exhibitions and trade show analytics.

Libraries and museums

Here there is no transaction to convert, so the KPI moves to visits, dwell and space utilisation: which rooms and study areas fill up, at what hours, and whether opening hours or staffing match actual demand. Counts also support funding and reporting obligations, which is why accuracy and a defensible method matter as much as the dashboard. See visitor counting for libraries and museums.

Transport hubs

Stations, terminals and interchanges measure flow rather than shopping: passenger volumes by entrance and platform, crowding and occupancy in concourses, and queue length at security, ticketing and boarding. Conditions are harsh and coverage areas are wide, which is where outdoor-rated and 3D stereo sensing options matter more than reporting features. See transportation and transit counting.

Case-study style: what a store insights review looks like

Imagine a fashion retailer with a group of stores, sensors on every entrance, zone counting in three areas per store and POS data connected to BoostBI. A monthly store insights review would run roughly like this.

1. Start at the estate, not the store. Open the multi-site roll-up and sort stores by conversion rather than by sales. Sales rank tells you which store is biggest; conversion rank tells you which store is wasting the traffic it already has. The bottom of that list is your review list.
2. Split traffic from performance. For each store on the list, ask whether footfall fell or conversion fell. A footfall problem is usually external — the location, the centre, the season, a road closure — and is answered with marketing or landlord conversations. A conversion problem is internal and is answered inside the store.
3. Go to the hour. Pull the hourly view for a weak store and compare traffic curves with conversion curves. The pattern people most often find is a conversion dip that lines up with the busiest hours, which points at coverage rather than demand.
4. Check the rota against the curve. Lay the staffing schedule over the hourly traffic. Shifts are frequently built around opening hours and break logistics rather than around when customers actually arrive, and that mismatch is visible the moment the two are on the same axis.
5. Look inside the store. Use zone and heatmap data to see whether visitors reach the areas the store is being judged on. A promotional zone with high traffic and low dwell suggests the display is passed but not engaging; high dwell with weak conversion suggests interest that is not being closed.
6. Check the queue. If waiting time climbs in the same hours conversion drops, the problem may be at the till rather than on the floor — a fixable operational issue that footfall alone would never reveal.
7. Sanity-check the audience. Anonymous demographic distribution shows whether the people entering resemble the customer the range and marketing were built for. A persistent mismatch is a merchandising question, not a store-team question.
8. Change one thing and re-measure. Agree a single action per store — a shift moved, a display relocated, a second till opened at a set hour — then compare the following period against the same period before the change and against stores where nothing changed. The control group is what keeps the review honest.

Two things make this kind of review work in practice. First, comparison discipline: a number is only evidence when it sits next to a baseline, a benchmark or an unchanged store.

Second, one change at a time — estates that adjust staffing, layout and promotions in the same week can never attribute the result. If you would like to see this run against your own data, book a BoostBI demo and bring one underperforming store with you.

Visitor analytics FAQ