DataBS Conf attendee registration is open!
Let's cut to the chase, the link to where you can register is below:
Team Tito
Some important details:
- The event will be on Wednesday, Sept 24, 10AM - 5PM EDT (UTC-4)
- Registration is technically "pay what you want"
- Zero dollars is a perfectly VALID entry for a ticket
- Registrations are capped at 3000 because I don't have the budget to pay beyond that number for a zoom webinar license for a one time event
- Any ticket sales/donations will go towards offsetting the direct costs of the event, most of it is the license
- The subscription fees from this newsletter you're reading right now is actually what is providing the budget for this. (So buying a subscription puts money into the same place and also gets you extra newsletter posts as a bonus!)
- There will be recordings uploaded to YouTube after
As of this writing, 6 of the 14 speakers have already confirmed that they're going to speak at the event. The final program, speaker profiles, talk descriptions will be updated on The Website when that information comes in. You can be sure that I'll make mention of updates as they come in.
Pre-show recording invitations to come
There's never enough time slots to give to very compelling talk proposals, and many of the talks submitted to the conference are super niche to this particular conference's theme. So the volunteers and I are going to pick out a selection of proposals that we regret having to pass on for time reasons and invite those speakers to record a version of their talk as a pre-show thing. That process is going on this week, which is why rejection emails have not been sent out yet.
Who's speaking?
You'll find out later once I have all the profile info!
But here's a quick preview based on what people have submitted to me already, in no particular order. These folks will have profile pictures and links to their socials, and other details later. For now, here's just the talk description.
Be warned that all this is subject to change! Speakers can make changes up to the last minute.
Lunchtime project to cabinet briefing – David Hood
The acute covid response had lots novel, rapidly changing data. My personal lunchtime and evenings project summarising NZ’s data had a weirdly high national profile. I have thoughts about why, and how to sustainably contributing usefully in times of crisis.
10 Red Flags That Tell You Your Multi-Market Research Data Went Sideways (And You're Only Finding Out Now) – Ramona Daniel
We've all been there. Staring at survey data that looks technically correct but feels fundamentally wrong. This is the session for people who've ever had to explain why the data from three countries is completely unusable, and you're only figuring this out now. This isn't about pointing fingers; it's about recognising that multi-market research is inherently messy, and sometimes the most valuable skill is knowing how to avoid all the pitfalls.
(Not Quite) Instant Awesomeness: corporate genAI adoption gone wrong, and how to put it back on-track – Q McCallum
Executives often see genAI as an instant awesome switch for their company. Their desire to use this technology overshoots its capabilities, leading to stalled projects and high-profile mishaps. Why does this keep happening? Why don't the execs see the danger signs? And what can you, the data practitioner, do about it? Using real-world genAI goofs, I'll share how to guide your stakeholders to better decisions and more successful rollouts. Who knows? You might talk them out of using AI altogether.
The Pitfalls of Prototyping Proof-of-Concepts in Prod – Mark Rieke
Sometimes, the best way to test a new piece of software is to run tests on a subset of real users in prod. You'll catch the weird edge cases that you hadn't considered during development and can iterate so that the next release is more resilient. But watch out! There is no such thing as a "one time proof of concept" in a production environment — once you deploy, you have a production system. In this talk, I'll share an example of how I learned this lesson the hard way: a proof of concept experimentation engine that continued to haunt me ~18 months after its original deployment.
Tailoring satellite images – Krishna (thechaoticneuron)
What will you do when you find a satellite image broken down into a collage of PNGs? You should become a tailor and stitch these images together. Then you should become a cartographer and georeference this stitched image on the map. That is what I did in a project where we got access to 13,680 PNG images. Bonus—these images had watermarks as well, which I had to remove without destroying the information in the image. This cleaning work helped me learn a lot about image processing and GIS. I'm eager to share my learnings!
Churn Prediction: We thought it was hard but it was easy – Matthew Brandt (mattytwoshoes)
Back in 2017, it was great to work on a project for a B2B SaaS company, trying to predict customer churn - the holy grail of all problems! The goal was to save the company a huge amount of money by stopping customers from cancelling, using machine learning to identify the customers to reach out to. After 6 months of in-the-weeds data work, the solution ended up being something we built in less than 60 minutes...
I told you we got a lot of really really good submissions. Just wait until the rest come in!
Standing offer: If you created something and would like me to review or share it w/ the data community — just email me by replying to the newsletter emails.
Guest posts: If you’re interested in writing something, a data-related post to either show off work, share an experience, or want help coming up with a topic, please contact me. You don’t need any special credentials or credibility to do so.
"Data People Writing Stuff" webring: Welcomes anyone with a personal site/blog/newsletter/book/etc that is relevant to the data community.
About this newsletter
I’m Randy Au, Quantitative UX researcher, former data analyst, and general-purpose data and tech nerd. Counting Stuff is a weekly newsletter about the less-than-sexy aspects of data science, UX research and tech. With some excursions into other fun topics.
All photos/drawings used are taken/created by Randy unless otherwise credited.
- randyau.com — homepage, contact info, etc.
Supporting the newsletter
All Tuesday posts to Counting Stuff are always free. The newsletter is self hosted. Support from subscribers is what makes everything possible. If you love the content, consider doing any of the following ways to support the newsletter:
- Consider a paid subscription – the self-hosted server/email infra is 100% funded via subscriptions, get access to the subscriber's area in the top nav of the site too
- Send a one time tip (feel free to change the amount)
- Share posts you like with other people!
- Join the Approaching Significance Discord — where data folk hang out and can talk a bit about data, and a bit about everything else. Randy moderates the discord. We keep a chill vibe.
- Get merch! If shirts and stickers are more your style — There’s a survivorship bias shirt!