Does your app require low latency? Here’s some research that will show you if that’s possible or not.
This is good work to hold the telcos and cloud providers accountable. More please. Also, some way late proof of product manager lies is personally enjoyable to me.
There’s a difference between online and reality and it’s not quite what you think.
Good considerations when thinking about how influence and market education really work.
81,000 interviews. Stunning number on it’s own. Like it says on tin - how they use AI, what they dream it could make possible, and what they fear.
Gonna have to spend a little time with this one.
Real data on how GTM teams use AI
Benchmarks on how well LLMs search the web. Regularly updated.
Key for those of us building agents that search.
Pretty nice database of all cybersecurity investment and M&A deals for the last 5ish years.
Whoa. Nice.
Measured mobile app usage (downloads, daily and monthly active users, and consumer revenue —- No API usage, enterprise contracts, or desktop use) for 2.5 years through multiple model launches.
None of the big 3 one have “it” figured out yet and everyone is “winning” on different trajectories.
6000 execs said “meh” so far to AI driven productivity gains. BUT they have high hopes, so party on Wayne! Some nice exec vs worker bee contrasts in this research.
This is quite a quote, “This contrast implies a sizable gap in expectations, with senior executives predicting reductions in employment from AI and employees predicting net job creation.”
A look at conversion metrics across 200 SaaS companies.
Welp, I’m clearly going to have to level up the positioning agent.
Interesting study that argues
Large global developer survey into the kinds of AI apps developers are working on. While this is a teaser for their research, they argue 29% of developers are working on AI apps on edge devices like phones and IoT… which feels high AF to me, but maybe.
How many professional developers are there worldwide? Slashdata says 38.4 million.
How are marketeers using generative AI.
Looks like a good mix of content generation (meh), content repurposing (for sure), and fear of being replaced (low confidence in your own creativity or working for execs who already think marketing is “easy”) - also kudos to the designers on this. It’s a nice mix of data and external examples pulled from social media.
How do content and products get discovered? Search still rules, but fragmentation is upon us. Great data as always.
A surprisingly sober take on the role of AI in cooking our brains.
“Enterprise listening” company Momentum dove into 2,000 conversations with GTM teams to understand the technology issues helping and hindering them. The punchline? CRM and customer data are a mess, leading to bad experiences for prospects, GTM teams, and others. Also, AI use right now is a mess of indiviual and inconsistent use along with a pile of useless pilos.
I’ve lived this and been part of massive projects aimed at fixing it. Since then, the number of data sources and the criticality of understanding the GTM team reality has only increased by 10x.
Every marketer has seen click fraud, Lunio shed some product data on it and whoooeee, even if they are exaggerating a bit (I’m not saying they are, I haven’t even looked closely) the big greedy ad platforms should be ashamed of themselves.
Never underestimate the ability of a man to be blinded by incentives.
I didn’t have time to read this, but the first bit was hella wordy with not a lot of data or demographics. Sorry.
PwC’s 29th! CEO survey - 4,454 CEOs across 95 countries and territories. I likely missed this in the past because the track record of big consulting and surveys is shaky - and they likely spent their time on topics I didn’t care about, but here we are, they care about AI for shnizzle - 30% even claim revenue increases due to AI.
While this is clearly a pitch to help big companies figure out their AI, it’s a decent take on how CEOs feel about AI’s recent past and future.
A dive into the state of energy from all the different angles with, of course, lots of dissection of where AI infrastructure fits into the picture.
Highly recommend. This is a great way to get a good level set on the state of energy, climate change, geopolitics, and more.
A little different from my standard developer and IT infrastructure focused fare, but when my favorite social network dives into their data and does a survey of 30k users, I’m paying attention. How are folks working to keep their minds and bodies in shape? Strava has lots of answers to that ranging from everyone running more to what shoes are on their feet.
As with any product data focused study, how Strava users use strava can be all over the map and influenced by the growth in smart watches and such, although it’s nice to see them tackle some of that in their analysis. It wasn’t clear in the analysis what data came from the survey vs product data. On the other hand, they did a nice job of promoting a few app features alonside the analysis without going overboard.
Another, deeper than most, effort to understand developer AI adoption.
Mckinsey has their say on AI
Common trend on how execs view the impact of AI on employment: 32 percent expect decreases, 43 percent no change, and 13 percent increases. … TLDR: Nobody knows yet.
Is traditional search dead? AI search everything now? Inquiring minds want to know.
I haven’t read this yet, but past performance is a good indicator that this is good.
