Real Estate Analysis using PowerBi

Real Estate Analysis using PowerBi

About this project

Project Goal

My project was centered around comprehending the intricate trends and patterns within the real estate market in Connecticut. Using Power BI as my primary tool, I aimed to create an interactive dashboard that could not only visualize but also bring to life the nuances of property values, sales ratios, and more.


Business Needs

  1. Optimize Investment Strategies: By analyzing trends in assessed values and sales amounts, the company can identify towns or property types that consistently exhibit high sales ratios. This insight guides investment decisions, directing resources toward areas with strong growth potential.
  2. Tailored Marketing Campaigns: Understanding property type distributions and their performance can guide marketing efforts. The company can create targeted campaigns that resonate with specific buyer preferences, boosting engagement and conversion rates.
  3. Risk Mitigation: Analyzing trends over time and assessing how residential types influence sale amounts can provide insights into potential risks. This information aids in identifying properties or locations that might pose a higher risk due to volatile market conditions.
  4. Strategic Expansion: Insights derived from geospatial analysis can guide the company's expansion plans. It helps identify areas with untapped potential, ensuring that new developments align with market demand and preferences.
  5. Data-Driven Decision-Making: Equipped with an interactive Power BI dashboard, various departments can make informed decisions backed by data. From sales to marketing to finance, teams can align strategies with market realities.
  6. Resource Allocation: By focusing resources on property types or locations with strong sales ratios and assessed value trends, the company optimizes resource allocation for maximum returns on investment.
  7. Cross-Functional Collaboration: Sharing insights from the analysis across departments encourages collaboration. Sales teams can align their efforts with marketing strategies, while finance teams can use data to assess potential returns on investment.


Data Exploration

I was looking for a “dirtier” dataset that could be used to solve a real-world problem. The dataset had a total of 997, 213 rows, and 14 columns. I got the perfect dataset from data.gov

When exploring the dataset, I noticed something interesting. The people whose property type is Commercial, Vacant land, and apartment, did not provide info about residential types. I picked that pattern.

My missing values could be categorized as MAR(Missing At random). What that means is some missing values from the dataset could be explained by other variables within the dataset.

I checked for outliers. Outliers don’t mean that the data is bad. In my case, real estate outliers can be explained in unique property features, luxury markets, investment or speculation, etc. There are many ways to do this I used the boxplot method.

Data Cleaning using PowerBi — Missing Values

I began the cleaning using PowerBi. I dropped the columns that had a large number of missing values and the ones I wouldn’t need. Four columns in total. I checked for duplicates and removed them to ensure data consistency.

For the missing dates, I did a forward fill since the method works well with time series data. For my other two columns; property type and residential type, I used a DAX formula to find the mode in order to replace the missing values. I did a whole thread on my Twitter now known as X on what informed my choice of mode.

Insights I uncovered

  1. I enhanced interactivity by implementing user-selectable options for both year and month, empowering users to view specific periods of interest.
  2. I conducted calculations to derive the average assessed value, average sales amount, and average sales ratio, offering valuable insights into the dataset.
  3. Using visualizations, I showcased the distribution of property types, effectively highlighting the performance disparities among different categories.
  4. Leveraging correlation analysis, I investigated the impact of residential types on sale amounts, revealing intriguing relationships.
  5. By visualizing trends in assessed value and average sales amount across multiple years, I captured the dynamic nature of these metrics.
  6. Lastly, I explored town-specific property performance, analyzing assessed values and sales amounts to unearth distinctive trends and behaviors.

Visualizing using PowerBi

After filling in the missing values, I was able to create an interactive dashboard using the slicer options. I then noticed that the visualization was looking rather bland. Here is how it looked


So I went over to PowerPoint and created a simple background as per my need. Microsoft Suite saving the world. And that is how I was able to come up with this visualization. The background made the difference.

undefinedChallenges faced

The main challenges were:

  1. Laptop crashing
  2. Missing data

I have explained above how I handled the challenges. Moreover, I sought inspiration from online communities and tutorials to fine-tune my visualization skills, ensuring that my dashboards resonated with audiences.


In conclusion, this comprehensive analysis of the real estate dataset offered insights that can significantly inform strategic decisions within the company.

The interactive Power BI dashboard has transformed raw data into a visually engaging narrative, offering stakeholders an intuitive tool to navigate the complexities of the real estate market. From assessing the impact of residential types on sale amounts to visualizing trends in assessed values and sales amounts over time, each facet of the analysis contributes to a holistic understanding of the industry.


  1. Market Segmentation: Based on the insights gained, I would recommend segmenting the market based on property types, residential types, or locations. This can help the company tailor their marketing and investment strategies to different segments.
  2. Investment Opportunities: After identifying towns or areas with consistently high sales ratios and growing assessed values, I would recommend focusing on these areas for potential real estate investment opportunities.
  3. Data-Driven Marketing: I would propose utilizing the insights from the property type distribution visualization to tailor marketing campaigns. Highlighting the most and least-performing property types to target the right audience.
  4. Geospatial Analysis Expansion: If not already done, suggest expanding geospatial analysis to include factors like proximity to amenities, schools, and transportation. This can provide a comprehensive view of property value determinants.
  5. Collaboration and Reporting: Encourage cross-functional collaboration by sharing the dashboard and insights with various departments, such as sales, marketing, and finance. This can facilitate data-driven decision-making across the organization.
Discussion and feedback(0 comments)
2000 characters remaining