Paridhan Couture, where I work as Chief Technology Officer, told me they would cover Black Friday. They told me they wanted to provide sales on category. Interesting fact is there is nothing in Shopify. There are collections in Shopify but no categories. Though these two terms seem the same. There are two options for providing sales. Either an appropriate app needs to be used or this criteria needs to be done through CSV. One of the most popular apps is Pumper Bundles Volume Discount which costs $9.99 per month. I mean it's only for Black Friday but need to pay for the full month? No Way!
So I just came up with a CSV manipulator coded in Python. I used Pandas to manipulate CSV data. Pandas is great at the time of manipulating big chunks of data, right? So the conditions were, If a product belongs to this (named the category which would have a 30%), then it would have a 30% discount or if a product belongs to this (named the category which would have a 40%), then it would have a 40% and the rests would go for 50%. So I categorized these into 2 lists.
tags30 = [""] # name of collections which would have 30 tags40 = [""] # name of collections which would have 40
except for these two lists' collections, every collection would go through 50% discounts. But now I faced a problem with this. Shopify doesn't provide collection names on its exported CSV of products. Thankfully we were using tags to categorize (putting products in collections) through tags. So now, it's tags that I considered instead of collection names because I didn't have them. I needed to know tags per product so basically I got the full length of the CSV and went through a loop for this using Pandas.
df = pd.read_csv('products.csv') count = len(df.index) for i in range(count): if type(df.values[i][6]) == str: tag = (df.values[i][6]).split(",") tag = [s.strip() for s in tag] containOld = tag tag = containOld
Here, df.values[i][6] is the column of tags and the tag variable has tags of products. containOld = tag in if condition and outside of if condition, tag = containOld! a little bit strange, right? I did this because in the CSV, only the base product contains the information of tags and other things, variant products don't. So before getting new tags for new products, base product information is available in the containOld variable.
price = df.values[i][23] basePrice = df.values[i][24]
After getting a product's base price and sale price field, I needed to insert discounts according to the conditions. Here, price is Variant Price column and basePrice is Variant Compare At Price of CSV.
if pd.isnull(df.values[i][24]) or df.values[i][24] == 0: mainPrice = price else: mainPrice = basePrice
I applied this condition because some products could have discounts already and I couldn't provide a discount on discount prices, right? So, for discounted products mainPrice will be basePrice instead of price.
if pd.isnull(price): pass else: if any(e in tag for e in tags30): df.loc[i, 'Variant Price'] = mainPrice / 100 * 70 elif any(e in tag for e in tags40): df.loc[i, 'Variant Price'] = mainPrice / 100 * 60 else: df.loc[i, 'Variant Price'] = mainPrice / 100 * 50 df.loc[i, 'Variant Compare At Price'] = mainPrice df.to_csv("result.csv", index=False)
In the end, I applied the conditions my boss told me for categories (collections) that were in tags30 and tags40 lists. math.ceil() could be used to get rid of cents issues, but that was pretty ok with the sale price and I just saved $9.99 of my company. I will talk more about this and how I will reverse these sale prices in a new blog.