Andrew Williams

🔥 2-Min Survey: Help Fix the Insights Pipeline, Boost Your Next Review At Ara Platforms we're *obsessed* with the data -> insight -> action pipeline and removing roadblocks. On the data (insight producer) side, we want to make it easier to generate and publish useful insights. On the stakeholder (insight consumer) side, we want to make it easier to ask questions and get useful answers quickly. The process feels like a broken market—data people produce insights that don't get used while stakeholders don't can't get insights to important questions in a timely manner. If you're on either side of this equation I'd love to chat—please reach out via DM! If you can't it would mean a ton if you could fill out one of these quick Google forms. The end goal is to improve everyone's performance reviews Insight producers (data scientist, analysts, researchers): https://lnkd.in/emtMV6um Insight consumers (product managers, leadership, anyone who uses data to make decisions): https://lnkd.in/eEJ3Wse7

4

1 Comment

"We've previously invested in data clean rooms, but we’ve struggled with flexibly moving data in and out for experimentation. Our data scientists are eager to leverage new data sources frequently but the pace of compliance and the delays make new data stale by the time it gets into the clean room. We’re not sure how to get the flexibility we need for the data purchase ROI." This is a common challenge we hear from organizations leveraging privacy-sensitive, regulated data for commercial use cases. At Integral, we're on a mission to enable companies to safely leverage regulated data with unprecedented efficiency and maximize the value of downstream solutions like clean rooms or models. Here's how we approach these challenges: 1. Streamline Compliance: Our real-time compliance process significantly reduces time to value for deploying and experimenting with regulated consumer data. 2. Maximize Data Utility: While traditional methods often compromise data, our solution preserves up to 98.8% of data utility - richer, more valuable datasets. 3. Enhance Collaboration: We power seamless secure data movement into and out of clean rooms, enabling necessary collaboration with internal teams and external partners. 4. Ensure Continuous Compliance: Our platform provides ongoing compliance assessments of data flows, detecting weaknesses and recommending enhancements as data is updated or refreshed. 5. Provide End-to-End Support: From executing necessary data remediations to delivering compliant data quickly, we offer a comprehensive solution supported by our best-in-class clean room partners. The result? Organizations can quickly purchase regulated data, deploy it, experiment, and innovate all while maintaining privacy and compliance. Are you looking to maximize the potential of your regulated consumer data while navigating the complexities of compliance? Let's connect and explore how Integral can transform your clean room strategy, enabling unprecedented data collaboration without sacrificing privacy. https://lnkd.in/e8xgbxap #DataCleanRooms #RegulatedData #DataPrivacy

26

1 Comment

Complexity tackled with simplicity = Product + Design + Engineering WIN See how Mixpanel is committed to the concept of "lego for analytics" to add powerful reporting but with simplicity. Preview this concept in our recent shipped stuff. Happy testing! - 1. More powerful funnels: https://lnkd.in/gGUq2bZK 2. Revenue-weighted funnels: https://lnkd.in/g7JBer3q 3. Mix and match metrics in a chart: https://lnkd.in/gmXxgSj4 4. Standardized metric and behavior definitions: https://lnkd.in/g9Z7yucD 5. Behavioral segmentation to find "aha" moments: https://lnkd.in/gr-eFh7M

3

If you're connected with me, you know I scream about how necessary integrations are. Check out this piece where I explain more for insideBIGDATA!

21

There’s a new wave of data buyers in the market: the "Corporates"... also known as every company that's not a hedge fund... CPG companies, Marketing agencies, Biotech, Tech firms, and more - And that means that data providers need to adjust their sales approaches accordingly. The OG data buyers are investment firms, and they don’t need as much personalization with their data. That’s why they hire a ton of analysts and they set aside lots of resources: because their whole revenue model is based on using data to make investments. So getting a ton of data - in whatever format - is as good a start as any for them. But the newcomers to the data-buying party A) don’t have the same revenue model that hedge funds do and B) therefore don’t have the same pool of resources to draw from to sift through all this data. They’re too busy manufacturing products, or developing drugs, or figuring out their supply chain. So if data providers want to sell effectively to the flood of new corporates in their market, they need to get onboard with personalization - both in content and delivery - to meet their new customers where they are. In other words, if you’re selling data to a hedge fund, you can throw the whole kitchen sink at them (within reason). But if you want to sell to the new wave of buyers, you’ll have to choose your ingredients more carefully.

