---
title: "GoTrack | Computer vision retail analytics"
description: "Computer vision on standard IP cameras or USB webcams: see which products shoppers pick up, hold and put back. No special AI camera, no face recognition."
url: https://gotrackgo.com/en/
language: en
updated: 2026-10-14
publisher: GOTONOM Yazılım Teknolojileri A.Ş.
---

# The store, aware.

See how shoppers interact with every product. Understand what happens before the sale. **Act in real time.**

## Your POS tells you what sold. What happened before that?

The sale is only the final event. GoTrack reveals the journey before checkout, product by product.

- Your POS: 109 units sold (One number, at the till.)
- GoTrack: 892 product interactions
  - 342 picked up
  - 147 compared
  - 28 s average hold
  - 217 put back

_Example data_

## The physical product funnel

Ecommerce has measured product behaviour for decades. GoTrack brings that visibility into the physical store.

1. **On the shelf** (48, products in view): Every product in view, including the ones nobody touches.
2. **Picked up** (342, pickups): Which product and colour, from which shelf zone.
3. **Held** (28 s, on average): How long it stays in a shopper's hands.
4. **Compared** (147, 43% of pickups): Which products are weighed against each other.
5. **Put back** (217, 63% of pickups; 125 not returned, 37%): Tried, then returned to the shelf: the put-back rate.
6. **Purchased*** (109, from your sales data · 32%): Joined with your POS or BI data: how many of the products that did not return were sold.

* Purchase is available when combined with sales data.

_Example data_

## A camera sees the product. GoTrack turns it into data.

Cameras placed near the products recognise which product and colour is in a shopper's hand, and record what happens to it. Your existing store becomes a product-interaction layer.

Camera → Product detected → Product & colour matched → Interaction recorded → GoTrack data

- No RFID tags
- No shelf sensors
- No face recognition
- Processed in the store
- No live video to the cloud

## Every product has a story.

See what gets picked up, held, compared and put back, by product, colour, store and hour. Product-level behaviour, not people counting.

- **White essential tee** (TS-WHT-001): 482 pickups, 31 s avg. hold, 22% put-back rate
- **Black oversized tee** (TS-BLK-002): 731 pickups, 43 s avg. hold, 11% put-back rate
- **Green linen tee** (TS-GRN-003): 216 pickups, 17 s avg. hold, 39% put-back rate
- **Gray relaxed tee** (TS-GRY-004): 394 pickups, 26 s avg. hold, 18% put-back rate
- **Which age and gender pick up each product** (Building now): Aggregate reports per product, never individual profiles.

_Example data_

[See the analytics](https://gotrackgo.com/en/retail-analytics)

## High interest. Low conversion.

The green linen shirt is picked up 623 times and considered for 41 seconds on average; your sales data shows only 18 purchases. GoTrack adds the missing behavioural layer to the sales data you already own.

**Green linen shirt**, Combined with your POS or BI data: GoTrack: 623 pickups, 41 s consideration + Your sales data: 18 purchases = High interest · Low conversion

**What to check next**

- Check the price
- Check size availability
- Compare its placement in the store
- Review the campaign
- Compare it across stores
- See which products it is compared with

_Example data_

[Explore insights](https://gotrackgo.com/en/retail-analytics)

## See what's working. Everywhere.

Compare the same product across stores, zones and shelves. GoTrack scales from one interaction to your whole network.

**White essential tee · TS-WHT-001**

- Flagship store (İstanbul): 687 Interactions, 42 s Avg. hold, 18% Put-back rate
- Mall store (Ankara): 411 Interactions, 29 s Avg. hold, 31% Put-back rate
- City store (İzmir): 892 Interactions, 47 s Avg. hold, 12% Put-back rate

Network → Store → Zone → Shelf → Product → Colour

_Example data_

## Insight doesn't have to wait.

GoTrack can stop at analytics.

Or it can act the moment a product is picked up. The screen beside it is one option, not a requirement.

Picked up → Product & colour matched → Event created → Screen shows it

**What GoTrack can do next**

- Show the product's content on the nearest screen
- Show two products side by side
- Play a campaign
- Send the event to your systems (API, webhook)

[See smart screens](https://gotrackgo.com/en/smart-signage)

## From product decision to screen acknowledgement

- **0.35 s**: Decision to screen acknowledgement, median with one product
- **0.51 s**: Decision to screen acknowledgement, median with two products
- **0.71 s**: The slowest of 24 acknowledged fires

Measured on 30 September 2026 in GoTrack's test setup: 25 fires across 8 products; 24 reported their render back (one two-product fire did not). Timed on the store device's clock from the moment GoTrack decided which product was in hand until the screen reported the product content painted, including that report's trip back. The time from lifting a product to that decision is not included here; we will publish it once it is measured.

[How we measured it](https://gotrackgo.com/en/screen-reaction-time-measured)

## Don't just run experiences. Measure their effect.

GoTrack can change what a screen shows, then measure whether shoppers picked up and held the product more than in a period without it.

- Before content: 182 interactions / hour
- After content: 246 interactions / hour
- +35% product engagement

_Example data_

[How to run an in-store A/B test](https://gotrackgo.com/en/in-store-ab-testing)

## If it can be seen, it can become a signal.

Every category comes with its own question.

