Aisle Be Back: Using AI in Retail Real Estate
How to use ChatGPT to help you go from clicks to bricks
Using AI to help you go from clicks to bricks
It’s hard to talk about DTC brands without also talking about innovation. And it’s hard to talk about innovation without talking about AI.
And while there’s been a lot of chatter on AI for marketers, there hasn’t been a ton (that’s accessible) in the world of “clicks to bricks.”
Some really cool companies using include spatial.ai and placer.ai, but they cost tens of thousands of dollars per year; for cash-strapped DTC brands, that’s a big cost to budget for on a channel that doesn’t exist yet (retail), or that doesn’t make up a ton of revenue (although the counterpoint would be that tens of thousands of dollars is a small insurance policy to ensure your investment is a good one — more on the cost of a mistake here).
I’ve been messing around with ChatGPT and various other AI tools a TON lately and, with the release of OpenAI’s latest Vision capabilities, I couldn’t help but want to share everything I’ve done with it and recommendations it also gives about itself. Let’s dive in, and in the familiar structure of each functional area:
Strategy & Finance
Real Estate & Lease Administration
Design & Construction
Store Operations & Management
And in case you’re wondering, the below list was generated with ChatGPT 4 (with a few iterations on the prompt). I’ve highlighted the broader use cases I’ve actually tried successfully by appending the example with *** at the end.
As you try any of these prompts, first send this “pre-prompt” to Chat GPT:
"Act as if you're an experienced store development leader with expertise in integrating AI solutions into retail strategies."
And note: many of these prompts require GPT 4 which has various plugins (PDF readers and web browsers). It’s a month-to-month subscription, and very affordable ($20/month). GPT 3.5 may get you decent results, but the slight cost is very much worth it in my opinion.
Strategy & Finance
1. Process and analyze vast amounts of data
Prompt: "Analyze this dataset to identify patterns related to geospatial analytics."
Input Data: A dataset containing store locations, sales figures, and customer demographics (Best format: Excel/CSV).
Example Output: "Based on the data, there's a higher sales volume in urban areas with a dense population aged 25-35."
2. Create financial models and predict future performance
Prompt: "Based on this historical data, forecast the next quarter's performance."
Input Data: Historical sales and expense data for the past year (Best format: Excel/CSV).
Example Output: "Predicted sales for the next quarter are approximately $1.2M, a 10% increase from the same quarter last year."
3. Gather information on competitors and market trends***
Prompt: "Provide insights on competitors' store locations and market trends in the retail sector."
Input Data: A list of known competitors and relevant market sectors (Best format: Text/Word).
Example Output: "Competitor A has a strong presence in the West Coast, while market trends indicate a growing preference for eco-friendly products."
4. Simulate different retail scenarios
Prompt: "Simulate the financial implications of opening a store in these three locations."
Input Data: Details of the three potential store locations, including rent, expected footfall, and local demographics (Best format: Excel/CSV).
Example Output: "Location B offers the best ROI given its lower rent and higher expected footfall."
5. Provide recommendations on budget allocation***
Prompt: "Recommend a budget allocation for marketing and operations based on past performance."
Input Data: Previous year's budget and performance metrics (Best format: Excel/CSV).
Example Output: "Allocate 60% to marketing to tap into the identified growth segments and 40% to operations to maintain efficiency."
Real Estate & Lease Administration
1. Understand and interpret lease provisions***
Prompt: "Explain the implications of this lease provision."
Input Data: A specific provision or clause from a lease agreement (Best format: PDF/Text).
Example Output: "This provision indicates that any structural changes to the property require the landlord's approval."
2. Assess potential real estate deals***
Prompt: "Evaluate this real estate deal based on our set criteria."
Input Data: Details of the real estate deal and the company's criteria for property acquisition (Best format: Excel/CSV).
Example Output: "This deal scores 85/100 based on our criteria, making it a viable option."
3. Provide insights on market rates during negotiations***
Prompt: "What are the current market rates for retail spaces in this area?"
Input Data: Location details and type of retail space (Best format: Text).
Example Output: "The average market rate in this area is $25 per square foot."
4. Provide plain-language explanations of complex legal provisions:
Prompt: "Translate this legal provision into simpler terms."
Input Data: A specific complex legal provision from a document (Best format: PDF/Text).
Example Output: "This provision means that both parties agree to resolve disputes through arbitration rather than court."
Design & Construction
1. Analyze blueprints and identify potential issues***
Prompt: "Review this blueprint and suggest any optimizations."