24

3 Comments

I think people are missing the arbitrage opportunity for ChatGPT Pro tier: To create "human wrappers" around ChatGPT. ChatGPT Pro costs $200/month and offers unlimited o1 Pro (near superhuman intelligence) and unlimited advanced voice mode (unlimited conversational AI). Yes, $200/month is beyond what many companies might pay per user, at least initially, because they might not have sufficient need. But those companies might pay a consultant $20,000 for a valuable task that the consultant could use ChatGPT Pro to complete in less than a day.

20

14 Comments

I’m betting (literally, in a Manifold prediction market) that we’ll see GPT-5 (or whatever it’s called) in November of this year. I might be wrong, but here’s my thinking: - Prediction markets have been trending to Q4 2024 or Q1 2025, incorporating various signs and signals. Some people in these markets are likely insiders who have non-public information. - OpenAI seems to be shoring up weak spots and addressing criticisms. For example, they recently appointed a former NSA head to the board to address security criticisms. I don’t think they’ll release a much more powerful model before improving public trust. - They need to get more compute in place. Hence their recent deal with Oracle about this. - They held their developer day in November last year, and will likely do the same this year. That seems like the right place to make a big announcement. - They still have to roll out the new GPT-4o voice mode. At best, we’ll get an alpha in the next few weeks. - They’re likely going to soon roll out some kind of search competitor to Perplexity and Google’s AI Overviews. So, I think what’s going to happen, is that OpenAI will maintain its dominance on performance by releasing incrementally better models only when forced by competitors to retain its top position on leaderboards, but reserve its biggest next model release, or an announcement of it, for November. I’ve bet on Manifold accordingly, putting my (virtual) money where my mouth is, on a release prior to December 2024. Though I didn’t make a huge bet, because I still have a lot of uncertainty here.

7

1 Comment

Excellent study from Damian Rollison and the SOCi, Inc. team which seems to confirm my longtime hypothesis around citations (https://lnkd.in/gxZQwdx3): the only citations that are valuable are the ones that rank in Google for the keywords you want to be known for. https://lnkd.in/g49DMSeN

38

11 Comments

📺 📺 📺 75% of in-store retail media will be deployed in the next 18 months!!!! This Grocery Doppio research paper is a wake up call to grocery, big box and large #retail chains. The time is now to get to moving or be left behind on one of the greatest profit drivers for brick-and-mortar retail. Some takeaways: --- Retail media is good for shoppers, brands and retailers --- 🛍 There is a 38% increase in shopper engagement with in-store media with 61% of shoppers saying the media content was useful! 💚 27% increase in impulse buys at grocers who enable in-store media 🛒 97% of grocers intend to adopt white-label or thirds party rather than build 📊 93% of #CPG brands want to to tie digital and in-store engagement. 📈 The number of #retailmedia networks will double in next 18 months --- Challenges to scale --- 👩‍💼 79% of retailers say they don't have the talent to scale their #retailmedia biz 💰 Only 3% of CPGs have significant additional budget for grocery media in 2024, and 89% want greater data accuracy for that market spend. Download the report here --> https://lnkd.in/ejJQmeSu Thanks for sending my way Lisa Goller, MBA

28

14 Comments

Data teams are unique. Often invisible until shit hits the fan, they are unloved and unappreciated—creating a challenge for the leaders wanting to empower them. If you're nodding your head in agreement, you may get something out of C.J. Jameson's talk from this year's Data Council. C.J., Engineering Manager at Monte Carlo Data, has created a step-by-step playbook for data team managers looking to better support their teams. Here are my notes: 1️⃣ Check your pebbles C.J. discussed the classic "rocks, pebbles and sand" metaphor for prioritization. Most data engineers obsess over the big rocks—like re-platforming and quality initiatives. But C.J. made a compelling case for not overlooking the pebbles—those pesky ad-hoc requests that eat up our time. Pebbles can point to what stakeholders genuinely need and where systems are falling short. Exceptionally great managers and engineers find ways to tackle the small stuff while still delivering on the big initiatives. They're automating repetitive tasks, connecting dots between small requests to drive larger improvements—the "pebble superstars." 2️⃣ Say what you mean There's a shortage of data PMs out there. Technical leaders often need to step up and articulate exactly what needs to be built. C.J.'s point resonated with me—don't be shy about sharing your vision for how things could work better. We need to advocate for the right tools and approaches. 3️⃣ Find your “want” This one hit home. To be an authentic leader, you've got to tap into what drives you. What problems get you fired up? When you're clear on this it becomes contagious and your team picks up on that energy. Thanks for a great talk, C.J.! The full talk is linked in the comments.

