Menu

Weekday or Weekend: When Do Shoppers Really Click “Buy”?
Sandra G Sandra G
22 October 2025

A Deep Dive into BigQuery’s thelook_ecommerce Dataset Using Terno AI

Weekday or Weekend: When Do Shoppers Really Click “Buy”?

A Deep Dive into BigQuery’s thelook_ecommerce Dataset Using Terno AI

Introduction

In the fast-moving world of e-commerce, timing can be everything. From the perfect moment to launch a flash sale to identifying the right day to restock, understanding when customers are most active can shape an entire sales strategy.

That’s the question we set out to answer with Terno AI, the no-code conversational data-science platform that lets anyone analyze complex datasets instantly. Using Google BigQuery’s public dataset — thelook_ecommerce, we explored a simple-sounding but powerful question:

Do customers shop more during the week or on weekends — and what does that reveal about their behavior?

This seemingly small question opened a window into a much broader story of shopper habits, loyalty, and sales timing.

About the Dataset: thelook_ecommerce on BigQuery

Google’s thelook_ecommerce dataset is a treasure trove for data scientists and business analysts alike. It simulates a real retail business with rich, relational tables containing customer, order, and product information.

For this analysis, we used:

  • orders – Each transaction with order timestamps and item counts

  • order_items – Line-item level details including sale prices

  • users – Customer demographics like age, gender, and location
    By connecting Terno AI directly to the BigQuery dataset (no CSVs, no manual uploads), we could run conversational queries and auto-generate visualizations instantly — bar charts, pie charts, and even geo-maps — all in one interactive workflow.If you haven’t set up the connection yet, check out our step-by-step guide to connecting BigQuery with Terno AI.

How Terno AI Makes Analysis Effortless

Traditional analytics usually means hours of SQL coding, cleaning data, and building charts manually. With Terno AI, it’s as simple as asking a question in plain English — just like talking to a human data scientist.

When we started with:“I’m curious — do customers tend to shop more during the week or on weekends?”Terno AI automatically identified the orders table, detected the timestamp field, classified days into weekdays vs. weekends, and generated both the results and a bar-chart visualization within seconds.

Click https://sandra1.app.terno.ai/chat/share/328866b3-f160-4f48-bedf-3cf5202acdb1    to see the chat summary of our analysis.

Question 1: Do Customers Shop More During the Week or Weekends?

Prompt:_ " I am curious to know, do customers tend to shop more during the week or on weekends?"_

AI Insight Image
AI Insight Image
AI Insight Image
  • Weekdays: 89 827 orders

  • Weekends: 35 161 orders
    Insight: Customers overwhelmingly prefer shopping during the workweek, possibly when routines and digital engagement are highest. The weekend slump might indicate offline activities or reduced marketing campaigns.The numbers tell a clear story.

That’s nearly 72 % of all orders placed Monday – Friday.

Question 2: Do Shoppers Buy More Items per Order on Weekends?
Next, we wanted to know whether weekend shoppers compensate by buying more per order.

Prompt: “Can you check if people buy more items per order on weekends, or if it stays about the same as weekdays?. Maybe a side-by-side bar chart would help visualize it clearly."

AI Insight Image
AI Insight Image
AI Insight Image

Terno AI queried the num_of_item field in seconds and revealed:

  • Weekday Average: 1.45 items/order

  • Weekend Average: 1.44 items/order
    Basket sizes remain almost identical. Customers may be buying fewer times on weekends, but not necessarily smaller orders. This consistency reflects stable product interest and well-balanced inventory.

Question 3: Who Drives Weekend Sales — New or Returning Customers?

Customer retention is the heartbeat of any online business. So we asked:

Prompt:Are most of the weekend sales coming from new shoppers or returning customers?”

AI Insight Image
AI Insight Image

Terno AI classified each order based on whether it was a customer’s first purchase and found:

  • New Shoppers: 22 651 orders

  • Returning Customers: 12 510 orders
    That means roughly two-thirds of weekend sales come from new shoppers.

Insight: Weekends attract fresh audiences — likely driven by ad campaigns, social media discovery, or leisure browsing. Retention efforts could focus on converting these weekend first-timers into loyal weekday buyers.

Question 4: Does Gender Affect Shopping Patterns Across Weekdays and Weekends?

We explored whether the gender mix shifts between weekdays and weekends.
Prompt:Does the weekend attract a different mix of male and female buyers compared to weekdays?”

AI Insight Image
AI Insight Image
AI Insight Image
AI Insight Image
AI Insight Image

Insight: The answer surprised us with its balance. The gender split remains almost perfectly even. Weekends see a tiny uptick in female shoppers, but overall engagement is balanced.

Question 5: Which Age Groups Are Most Active on Weekends?

Demographics often tell a deeper story.

Prompt: “Which age group seems most active on weekends — younger customers or older ones?”

AI Insight Image
AI Insight Image

Using customer birthdates from the users table, Terno AI calculated a median age split and compared orders:

  • Younger (< median): 17 379 orders (49.43 %)

  • Older (≥ median): 17 782 orders (50.57 %)
    Insight: Older shoppers hold a slight edge. While activity is nearly even, this suggests that mature customers use weekends to browse at leisure, while younger audiences might already complete purchases during the week.

Question 6: Which Regions Generate the Most Weekend Sales?

Finally, we wanted to see where the weekend action happens geographically.

Prompt: “Which regions or states bring in the most sales during weekends?”

AI Insight Image
AI Insight Image
AI Insight Image

By joining orders, order_items, and users, Terno AI computed weekend revenue by location.