- **Apparel** (In stores): Which colours get picked up, and which go straight back?
- **Footwear** (In stores): Which models are compared most often?
- **Fragrance** (Pilot first): Which fragrance is held the longest?
- **Cosmetics** (Pilot first): Which shades are picked up and put back?
- **Watches** (Pilot first): Which pieces are held longest before they go back?
- **Bags** (Pilot first): Which bags draw hands in which store?
- **Electronics** (Pilot first): Which products are weighed against each other?
- **Grocery & FMCG** (Pilot first): Which products leave the shelf most often?

## One camera system. Many ways to use it.

Screens are optional: the analytics works with cameras alone.

### Shelf analytics, no screens

Pickups, holds and put-backs from cameras alone.

[Learn more](https://gotrackgo.com/en/shelf-analytics)

### Lift and learn

Pick up a product; its story appears on the screen beside it.

[Learn more](https://gotrackgo.com/en/lift-and-learn)

### Showrooms & trade fairs

Hold a product up to the camera and the screen presents it.

[Learn more](https://gotrackgo.com/en/showroom-product-presentation)

### Screen content testing

Which content goes with longer holds, per asset and store.

[Learn more](https://gotrackgo.com/en/in-store-screen-content-testing)

### Visitor counting

On request: how many came in, anonymously.

[Learn more](https://gotrackgo.com/en/visitor-counting)

### Store and chain benchmarking

The same reports, compared across every branch.

[Learn more](https://gotrackgo.com/en/retail-analytics)

[All use cases](https://gotrackgo.com/en/use-cases)

## Measure product interactions. Build no shopper identity profiles.

To tell that a product is being held, GoTrack follows the positions of people and hands within a single camera, using short-lived tracking numbers that belong to that camera only. No face recognition, and no shopper is recognised from one visit to the next.

- **No face recognition**: No face templates and no biometric matching.
- **No identity profiling**: No shopper is recognised from one visit to the next.
- **Processed in the store**: Live camera images are analysed on a device in the store, in memory.
- **No live video to the cloud**: The cloud receives product events and device health. A product photo from a camera goes there, by default, only when an authorised user starts a catalogue or setup step.
- **Works offline**: Recognition and screens keep running when the internet drops.
- **No images kept**: By default no camera image is written to disk.

[Read the privacy approach](https://gotrackgo.com/en/privacy-by-design)

## Designed for real stores, not for a lab.

### Standard cameras

No AI camera needed: a standard IP camera or USB webcam, at a distance where it sees the products clearly. We supply them, or you buy them.

### Product- and colour-level

Not “someone touched the rack” — the exact product and colour variant.

### Add products yourself

A guided step in the panel — no data-science project, no waiting on a vendor.

### Offline-first

Recognition and screens keep running when the internet drops; nothing is lost.

### Managed remotely

Every store device reports its health and receives updates without a site visit.

### Local team, three languages

Built and supported in İstanbul. Panel and site in Turkish, English and Spanish.

[See how it fits](https://gotrackgo.com/en/integrations)

## What is GoTrack?

**GoTrack is camera-based in-store product analytics software developed by GOTONOM in İstanbul.** It measures product interaction on the shop floor: which product and colour was picked up, from which shelf zone, how long it was held and whether it went back on the shelf.

Its focus is products: what shoppers do with them. It uses no face recognition, images are processed on a device inside the store, and screens are optional. The full picture: [What is GoTrack?](https://gotrackgo.com/en/what-is-gotrack)

## Common questions

### What is GoTrack?

GoTrack is computer vision retail analytics software developed by GOTONOM in İstanbul that runs on standard cameras. Using standard IP cameras or USB webcams placed where they see the products clearly, it measures product interaction on the shop floor: which product and colour is picked up, how long it is held and whether it is put back. Optional screen integration can show that product's content on the nearest screen; the analytics works without screens.

### How quickly does the screen change when a product is picked up?

In our test measurement on 30 September 2026, the screen acknowledged painting the product within 0.71 s of GoTrack deciding which product was in hand, in all 24 acknowledged fires: a median of 0.35 s with one product and 0.51 s with two. One of the 25 fires sent no acknowledgement. The time from the lift itself to that decision is not included and not yet measured separately. Method: [the reaction time note](https://gotrackgo.com/en/screen-reaction-time-measured).

### How is GoTrack different from people counters and heatmaps?

People counters count who comes in, heatmaps show where people walk, and shelf recognition checks what is on the shelf. GoTrack measures what shoppers do with the products themselves: which product and colour is picked up, how long it is held, and which are put back or compared. Visitor counting can be added on request.

### Does GoTrack use face recognition?

No. There is no face recognition, no face template, no biometric matching, and no shopper is recognised across visits.

### Do we need screens to use GoTrack?

No. GoTrack's product analytics works with cameras alone. Screens are optional: add them when you want the shelf to react to the product in hand.

### Does GoTrack count people?

On request, yes. GoTrack's standard reports are about products: what shoppers pick up, hold and put back. For stores that also need visitor numbers, GoTrack can count visitors with the same cameras, anonymously and without face recognition; it is set up per project.

## GoTrack is the intelligence layer between the shelf and the sale.

Measure interactions. Understand product performance. Activate better retail experiences.

[Explore GoTrack analytics](https://gotrackgo.com/en/retail-analytics)

---

GoTrack · by GOTONOM · https://gotrackgo.com/en/ · info@gotonom.com