Deep.bi - Real-time, Deep Data Analytics Platform For Ecommerce
0 likes853 views
This document provides an overview of the deep.bi analytics platform for ecommerce companies. It describes how deep.bi collects detailed ("deep") data on products, customers, and customer behaviors. This deep data is analyzed to provide real-time insights. Deep.bi helps ecommerce teams improve performance in areas like merchandising, marketing, customer service, and site experience. It does this by tracking custom metrics and providing customizable dashboards. Deep.bi can be used as a standalone tool or integrated with other systems through its API.
![Enrich web data with business information
2015-05-15T00:26:41.328Z,3,D,[ip_hidden],i1xszg0f-19hqrje,"Mozilla/5.0 (Windows NT 5.1)
AppleWebKit/537.36 (KHTML, like Gecko) Chrome/42.0.2311.152 Safari/537.36",”[url_hidden]",
7279848891,@906,"https://www.google.pl/",vuser-history-allegro-1-
hc20150509.1,"122_100003_Park@700:html_620x100_single_banner:See offer"
IP, URL, cookie, user-agent, timestamp, text…
Raw, web browsing data
Real-time data enrichment
Structured data
Enriched data
see the next slide](https://image.slidesharecdn.com/deepbiecommerce-160107204751/85/Deep-bi-Real-time-Deep-Data-Analytics-Platform-For-Ecommerce-20-320.jpg)
![Business context based on internet provider
Example use:
• [telco] competitors’ users-
> acquisition
• [telco] our users ->
retention/up-selling/cross-
selling
• people from particular
company or company type
• ISP name or Organization name
• Organization type:
• Commercial
• Organization
• Government
• Military
• University/College/School
• Library
• Content Delivery Network
• Fixed Line ISP
• Mobile ISP
• Data Center/Web Hosting/Transit
• Search Engine Spider
• Reserved
• Mobile brand
• Net speed](https://image.slidesharecdn.com/deepbiecommerce-160107204751/85/Deep-bi-Real-time-Deep-Data-Analytics-Platform-For-Ecommerce-23-320.jpg)
Deep.bi - Real-time, Deep Data Analytics Platform For Ecommerce
- 1. deep.bi for ecommerce © 2015 by deep.bi
- 2. Every business is different So, there’s no “one-size-fits-all” analytics solution
- 3. Deep.bi is a real-time, deep analytics platform It helps ecommerce teams improve their performance by providing current and precisely tailored insights.
- 4. Operational excellence and performance Category Managers / Merchandisers Find new revenue opportunities, improve rotation, promote best sellers, eliminate non-sellers, up-sell, cross-sell. Marketers Deeply understand your customers and what brings them to purchase. Then optimize your marketing spent across markets and channels. Customer service Maintain more personalized customer relationships. Fulfill customer needs based on their value. UX / Design Team Improve your site constantly by micro- updates and monitor their impact on conversion in real-time. Tech / IT Monitor site performance. Quickly deliver new reports and functionalities. Focus on business value, outsource heavy engineering stuff to experts. Executives / Managers Get the up-to-date big picture of your store performance. Monitor key metrics like conversion rate, revenue, retention in real-time.
- 5. Custom defined, real-time metrics
- 6. Historic data is excellent for strategy and planning. Real-time data gives the right, current context and helps take action intraday to improve daily profits. Is real-time data useful in ecommerce?
- 7. People are complex. We make decisions based on many factors. deep.bi helps extract product, customer and behavior detailed attributes to find real purchase patterns. What’s deep data and why it’s important?
- 8. Example of tracked product dimensions: • Name • Brand • SKU • ID • Offer Price • Regular Price • Save Amount • Availability • Specs • Weight • Color • Size • Box Size • categories Deep data: Product Deep.bi extracts detailed product characteristics and make it available for analysis with all other data. Example insights: • Most viewed but unavailable products by brand • Worst selling shoes by color • Worst selling smartphones by battery life, weight and size • Top selling at full price Nike men running shoes in California by detailed features (size, color, weight)
- 9. Deep data: Customer Example of tracked user dimensions: • User • Cookie • Email • First Name • Last Name • Gender • Location: • ZIP • City • Region • Country • Population • Device • Type • Brand • Model • Version • Internet Provider • Type • Name Deep.bi can enrich raw user data (like IP) with rich, business-useful information. That allows to deeply understand customers and prospects. By combining this data with behavior and product data, store managers can optimize marketing spend, traffic sources and inventory.
