<img height="1" width="1" style="display:none;" alt="" src="https://dc.ads.linkedin.com/collect/?pid=398730&amp;fmt=gif">

Driven By Data

What’s their secret? Exploring Victoria’s Secret Dataset (Part 2)

Posted by Data Society on Apr 1, 2019 5:14:52 PM

Last week, we dove into the Victoria’s Secret dataset and discovered that they had a huge variety of not only bra types, but also colors, sizes, and materials. Whenever you start working with a new dataset, always remember to explore it first so you have a good idea of the distribution of the data and you may even have new questions. Now that we have a better understanding of Victoria’s Secret’s inventory, we can ask more questions about their pricing strategies, such as:

  • What’s the price distribution for each product category?
  • Which cup size is cheaper? And which cup size is the most expensive?
  • If you want to save money, which color is cheaper? 

VS-2

What’s the price distribution for each product category?

The boxplot visualization is a great way for us to view the distribution of many categorical attributes. The longer the boxplot, the more variation there is. The small dots on either end represent outliers. It turns out Multi-Way Bra has the highest average price of all the categories, while the Bralette has the lowest average price. 

Product price distribution visualization

However, the Bralette has a few outliers that are much higher than the average price. Why are they much more expensive? The product names are ‘Sexy Crochet Lace Triangle Bra' and 'Dream Angels Crochet Lace Strapless Bralette'. Both products are made with crochet fabric, so this could be what is impacting the higher price point. A possible reason for the increased price is because the crochet is done by hand or may be perceived to be more luxurious. Interestingly, the most popular bras (Push-Up) range on the more expensive side. This is where domain expertise comes into play, as the manufacturer and distributor would have a better idea of these nuances.

Which cup size is the most expensive?

When we look at the pricing compared to sizing, we see that the higher prices appear at the smaller and larger sizes.

size vs. price

If we just look at cup size and ignore band size, data analysis shows uncommon bra sizes are the most expensive. AA cups’ average price is $55.1 and DDD cups’ average price is $43.6. For cup sizes B, C, and D, the price is almost the same. It is interesting that DDD cups are less expensive, on average, than AA because presumably they’d need to use more fabric for the larger sizes. But as we saw with the boxplots, one other factor of price is the material of the bra, so that could impact the price. Alternatively, if it’s less popular, they may need to price it higher because they might not get as much of a discount in manufacturing in bulk. This is where some more information about the supply chain and clothing manufacturing process would be valuable.

size vs. price

If you want to save money, which color is cheaper?

Can you save money simply by choosing a different bra color? We simplified most (73.6%) of the 542 very specific colors into 12 condensed general categories and then checked the median price. As it turns out, orange is the cheapest color, while brown is the most expensive! That was unexpected because pink is the brand’s signature color, although we can see that it is only a dollar less expensive on average than brown. When we investigated further, we saw that a large portion of the ‘brown’ color was called ‘bronze brown lace’ or ‘bronze brown crochet lace’. So both ‘crochet’ and ‘lace’ could be responsible for the price increase - this just goes to show that it’s important to ask ‘why’ certain results occur to make sure you are reaching the correct conclusion.

Color

 

We’ve only scratched the surface on the questions you can answer with a dataset like this. Here are some more questions if you want to keep exploring:

1. Product description
Topic modeling on the product description: What are the common style attributes? What’s the difference between different bra categories?

2. Popularity
What’s the rating for each product category? How many reviews do they have? Is there any relationship between rating and review numbers?

3. Brand price differences
You can cross-reference this data with other underwear brands (Btemptd, Calvin Klein, Hanky Panky, American Eagle, etc.) to see how does the price differ depending on the brand

At the End

Don’t forget the best practices that we’ve demonstrated over the course of this investigation, including asking the right questions and looking at an overview of the data instead of diving in. Make sure that you take the time to validate your findings or research it further to ensure that you’re coming to the right conclusions. 


Curious how we can help you and your team ask the right questions of your data? Contact us today to schedule a time to speak with us.

Topics: data visualization, data training, data communication, data strategy, data-driven decision-making

Who are we?

Data Society provides tailored data science training programs to help organizations build their data science capacity and instill a data-driven culture. We work across all levels of a company to bridge the data communication gaps, and we're preparing people for the data revolution!

Subscribe Here!

Recent Posts