19

2 Comments

Being the first analytics tool (that I know of) to offer up-to-date benchmark data (from 70+ different software tools) built right into the platform, we had to come up with a way to show benchmarks on our dashboards (that made sense to our users.) Feedback very appreciated... Below is the visual. Here’s the math and logic that goes into creating that visual. 1️⃣ First, we calculate benchmarks for the cohort that the user selects (company size, industry, etc) based on the sample of companies in that cohort. The benchmark shown on a line chart like the one below that shows "This Month's" data is calculated from the previous month's data, since it is impossible to calculate a benchmark for a "this month" time range, given that the month isn't over. 2️⃣ To visualize a monthly benchmark on a daily line chart, we divide the previous month’s benchmark by the number of days in the current month. This gives us a “daily average benchmark” line. 3️⃣ To help someone compare their daily performance to the "daily average benchmark line", we added a "median" line option for the current month's performance. Why? Most metrics visualized over a 30 day period fluctuate quite a bit each day. As you can see in this chart, the value for sessions varies a lot on a daily basis .So, to get a good comparison, it's best to compare the median value for the company’s monthly sessions to the 'daily average benchmark'. 4️⃣ What's the takeaway? This company is outperforming the benchmark. As you can see by the daily values of the current month's performance, it's pretty much always higher than the benchmark line for all but a few days in the beginning of the month. So, you’d expect the median value to be above the daily average benchmark too, as it is. In this case, it might makes sense for the company to compare themselves to a cohort of bigger, more successful companies, if they want to challenge themselves. Or just take the win and keep trucking. What do you think of our logic and how we're displaying benchmarks on line charts? PS. Credit to the team for coming up with this: Katja Pozeb, Mateja Verlic Bruncic, Gasper Vidovic among others, I'm sure. PPS. If you'd like to try this for yourself, here's more about how it works: https://lnkd.in/eRN-xnEB

37

11 Comments

Market research is evolving. The conversations paint a clear picture...traditional web monitoring solutions are falling short. High volumes of irrelevant data. Manual data verification slowing things down. Missing real-time sentiment shifts. Fragmented insights from scattered sources. These challenges are costing researchers time, accuracy, and opportunities. But there's another way. Imagine a world where AI filters out the noise, leaving you with only relevant data. Where data verification is automated, freeing you to focus on analysis. Where sentiment shifts are tracked instantly, informing your strategy as it happens. Where fragmented data sources come together, giving you a complete picture. This isn't just imagination. It's happening now. AI-powered open-source intelligence is transforming market research. It's not about doing something different. It's about doing something that matters to researchers. Faster insights. Cleaner data. Real-time intelligence. The future of B2B market research is here.

5

1 Comment

Interesting findings from DocSend, which looked at successful and failed Seed and Pre-Seed pitch decks. "Successful seed decks in 2023 rearranged some of the classic building blocks: instead of foregrounding problem/solution sections, they often inserted the why now? section after an opening company purpose slide. They also tended to place their product and business model sections earlier slightly earlier in the narratives than unsuccessful decks." So timing, or "why now," was judged as even more important... If you want to learn more about why now, I have a book you will like. Links to the DocSend and the book in the comments.

18

5 Comments

Announcing our new feature: Clean Up Your Documents with RaijinAI’s Intelligent Formatting! 🌟 We know how much y’all love nice, clean transcripts😎. At Raijin.ai - AI Analysis Assistant, we are constantly pushing ourselves to make the best experience for anyone who wants to do quick + good research and insight mining. We can now automatically format documents and web links that you upload into our platform without any additional prompts or instructions from you. What this means is that you get enhanced readability without the hassle of trying to split your notes or web content into paragraphs manually. The example here is from the recent Fortune magazine interview with NVIDIA's legendary CEO Jensen Huang. This was done as simply as copying and pasting the URL link into our platform. Come try this out by signing up for our Beta Launch (link in comments section)! #AI #generativeAI #LLM #userresearch #discovery #productivitytools #insights #UIUX #productmanagement #productdevelopment #voiceofcustomer

14

1 Comment

There’s a lot of talk about how to integrate media mix modeling and lift tests, but how specifically can they work together? Here’s 3 use cases our customers find valuable: 1) - Use lift tests to calibrate your MMM 2) - Use lift tests to proactively validate your MMM 3) - Use MMM to get additional insight into incrementality Now, I know these are complex use cases that are often followed up with plenty of questions: - How should you incorporate the results from your lift tests when the vast majority of MMMs make the assumption that marketing performance doesn't change over time? - When an MMM and a lift test disagree, which is wrong more often? - What are the drawbacks of lift tests and how should you use an MMM to compensate for them, and vice versa? - How should the MMM pull in all these different pieces of evidence into one single view? Michael Kaminsky and I are putting together a live Office Hours session for marketing-focused data scientists who are interested in MMMs and lift tests. We’ll talk about how to tactically implement each use case, and we’ll open it up for questions and go through any measurement challenges you’re dealing with. If you’re a marketing-focused data scientist, I think you’ll find this session valuable and I’d love to see you there – I’ll put the link to register in the comments.