Insight: California dominates, followed by New York and Texas. Together, these three states generate over 40 % of weekend sales, mirroring national population and digital-shopping intensity.

Pulling It All Together: The Complete Picture

What Businesses Can Learn from This Analysis

  • Run Core Promotions Mid-Week (Tue–Wed).**
    **Engagement and AOV peak early in the week — capitalize with targeted campaigns and premium bundles.

  • Use Weekends for Acquisition.**
    **Since most weekend orders come from new shoppers, focus ad spend and social content on brand discovery and onboarding.

  • Add Flash Deals Friday Evening → Saturday Morning.**
    **Weekend orders dip, but short, time-sensitive offers can boost conversions when traffic is low.

  • Restock on Thursdays.**
    **With weekday turnover high and weekends still drawing new buyers, Thursday restocks ensure shelves stay full for weekend browsing.

  • Target Older Shoppers with Loyalty Offers.**
    **Slightly higher weekend participation among seniors (55 +) signals a receptive audience for personalized email campaigns.

  • Localize Marketing for Top States.**
    **Prioritize high-performing states — California, New York, Texas — for weekend-specific ads and inventory planning.

  • Introduce Measurable Discounts.**
    **Flat pricing limits analytical insight. Recording promo usage will help track campaign ROI and future uplift.

Why Terno AI Got It Right

What makes this analysis impressive isn’t just the findings — it’s the fact that every query ran cleanly, without special permissions or manual coding.

Each question, from “Do people shop more on weekends?” to “Which regions perform best?” was processed directly through Terno AI’s conversational layer. The platform automatically generated the right SQL joins, created professional-grade Plotly visualizations, and summarized insights in plain English.

In short, Terno AI turned analytics into a conversation, not a codebase. See the full Terno analysis here** **@ https://sandra1.app.terno.ai/chat/share/5bf2c1b7-971d-46cd-a94e-5510619f394e

Smarter insights start with a demo, Book a demo to explore how Terno AI turns data into decisions.

Final Thoughts: The Future of Conversational Analytics

This case study demonstrates how accessible and powerful no-code analytics can be. With BigQuery’s rich, production-grade datasets and Terno AI’s natural-language interface, anyone from business managers to marketing analysts can uncover deep insights in minutes.

The takeaway is simple:

“Timing matters and understanding your customers’ weekly rhythm is key to smarter decisions”.

Whether you’re planning promotions, managing inventory, or mapping growth, insights like these turn ordinary data into business strategy. So the next time someone asks “When do customers actually buy?”, you’ll have the answer and the visuals to prove it.

Ready to unlock insights from your data? Book a demo at Terno AI to get started.

Dataset Credit

All analysis is based on **Google BigQuery’s public dataset – **thelook_ecommerce, accessed directly via Terno AI’s BigQuery integration.

The Honest Number Was 83%: Leakage, Abstention, and a Complaint Router You Can Actually Deploy

18 August 2026

The Honest Number Was 83%: Leakage, Abstention, and a Complaint Router You Can Actually Deploy

The same complaint-routing model scores 96.3% or 83.2% depending on which three columns you leave in the training data. The high number is the intake form being read back to you. This is what the leakage audit found before a single model was trained, why 83.2% is the honest figure, and how the same model — given permission to say "I don't know" — becomes deployable at 90.7% accuracy on 79.5% of traffic.

Read More
EdgeGuard: AI-Driven Predictive Maintenance for Power Transformers

29 July 2026

EdgeGuard: AI-Driven Predictive Maintenance for Power Transformers

Power transformers are among the most critical assets in electrical distribution infrastructure. Their unexpected failure can result in power outages, safety hazards, equipment damage, expensive repairs, and long service interruptions. Traditional transformer maintenance practices often rely on periodic manual inspection, offline testing, or run-to-failure maintenance. These methods are expensive, slow, labor-intensive, and unable to detect rapidly developing faults in real time. EdgeGuard is an AI-driven, edge-computing predictive maintenance system designed to continuously monitor transformer health and forecast failures before catastrophic damage occurs. The system acts as a retrofittable “Digital Doctor” for distribution transformers by combining low-cost industrial sensors, an ESP32 microcontroller, local intelligence, machine learning-based risk prediction, autonomous relay control, and a real-time web dashboard. The proposed system monitors six major transformer health indicators: temperature, humidity, vibration, oil level, current, and voltage. These signals are normalized and processed through a Multi-Layer Perceptron neural network to classify transformer condition and estimate failure risk. If the predicted risk crosses a critical threshold of 80%, EdgeGuard automatically triggers a relay through GPIO 26 to isolate the transformer from the electrical network. The system also supports secure remote control, dashboard monitoring, API-key-based hardware authentication, JWT-based user access, WebSocket live updates, and automatic live-hardware detection. With an estimated deployment cost of approximately ₹3,850, EdgeGuard offers a low-cost alternative to conventional transformer monitoring systems. Its cloud-independent operation and edge-based decision-making make it especially useful for rural and semi-urban distribution grids where connectivity and maintenance resources are limited.

Read More
ANALYZING TOXIC USER BEHAVIOR AND RISK PATTERNS IN ONLINE GAMING PLATFORMS

28 July 2026

ANALYZING TOXIC USER BEHAVIOR AND RISK PATTERNS IN ONLINE GAMING PLATFORMS

This study shows that behavioral data alone can't reliably predict gaming toxicity — but a risk-based model combining behavioral and engineered features does a much better job of flagging the small segment of high-risk users driving disproportionate harm.

Read More

- Your AI-Data Scientist

Turn your data into decisions with Terno.

Check out Terno