Laptop Price Intelligence

Why this project

Comparing laptops across brands means untangling specs that are never formatted consistently — "8GB" vs "8 GB", "256GB SSD" vs "256GB Flash Storage" — before you can compare anything at all. This project is a good beginner-to-intermediate exercise precisely because most of the real work is in Python data cleaning, not the dashboard. You can build it with the public Laptop Price dataset on Kaggle, which mirrors real listing data closely enough that the cleaning steps below transfer directly to scraped data.

The raw data

A typical row looks like this before cleaning:

Memory mixes size and storage type in one string, and prices come in whatever currency the source used — both need parsing before anything downstream is trustworthy.

Step 1 — Clean it with Python

import pandas as pd
import re
 
df = pd.read_csv("laptop_data.csv")
 
# "8GB" -> 8
df["ram_gb"] = df["Ram"].str.replace("GB", "", regex=False).astype(int)
 
def parse_storage(mem: str) -> int | None:
    match = re.search(r"(\d+)\s*(GB|TB)", mem)
    if not match:
        return None
    size, unit = match.groups()
    size = int(size)
    return size * 1000 if unit == "TB" else size
 
df["storage_gb"] = df["Memory"].apply(parse_storage)
df["storage_type"] = df["Memory"].str.extract(r"(SSD|HDD|Flash Storage|Hybrid)")
 
# Adjust the multiplier to your target currency and current exchange rate
df["price_bdt"] = df["Price_euros"] * 127

Run df.isna().sum() right after this step — a handful of unparsed storage_gb values almost always means the regex is missing a format your dataset uses, and it's much cheaper to catch that here than to notice a gap in the Power BI chart later.

Step 2 — Model price bands in DAX

Rather than hardcoding thresholds into the data itself, build the band as a measure so it's easy to adjust as prices change over time:

Avg Price by Brand = AVERAGE(Laptops[price_bdt])
 
Price Band =
SWITCH(
    TRUE(),
    Laptops[price_bdt] < 40000,  "Budget",
    Laptops[price_bdt] < 80000,  "Mid-range",
    Laptops[price_bdt] < 150000, "Premium",
    "Flagship"
)
 
Laptops in Band =
CALCULATE(
    COUNTROWS(Laptops),
    Laptops[Price Band] = SELECTEDVALUE(Laptops[Price Band])
)

Step 3 — Build the report

Two pages carry the whole analysis:

1. Brand comparison — average price by brand as a bar chart, with RAM and storage as slicers so a viewer can compare like-for-like configurations.
2. Budget segments — a stacked bar of laptop count per price band per brand, so someone shopping with a fixed budget can immediately see which brands actually have options in their range.

How to build your own version

1. Download the Laptop Price dataset from Kaggle (or substitute your own scraped listings — the parsing logic is the same either way).
2. Run the cleaning script above and check for unparsed rows.
3. Load the cleaned CSV into Power BI.
4. Add the Price Band measure and the two report pages.
5. Re-run the currency conversion multiplier periodically if you're tracking a market where prices are quoted in a foreign currency.

Talking about it in an interview

The strongest thing to point at here isn't the dashboard — it's the regex parsing step. Being able to explain why you extract storage type and size separately (so you can filter "256GB SSD" laptops without accidentally including "256GB HDD" ones) shows you understand the data quality problem, not just the charting tool.