- 10. Deep data: Behavior Example of tracked event behavior dimensions: • Event Type • Timestamp • URL • Referrer Host • utm_source • utm_campaign • utm_mediium • Active Engagement Time By knowing detailed characteristics of users’ behaviors ecommerce managers can build detailed users profiles, discover patterns and track key metrics like: • RFM (Recency, Frequency, Monetary), • LTV (Customer Lifetime Value) • CAC (Cost of Acquiring Customer) • AOV (Average Order Value) • COS (Cost of Sale – Ad Spend / Revenue) • Shopping Cart Abandonment Track any important user behaviors like: • product views • adding items to cart or wish list • purchases • delivery confirmations • newsletter subscription • return request and any other. Measure detailed behavior characteristics, like active engagement time while reading product descriptions.
- 11. Find deep buying patterns and correlations Example: what buy and how behave people from small cities who use iPhone and came from Facebook and what product features are important for them, etc. Adjust prices and optimize stock Track best-sellers in real-time, optimize price to maximize profits. Find non-sellers and features that are common for them (maybe brand, color, size, etc.) Predict Having collected detailed, raw and enriched data you can build customer predictive models using machine and deep learning algorithms. Combining deep data with real-time insights creates unprecedented possibilities
- 12. deep.bi can be used standalone or as a part of your current solutions deep.bi dashboards Use our predefined dashboards and metrics or customize them to fit your needs. deep.bi API Use API to build your own, external dashboards or embed our analytics in your current systems. deep.bi data platform Use deep.bi as a data collector and enricher to build your customized solutions, e.g. prediction system
- 13. Quick and easy to implement Minutes/hours vs. weeks or months. Just embed our script and start tracking. Real-time Data latency <1s Super-fast Less than a 1s on TB-size data store Scalable Petabytes and more Highly flexible No relational data modeling (schema-flexible), no data pre-aggregation, no defining reports up front. Just dig into data and explore it. Add new dimensions (columns) and metrics on the fly. Why deep.bi is unique? (technical explanation in blue)
- 16. More analysis examples Merchandising insights: • Product, SKU, Category, Brand Report: views, added to cart, sold, CR, etc. • Hot Products (most viewed, most added to wish lists, most added to carts) • Non sellers, cold products • Most discounted products Lifecycle marketing: • Who viewed but did not purchase • 30/60 day repeat purchase rate • Best products/categories/brands for repeat purchase • Products/categories purchased together • Customer Lifetime Value by channel (day 1, 90, 180, 360) Conversion & performance: • Purchase funnel incl. abandoned cart • Conversion & attribution by traffic source/referral, campaign Customer intelligence: • Best customers by detailed characteristics (demographics, RFM, best full price customers, etc.) • Cold customers: not buying, low AOV • Customer Lifetime Value (Day 30, 90, 180, 360) • How long it takes customers to purchase UX improvements • Checkout improvement: find that most customers are abandoning carts when they reach a confusing section in your checkout process that you were not even aware of.
- 17. APPENDIX Going beyond data silos
- 18. Going beyond silos Website: Google Analytics Campaigns: Agency reports Apps: Dedicated monitoring tools Other systems: Call center IVR, emails Instead of integrating current reporting tools we gather all the single events that our customers generate. Data is stored in silos. Reporting tools provide aggregated reports impossible to integrate around single customer.
