Analyzing Sales Data for E-commerce

Ask the Community
,
1

Hello everyone, I have found this awesome data set on Kaggle: https://www.kaggle.com/datasets/benroshan/ecommerce-data/data

I want to develop a dashboard analyzing this sales data. I am a newbie and am unsure how to go about it.

Please help me with what sort of analysis can I do here. Secondly, how to go about building this on the AIP? How to build the workflow diagram?

4

Hi @p23rudraneel! I would start by uploading the dataset into Foundry and then use the Contour application to analyze your data. https://www.palantir.com/docs/foundry/contour/overview/

Contour also allows you to build dashboards which support things like chart-to-chart filtering, inline parameter references, a fullscreen presentation view, etc. https://www.palantir.com/docs/foundry/analytics/dashboards/#contour-dashboards

1 Like
5

Hey! Welcome to the community! I’ll flag off the top that in the future, you’ll get the best help here (and have the best luck building on Foundry) with more concrete project goals. That said, I’m certainly sympathetic with a sense of “where do I start” with a platform as broad and deep as AIP.

So here’s a quick rundown of some different ways to proceed:

Dive right in | I’d go this way if you just want to start “playing around” with this data, but I don’t think it’s the best way to really grasp what’s unique or powerful or important about building something with AIP.

  1. Download the Kaggle dataset as a CSV
  2. Create a new Project in Foundry to hold your work.
  3. Drag and drop the CSV into your project - this is a manual data upload
  4. Open the dataset in Contour and start exploring and visualizing

Build something interesting | AIP is all about putting your data into motion and then automating / enriching that motion with AI. To get there, though, you as the builder have to have more of an opinion about what’s valuable than to “develop a dashboard analyzing [this] data”.

You might instead imagine: How could I use this data to build a tool for a sales agent to use during sales calls that gives them:

  • Al the metadata of the customer and a history of their purchases
  • An automatic analysis of the value of this particular customer and where they fall into some different segmentation / cohorts
  • A suggested set of offers to make during the call to improve the sale
  • An AI agent assistant for the agent to interact with that has context on the customer and their history

These kind of user stories and high-value use cases are, in my opinion, the key to actually unlocking value on Foundry. Once you can articulate something like this, you can use the AIP Architect in Solution Designer to help all the components you might configure within AIP to start building this vision (and get a walkthrough on implementing them).

image
image1484Ă—1364 262 KB

image
image1920Ă—2056 202 KB

Follow a learning path

You may also benefit from grounding yourself in a few more “on rails” experiences in AIP before diving in with your own data: the Data Analyst would again probably help you get moving most quickly towards your stated goal, but I’d encourage you to check out the other tracks as well

Again, welcome to AIP and looking forward to seeing what you build!

3 Likes