21

3 Comments

👉 Analytics and Product trailblazers tell their real-life stories of insights that led to massive growth. Shoutout to Lenny Rachitsky / Lenny's Newsletter, Phil Carter, Maria Pergolino, Casey Winters, Japna Sethi, Ken Rudin, Darius Contractor, Jenna Lemonias, Andreas Gross, Elena Dyachkova for contributing 🌟 - your insights are driving innovation and shaping the future of data analytics. 🙏 I’m thrilled to share our collection of "Outta Sight" Insights from these leaders. 🚀. We asked our partners to share their geeky, cocktail party-worthy stories 🍸 - incredible journeys and impactful insights! The stories they shared are incredible. 🍻 🔥These trailblazers, who helped build Airbnb, Google, Grubhub, Quizlet, Zynga, Adobe Marketo, Dropbox, Otter.ai, Life360, Peloton Interactive, The Atlantic, Loops, and other companies, have seen firsthand the transformative impact that robust insights can deliver. Enjoy reading about their bold, analytics-driven moves in this Outta Sight Insights collection. 🔥 Here’s why we built this collection: At Loops, we know there's too much "data theater" out there, distracting from the robust insights that truly drive growth. 🎭 We wanted to inspire data analysts and product growth teams by showcasing these powerful wins. It’s easy to forget the thrill of a big win, especially when grappling with daily growth challenges. These stories remind us of the profound impact that data-driven decisions can have on a company's trajectory.🚀 👉 Together, let's move beyond data theater and focus on what really matters: driving actionable insights that fuel success! 💡📈 https://lnkd.in/d4QsxKgf #DataDriven #ProductAnalytics #GrowthInnovation #CausalML #DataDrivenDecisions

45

4 Comments

Don't let your CRO strategies fall short! If your CRO efforts aren't delivering results, you might be overlooking a critical element: qualitative research. Quantitative data shows you what users are doing, but qualitative insights: ✅ Uncover user motivations & pain points ✅ Identifies root causes of behavior ✅ Guides impactful optimizations Start combining qualitative research techniques (user interviews + usability testing) with your analytics data. It allows you to deeply understand your audience, prioritize high-impact changes, and create user-centric, conversion-driving experiences. How will you implement qualitative research to drive real results?

2

If you’re doing multivariate testing on messaging like email/text/push remember that day/time of send is likely the most impactful variable. So if you send version A at time X and version B at time Y, you’re doing more XY testing than AB testing.

2

Considering vertical segmentation in 2025? Here's why you shouldn’t: 1️⃣ Messy Unequal TAMs and SOMs lead to uneven comp plans and convoluted territory carves. Every horizontal org considering verticals has a couple core verticals where they're mature and some new verticals they want to break into. Those require different investments and have different outcomes. The compromises pile up. 2️⃣ Risky We tend to only see the upside of specialized seller knowledge. But there’s a downside. Your expert seller with the network walks out the door? That vertical’s dead for the next 6-12 months. 3️⃣ Fragmented True verticalization demands unique GTM efforts. Are you ready for distinct ICPs, personas, demand gen, product marketing, case studies, enablement and sales processes? Most orgs aren’t—they just want a rep to “speak the language”.[1] I met a company yesterday where verticalization is a good fit. Half century old, big customer base, super specialized products and a will to put enough structure across the GTM org to see it all the way through. You probably aren’t like that. Unless you have crystal-clear vertical-specific sales differences and the structure to support them, just don’t do it. Instead, focus on “light specialization”—vertical overlays, tailored case studies, vertical reference customers and focused team-wide plays. You’ll always benefit from being hyper-relevant to your prospects, but don’t let it dictate your entire GTM motion. -- [1] One reason I like Vertical SaaS is that all of this tension dissolves. You get the benefits of specialization with just one core GTM motion. More on that in the next issue of the Uncharted Territory newsletter.

19

4 Comments