- 20. Enrich web data with business information 2015-05-15T00:26:41.328Z,3,D,[ip_hidden],i1xszg0f-19hqrje,"Mozilla/5.0 (Windows NT 5.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/42.0.2311.152 Safari/537.36",”[url_hidden]", 7279848891,@906,"https://www.google.pl/",vuser-history-allegro-1- hc20150509.1,"122_100003_Park@700:html_620x100_single_banner:See offer" IP, URL, cookie, user-agent, timestamp, text… Raw, web browsing data Real-time data enrichment Structured data Enriched data see the next slide
- 21. * Coming soon Get deeper insight - enrich raw user data 50+information from one interaction Purchase intent Device Time Location ISP Online context Weather*Demographics
- 22. User location details Example use: • international travellers • townspeople • people in mountains • rainy day • Country • Region • City • ZIP Code • Population • Latitude & Longitude • Time zone • IDD prefix to call the city from another country • Phone area code • Mobile Country Code (MCC) • Mobile Network Code (MNC) • Elevation • Weather at the moment of event
- 23. Business context based on internet provider Example use: • [telco] competitors’ users- > acquisition • [telco] our users -> retention/up-selling/cross- selling • people from particular company or company type • ISP name or Organization name • Organization type: • Commercial • Organization • Government • Military • University/College/School • Library • Content Delivery Network • Fixed Line ISP • Mobile ISP • Data Center/Web Hosting/Transit • Search Engine Spider • Reserved • Mobile brand • Net speed
- 24. Detailed information about user’s device Example use: • smartphone users • Apple users • Samsung Galaxy users • Google browser users • Device Type • Device Brand • Device Model • Device Operating System • Operating System Producer • Browser • Browser Producer
- 26. • Deep.BI provides two ways of ingesting data: using direct REST API or through JavaScript snippet. • In both ways data is sent in JSON format. • Deep.BI is event-based analytics what means that we collect full scope of information about every single tracked event that happens on your site. • We provide full flexibility of defining data dimensions: • You can define as many dimension as you want (e.g. 100+ dimensions for every event) • You can build infinite hierarchy of dimensions, e.g.: product.specification.size.width.unit, or category.category.category… • On the following slides we provide some examples of data dimensions for: product, user and user behavior. Deep.bi Data Ingestion API
- 27. { "type": "product", "title": "Nike Men's Dual Fusion Run 3 Running Shoes", "brand": "Nike" "sku": "B00LHX3IWG", "productId": "653596-012", "offerPrice": "$44.99", "regularPrice": "$199.99", "saveAmount": "$155.00”, "availability": true, "specs": { "weight": "2.5 lb”, "object_size": "11.0 x 7.0 x 4.0 in”, “color”: “Grey/Black/White”, “size”: “7.5 US”, "asin": "B00LHX3IWG", }, "categories": { "name": "Clothing, Shoes & Jewelry", "name": "Men", "name": "Shoes", "name": "Athletic", "name": "Running” }} Deep data: Product JSON
- 28. { ”user”: { ”cookie": ”43dad90aad9a88a", ”email": "seb@deep.bi" ”fist_name": "Sebastian", "last_name": "Zontek", "gender": "male” }, "location": { "zip": "94107", "city": "San Francisco", "region": "CA", "country: "USA", "population": "837442" }, "device": { "type": "smartphone", "brand": "Apple”, “model”: “iPhone”, “version”: “5S” }, "isp": { "type": "mobile operator", "name": "T-Mobile” }} Deep data: Customer JSON
- 29. { "event": { "type": ”product-view", "timestamp": ”2015121008550145" "url": "http://www.amazon.com/Nike-Fusion-Unvrsty-White- Running/", "referrer_host": ”facebook.com", "utm_source": "male”, "utm_campaign": "social_media_paid", "utm_mediium": ”cpc”, "active_engagement_time": “120s”} } Deep data: Behavior JSON
Editor's Notes
- #3: No two businesses are the same. Even if you sell the same kind of products or services, you and your competition have different marketing approaches, product presentations, store design, target demographics, brand loyalists, ethics on how the company is run, and employees.
- #4: optimizing acquisition channels (revenue growth, reducing marketing costs) buying and selling the right products at the right time servicing customers (returns, delivery) better understanding customers’ behavior to engage them
- #7: Ecommerce real-time analysis examples: Monitor intraday revenue targetsYou can act instantly when sales goals are below the expected level for example by blasting an email campaign or buying traffic. Optimize acquisition channels (referral traffic)Stop wasting money for low quality traffic/campaigns. Test channels and campaigns variations and shift budget to the best performers. Sell and promote the right productsOccasionally you’d want to focus your sales on particular products, category or brand. Monitor in real-time these special actions and make improvements to maximize profits.Also, watch real-time trends to boots promotion of the top selling products. Conversion ratesUnderstand how minor changes, social media promotions, email newsletters affect your conversions in real time. Impact of website updatesGet to know when to apply changes to your store (lowest traffic) as well as monitor in real-time how the updates affect your metrics (sometimes errors happen – know it instantly!).