September Digest: A Recap Of All Things Analytics
Case studies, reports, analysis, and tutorials you may have missed
Happy end of September!
and are here to wrap up the month with a whirlwind of all things analytics across product, data science, business, academia, and research.We put together some of the case studies, reports, analyses, and tutorials you may have missed, along with events, recaps, and new interesting publications.
🔊 Advocating for analytics
in How To Do a Root Cause Analysis“Quite a few companies claim their products can do the root cause, activation, or causation analysis in seconds (Heap Illuminate, Loops, Inner Trends). Yes, we can locate early predictive indicators of user churn or product abandonment and be proactive instead of reactive (with some questionable degree of trust and many… many exceptions). However, automating this is a challenge. Predicting user behavior should be tailored to your product, your tech stack, your market, and your user. Using industry benchmarks, competitors’ data, popular playbooks, or even your own historical data often doesn’t guarantee the answer to WHAT and WHY.“
🔥 September highlights
Contentsquare acquired Heap
: September ended with a bang for analysts, with news just coming in yesterday that Contentsquare (or what I call product-analytics-for-big-enterprise, or amplitude-but-even-more-expensive) acquired Heap:“Together, Contentsquare and Heap will provide teams with a 360 view of every experience across web and apps, offering an all-in-one analytics platform and advanced alternative to traditional analytics. Ready to deliver better experiences faster? We are.” - shared by Contentsquare team in their announement yesterday.
The Data Science job market has stabilized with a 25% M/M increase in openings
Hiring across data science and analytical roles is returning to the normal pattern, with almost all data science positions being up month-over-month, as shared by Interview Query in their September 2023 Data Science Job Market Update.
The Best SaaS Blog Posts and Resources Library
ChartMogul, a subscription analytics platform, revealed its list of top SaaS reads from the basics SaaS Metrics — 101 to more advanced topics like Go-To-Market and Product-Led Growth (PLG):
“The library encompasses hours and hours of reading, listening, and viewing material. Please, use it as a reference tool and don’t try to get it all in one sitting!“
(Thank you, ChartMogul team, for including my publication on such a wonderful list!)
Finalists Announced For Best Data Visualization 2023
An annual award celebrating excellence and beauty in data visualization and data storytelling presented their 2023 finalists to compete for the winner title (to be announced soon by the Data Visualization Society (DVS).
📈 Industry reports and new benchmarks
🤓 Analysis and case studies
How Strava Accelerated User Engagement - Inside analytics at Strava: its process and tooling. A case study on the Route Detail page redesign that improved user activity in the app.
Quo vadis, Data Open source from
- A case study and review based on Snowplow, dbt, Rudderstack, and Iceberg examples.How we accelerated the adoption of Amplitude at Preply - A new analytics tool is new and shiny, but the real value for a bigger scope only comes when you train the different teams to use the tool. This post published by Preply, an online language learning marketplace run by a small Ukrainian team, is a rare example of how to develop such an adoption and training plan.
⚙️Know your craft
Product and marketing analytics
How To Do a Root Cause Analysis - Olga’s approach and framework for running a root cause analysis to understand unexpected user behavior change or metric decline.
About Data User Experience - a deep dive into event analytics by Timo: its sources, challenges, and data structures. A consolidated guide on how to work on “data user experience” for your data customers.
Do Sequence analytics like writing SQL - from Timo: I have been playing with Motif Analytics for a while now and really like it. It’s a pretty engineering approach to sequence analysis, but this makes it quite powerful. It takes a while since you have to learn a “new” query language.
Mastering Customer Retention Strategy by Amplitude. From Olga: I don’t believe there are many strong deep-dives on retention out there (except mine, duh), but this recent Amplitude strategy guide is practical and easy to understand.
Develop a Cost-aware Culture for Optimized and Efficient SaaS Growth. From Timo: The post is better than its title. It has a good breakdown of different types of SaaS cost metrics. Especially the unit economics are often not really covered in SaaS metrics playbooks and setups.
Marketers, what you need to know about iOS17, LTP, and privacy trends. From Timo: I guess Marketing analysts already hate the yearly WWDC Apple event. Because Apple always drops something new that will reduce the usable data for campaign attribution. This time, it is Link Tracking protection. UTMs should be fine, but things like GCLIDs or Hubspot IDs should be tested when it comes out.
Data Science and ML
ML: Bias and Variance - ML Basics 101.
A gentle and quick introduction to Neural Networks from Photon-Lines Substack. What a gem.
Create Python plots in Matplotlib with LLM - PlotAI. Free, fast, and intuitive. It works in Python scripts and notebooks (Jupyter, Colab, VS Code). Props to Polish ML researchers.
⛔ Hot Seat
Recent publications that made us raise an eyebrow.
Why IT Professionals Choose Tableau for Modern BI and Analytics
From Olga: if you are up for a laugh and light entertainment, Tableau released a whitepaper (a whitepaper!) to boost self-promotion and remind us why dashboarding must be heavy and hard. In case we dare to think otherwise.
Tackling Data's Biggest Culture Problem by Chad.
From Olga:
“Tech isn't the reason for data quality issues. Humans are.” Improve data quality and …buy another piece of tech?
I greatly respect
, and I appreciate his advocacy for better data quality (make sure to check his newsletter ). And yet, I disagree that (1) contracts are the answer to data issues, (2) conventional solutions don’t work. They do. And (3) investing in the quality of people and processes rather than tooling should be a path to success. Regardless, congratulations to Chad on a new product launch, and good luck to the team with Gable.ai!Emotions aren't terrible by Hyperquery in
From Olga: No. We (analysts) are getting paid to represent due diligence, data accuracy, and facts. And emotions are terrible.
Airtable is focusing on enterprise, and the PLG world roars
From Timo: The Airtable announcement is not really important here. They said they want to focus more on enterprise customers; therefore, we can imply that they are moving away from a clear PLG focus. Naturally, that triggered all PLG influencers who all commented on it (and what kind of mistake it would be). My take: Companies need to make money, and the unit economics are not visible to us from the outside. So, enterprise contracts can earn a lot of money. It is more difficult if you are a content creator who puts all your eggs into a product or concept bucket. You have to fight for it, whatever it takes.
❤️ Favorite publications this month
Bookmarked to re-read favorite takes
- : So you want to sell to the enterprise?
Bessemer Venture Partners: Understanding churn and building an action plan to fix the proverbial “leaky bucket.”
- : A Primer: Subscription vs Usage-Based Pricing Models
📊 Monthly Chart Drop
🍸 Drink and Mingle
Upcoming events, meetups, talks, and webinars.
Oct 16-19, San Diego, London, Sydney, and online: Coalesce 2023.
Oct 17, San Francisco: SingleStore Now: The Real-Time AI Conference
Oct 10-12, Chicago: DAA's OneConference
Oct 24-26, virtual: CDP Week 2023
Oct 25-26, virtual: The DE And ML Summit run by
Oct 26, Bay Area: IMPACT: The Data Observability Summit
Oct 30, Santa Clara: DATA CLOUD WORLD TOUR
Sep - Nov, online: ML⇄DB Seminar Series
Nov 1-2, virtual: Snowday: The View Ahead
Nov 7-8, virtual: Causal Data Science Meeting 2023
Nov 8, virtual: Impact: The Data Observability Summit
Nov 29, San Francisco: DSS Using Gen AI & ML in the enterprise
Dec 1-3, virtual: PyLadiesCon 2023
Thanks for reading, everyone!
Olga + Timo
I guess I have to write another post called "Nvm emotions are terrible." 🤷♀️
Haha, join me in harvesting the unique generation of analysts-robots!