Nate Sooter ⛳️
@natesooter.bsky.social
Ops and data nerd, husband, dad of 2 littles and avid golfer
It’s been a bit since I posted but I have a career update! I am now the Senior Manager of the RevOps Insights (Analytics) team at 1Password. I think I’ve found my niche. I’ve worked on GTM/RevOps teams for most of my career and am excited to step into a leadership role in GTM data 🤓
No, Facebook, I don’t want to know the “memories” of the cringey song lyrics I posted 15+ years ago
A tale as old as time… Me: “hey data eng team i need some fields added to our Salesforce tables in the warehouse for some GTM analysis” Them: …no Me: what’s the process to submit for approval? Them: (laughs)
The world if central data teams pivoted into building an internal data platform
Golf equipment manufacturers need to work on their data viz and storytelling …
T-25 minutes to this -- you're all registered, right?
dagster.io@dagster.io · 2y agoJoin us tomorrow where @alexnoonan.bsky.social, @matsonj.com, and @colton.boo will discuss: 🦆 Building an end-to-end data platform ingesting Bluesky data into MotherDuck, and PowerBI 👈 Shifting Left in Data Engineering 🧠 And AI-powered Data Engineering best practices lp.dagster.io/deep-dive-sh...
When life tells you to compare column types on dozens of tables between their Redshift instance and their Snowflake instance...you make some version of lemonade in Hex
Does anyone else yell at their phone “unsubscribe me” when they get an irrelevant ad?
When a GTM leader asks for a churn prediction model, they often don’t actually want a churn prediction model. They want to understand levers available to reduce churn…and prediction models don’t tell them that. So much wasted time rebuilding models that never say what the leader actually wants.
YAML is everywhere, even as the automod code on Reddit. What else did I expect.
Great read from Jacob on the benefits and current limitations of using an AI coding companion.
Jacob Matson@matsonj.com · 2y agohad a couple of opportunities to actually use an AI-enabled coding workflow, and I wrote about the experience.
Something I've consistently observed in analytics across multiple companies: The central data team doesn't 'get' sales analytics The GTM Ops team doesn't 'get' how to scale analytics Which means the CCO / SVP of GTM Ops / SVP of CS consistently express frustration about data
Happy Thanksgiving, especially to all the Export to CSV buttons out there
When I was little, I always told myself “when I grow up I want to provide shareholder value by making pretty Google slides”
Must follow starter pack for any other PNW weather enthusiasts
Ryan Voutilainen@yvrwxwatcher.bsky.social · 2y agoWhile I have shared the #BCstorm #StarterPack (also pinned to my profile) - we gotchu covered across the #PNW for our offshore #WeatherBomb xpctd to impact #CA, #BC, #OR, & #WA. Find all your #BCwx, #WAwx, #IDwx, #ORwx and even #CAwx peeps here for coverage this week & beyond: go.bsky.app/CA19msB
Back at it again (annual debate over success metrics for customer success)
🚨New Blog Post🚨 It's a near-guarantee that your CRM company data like Industry is messy. What if you could leverage the knowledge and synthesis of an LLM to clean it up? And do it directly in SQL in your data warehouse? I recently tried exactly that: www.dataduel.co/llms-in-sql-...
LLMs in SQL? A real-world application to clean up your CRM data – DataDuel.co
Executive Summary Firmographic data, like industry classification, employee count, and location, is notoriously messy and challenging to analyze. Modern databases offer a unique opportunity to clean a...
dataduel.co
Dropping a blog tomorrow morning about using an LLM in SQL to clean up CRM industry data 👀
Nate Sooter ⛳️@natesooter.bsky.social · 2y agoUsed it to fill in Industry data for some accounts in the CRM, worked much better than I anticipated. Need to write up a blog post about this!
Anyone out there working in RevOps / Sales Ops / Marketing Ops / CS Ops? Would love to put together a starter pack for us and haven’t seen one yet.
Went back today and bought this, it’s so cool. Circa early 1918 according to the serial number.
Nate Sooter ⛳️@natesooter.bsky.social · 2y agoIf you aren’t doing your analytics on one of these, you’re ngmi
If you aren’t doing your analytics on one of these, you’re ngmi
We ran 80,000 simulations tonight. Analytics Engineers are called Data Engineers in 40,012 of them. They’re called Data Analysts in the other 39,988.
Poll line images like this are madness. We should be making voting as easy as humanly possible for registered voters. A line is a failure. A line means we aren’t gather all the votes we can. It’s like giving a user a 20 field form to fill out before getting into your app.
Stephen Wolf@stephenwolf.bsky.social · 2y agoIt's great that turnout here seems high, but a long line is also a sign of voter suppression because Ohio's GOP limited early voting to just 1 site per county regardless of population. Franklin County's 1.3 million people (heavily Dem) & Vinton County's 12,000 people (heavily GOP) each get 1 site
Happy “my kids didn’t get the time change memo” day to all parents who celebrate
Using LLMs to fill in Industry for accounts with missing / bad data is …. Really effective?? Even using snowflake-arctic is getting reasonably accurate data directly in-line of my query
Metric design question: We have a small team formed to deal with renewals that are flat or downgrades to take time off sales/cs plates. We want to measure team efficacy, but all their metrics look horrible since they deal with the hand selected worst/riskiest renewals. How would you handle this?
