Hello, Habr! On June 30 at 18:00 Moscow time, we will hold an online meetup for analysts. Speakers will discuss regional A/B tests, product listing management in online stores, predicting profits from new features, and data science in product delivery.
Below you will find, as always, the abstracts of the talks and all the necessary links.
Reports
Regional A/B Tests: Why They Are Needed and How They Work — Igor Krasovsky, Avito
What to do if the test group in an A/B test is exactly unknown, and the impact on the user occurs offline, for example, through regional TV advertising? How to form a control group for the skewed test? How to measure the effect and distinguish it from random error? I will share how we answered these questions at Avito and what problems we encountered.
About the speaker: Has been at Avito for just over 2 years, previously worked in eCommerce and IT consulting. Currently, I work in the Core Analytics team, which is responsible for areas such as Data Management, Strategic Analytics, Core Analytics Platform, and Key Account Analytics.
The Best Data Products Are Born in the Field — Marina Kalabina, Leroy Merlin
Most of our online orders are collected from store showrooms, not warehouses. This leads to discrepancies between what is displayed on the site and what we can actually gather.
Due to the high turnover of goods in stores and the complexity of inventory management systems, errors can be detected automatically. Based on our knowledge of systems and using social engineering, we proposed a solution that would automatically identify problematic products and adjust their stock before publishing on the website.
About the speaker: 9 years of experience at Leroy Merlin. Initially opened stores, then worked in them, and now I am organizing the inventory. Built a team and launched a data product in 6 weeks.
Growth Model: Predicting Profit from Features for Prioritization — Pavel Mikhailov, Ostrovok.ru
- We are building a growth model — a framework based on cohorts and key metrics that models revenue in the medium term.
- We convert product and business metrics into money using the model.
- We evaluate potential profit from features with examples.
About the speaker: Head of growth at Emerging Travel Group (Ostrovok.ru) with an analytical background. I generate, develop, and test growth hypotheses.
How data science helped Avito Delivery — Dima Sergeev, Avito
… Or a story about how to stop offering users to buy a "Kamaz Cabin" for Delivery. There are over 60 million products on Avito. It’s not easy to determine if a seller can fit it into a box sized 120×80×50 and send it to a buyer in another city.
Sometimes we make such mistakes: we offer delivery where it shouldn't exist and vice versa. I'll share a bit about how we tackle this issue and what results we've achieved.
About the speaker: For the past year, I've been focusing on analytics at Avito Delivery. Prior to that, I worked in analytics at OZON for three years.
Ask questions in the streaming chat — we will answer the most interesting ones live. After each presentation, you will have the chance to chat with the speaker.
Passwords and credentials
Broadcast starts on Tuesday, June 30 at 18:00. We plan to finish by 20:40. You can immediately hit the "remind me" button during the broadcast so you don't miss anything.
If you want a reminder with a link to the broadcast via email, you can .
See you online!
Source: habr.com