Input Data: A Lease Outline Drawing with optimal dimensions, room/pad minimum dimensions, etc. (Best format: PDF/Image).
Example Output: "Consider reducing the size of the storage area to allow for a larger sales floor, enhancing customer experience."
2. Estimate construction costs based on design choices***
Prompt: "Estimate the construction cost for this design."
Input Data: Design drawings and material specifications (Best format: PDF/Image).
Example Output: "The estimated construction cost for this design is around $500,000."
3. Track vendor performance and manage invoices
Prompt: "Provide a performance report for this vendor based on past projects."
Input Data: Previous projects' details, invoices, and feedback related to the vendor (Best format: Excel/CSV).
Example Output: "Vendor X has a 90% on-time delivery rate and an average feedback score of 4.5/5."
4. Provide feedback on test fits to enhance customer experience***
Prompt: "Suggest changes to this test fit to improve the customer experience."
Input Data: Test fit drawings and customer feedback data (Best format: PDF/Image).
Example Output: "Relocate the fitting rooms closer to the entrance for better accessibility."
5. Track and manage tenant allowances
Prompt: "How much tenant allowance do we have left from the landlord for this store?"
Input Data: Lease agreement details and previous tenant allowance usage records (Best format: Excel/CSV).
Example Output: "You have $10,000 remaining in tenant allowances for this store."
Store Operations & Management
1. Analyze sales data and provide insights on store performance
Prompt: "Analyze the sales data for this month and compare it to the same month last year."
Input Data: Sales data for the current month and the same month in the previous year (Best format: Excel/CSV).
Example Output: "Sales have increased by 15% compared to the same month last year."
2. Create staffing models based on store size and expected footfall
Prompt: "Recommend a staffing model for our new store opening next month."
Input Data: Store size, location, and expected customer footfall (Best format: Excel/CSV).
Example Output: "For the store size and expected footfall, you'll need 5 full-time and 3 part-time staff."
3. Draft and refine Standard Operating Procedures (SOPs)***
Prompt: "Review and suggest improvements to this SOP for inventory management."
Input Data: Current SOP document for inventory management (Best format: PDF/Word).
Example Output: "Consider adding a section on handling damaged goods to improve inventory accuracy."
4. Track inventory levels and provide reorder recommendations
Prompt: "Which products are running low on stock and need reordering?"
Input Data: Current inventory levels and sales rate of products (Best format: Excel/CSV).
Example Output: "Products A, B, and C are running low and should be reordered immediately."
5. Process and analyze customer feedback
Prompt: "Analyze this customer feedback and identify areas of improvement."
Input Data: Collection of customer feedback forms or reviews (Best format: Excel/CSV).
Example Output: "Customers have mentioned long wait times during peak hours. Consider additional staffing during these times."
Other
1. Answer queries on store locations, hours, and promotions***
Prompt: "Provide details on our store locations in New York and any ongoing promotions."
Input Data: Store database with location details and current promotional offers (Best format: Database/Excel).
Example Output: "We have 3 stores in New York. Our flagship store has a 20% off promotion running until the end of the month."
2. Train new employees on company policies and procedures***
Prompt: "Provide a summary of our company's return policy."
Input Data: Company's policy and procedure handbook (Best format: PDF/Word).
Example Output: "Our return policy allows for returns within 30 days of purchase with a valid receipt."
3. Draft communication and action plans during crises***
Prompt: "Help draft a communication plan in response to this product recall."
Input Data: Details of the product recall, affected batches, and customer communication channels (Best format: Text/Word).
Example Output: "Inform customers immediately via email and social media about the affected batches and offer a full refund or replacement."
Beware of ChatGPT’s limitations!
Out of date data: Unless you use the browser plug-in, the data only goes up to Sep 2021; it may not give you the most up-to-date information if your query is about something after that date
Everything requires specific context: You still need to learn real estate! Asking it for “market rents in Ohio,” for example, may give you a wildly different response the more relevant “retail market rents in Cleveland, OH for spaces between 2,000 and 4,000 square feet; assume 10 year lease term and no tenant allowance”. Deals are complicated, and while ChatGPT can be a great tool, you still need to know how to word your questions in the optimal way
Confidentiality concerns: be cautious of uploading sensitive documents or sharing confidential information with ChatGPT. I’m no data privacy expert, but keep in mind that you’re uploading/sharing information with another company when you provide any input data/files.