This is a solution I came up with for a problem in a fishing tackle store where I was working.

The Problem

The fishing tackle store is predominately online but does have a physical showroom that customers can visit. For a long time if customers wanted to know how much an item cost, they would either have to try and work out the product name and search the website themselves, or ask an employee how much it is. Having to do either was quite an inconvenience and would make the customers feel bad for constantly having to ask. The reason why no price tags had been previously added was the sheer number of products, how often new products are added to shelves, and that the price tags become out of date and needed to be changed each time prices or stock were updated.

A New Solution

I came up with an idea to try and address the concerns with traditional price tags. By utilising QR codes to the website's product page the customer in the showroom could scan the code on the flip tag above the item and go to the website where they could see the current price, without having to ask the staff. This solution means that every time a price update occurs in the online store it is immediately visible in the physical showroom. The added advantage is that it allowed not only easy price updates but also drives traffic to the website, which is great for SEO. Additionally, if someone doesn't close the tab in their phone's web browser, they can see the product they were looking at again at a later date. This potentially will generate more sales for the online store.

Automatically Generated QR Codes

Creating QR codes for every product in a store is surprisingly easy. For this I used python's QR code and pillow libraries. Inspired by QR code generating websites I wanted to be able to have a large company logo in the center of the image. Taking this into account I went with the following parameters:

qr = qrcode.QRCode(
    error_correction=qrcode.constants.ERROR_CORRECT_H,
    box_size=10,
    border=1,
)

Things that stand out here is the high error correction allowing me to use the biggest logo possible and the really small border size of 1. The quiet zone is very important for QR codes and typically you would leave this at around 4, but in my case the white paper I'm printing onto will provide the necessary extra blank space.

After generating the basic image, pillow can then be used to resize the company logo for layering over the top. I consulted this comparison table from pillow's documentation and went with LANCZOS resampling as I wasn't overly concerned about performance.

Getting Products Through BigCommerce's API

BigCommerce has a pretty well documented API which allowed me to easily loop through everything in stock to make a QR code. The API does utilise pagination for retrieving large quantities of products which is important to remember when it seems like your store suddenly has a fraction of the total products.

Filtering for only SKUs with at least one item in stock and utilising the product id and slug for file names, I was able to generate about 100 QR codes a second.

In the end this solution was much faster than I anticipated and has made a big difference in the showroom to aid customers.