I don’t know Dataiku, but I dig their research. This report asks data leaders at $1B+ companies about AI Agents and contrasts many of the answers against a similar survey they did with CEOs back in the spring of ‘25. They also break this out by country which is unique and cool.
“59% have already experienced a business issue or crisis stemming from AI hallucinations or inaccuracies in the past year.” Yowzers. Also, I know it can cost a few extra bucks to have someone like Harris Polls do a survey like this, but at first glance, this is very well done.
What’s actually happening online? This report focuses on consumer behaviors, but I love it from a B2B perspective because there is so much overlap in human behaviors. It’s always so well done. They analyzed 912 million social posts - impressive in its own right, especially given how the number of platforms has expanded. Go read it, I guarentee you’ll learn something about the current psycology of humans online.
Funny story - authenticity and integrity get rewarded and they have the data to prove it.
Survey of sales reps. How can you not read this? Just gold nuggets everywhere.
Attainment is up, but leads are harder to come by. The old demand gen playbooks don’t work anymore.
A nice foray into the elements of SaaS success and failure.
You should read Kyle Polar’s newsletter if you don’t already. It’s fantastic.
If you think no one should care much about fonts, you are wrong and old. This survey put data from 12,000 respondents behind that.
A global survey of over 8,000 people working for companies with over 500 people on AI adoption trends finds it’s still early with only 13% of companies actually getting value from AI projects.
This study dropped with an AI assessment tool and a bunch of other assets. It’s a good example of how to do these things well. I thought they underplayed the global nature of the survey a bit, but then again I only gave it a quick skim.
Q: In an age when buyers are turning to AI what’s a sales rep to do? A: Be crazy knowledgeable and curious about a customers problem (duh)
Mercury data on startups spending on AI tools
Interesting, although I wouldn’t place any bets based on this data.
What spearates the best software teams from everyone else? DORA is back for another year with fresh insights into developer AI adoption and more.
ChatGPT usage data y’all. Consumer only.
Lotta homework. 10% of the world’s population.
A product list with a discussion of features and capabilities. Probably skewed a bit to show off FillApp, but overall a nice overview of agent types and what they do.
An analysis of code analysis data of 7.9 Billion lines of code generated by 970,000 developers using AI tools with investigations into code maintainability, security, reliability, and languages.
This looks very cool on the surface, alas I don’t have time to dig into it much.
By reviewing a collection of mostly non-SVB data sources, SVB puts their own spin on the startup and tech capital markets. With large-scale AI companies incinerating money at unprecedented rates, the rest of the startup market is running lean with low “graduation rates”, and VC LP money threatened by institutional endowment taxes, SVB says, “Yo, be careful out there.”
Good read with some non-obvious insights.
290 Cybersecurity Funding Rounds, 160 Acquisitions, but only 16% of deals are AI?? Feels low, gonna have to dig into that. Comments dispelling quick returns in security are interesting too.
Security is back baby. (for now)
Part 1 of the Pragmatic Engineer reader survey (Software engineering and development leadership and staff)
Must read for anyone selling into dev and eng teams building products.
16 experienced developers in “a randomized controlled trial (RCT) to understand how early-2025 AI tools affect the productivity of experienced open-source developers working on their own repositories. Surprisingly, we find that when developers use AI tools, they take 19% longer than without—AI makes them slower.”
I really like the way the authors discuss the limitations and caveats attached to studying the impact of AI on developer productivity. I need to spend som emore time with this study, but on top of the ‘how fast?’ question, I wonder if AI had any impact on the quality of the code or architectural decisions? (And if the tasks included more important software challenges than code generation)
Study into AI adoption and comfort level across numerous types of work.
POE - a chatbot that can access numerous models shares their userbase data so you can see which models are gaining and loosing popularity.
More of a popularity contest than anything else, but still interesting, espeically on the rise of reasoning models and the types of “brands” models have among POE users.
Carta on the
Full methodology description here: https://clickstream.cc/deep-dive-methodology-google-ai-overviews-ux-study/
The OG of security reports. Every year, this report provides a reality check on how breaches happen. Excellent as always.
Second quarterly (so far) survey of application developers using AI to develop apps.
Nice quotes sprinkled throughout. Findings tout a transition from AI just because we need to do something to finding actual use cases. Little to no demographic description (seriously, Vercel? I dig y’all, but that’s lame. I find it hard to believe the demographics of the Vercel community wouldn’t be a bit of a flex) The findings on AI specialization, leadership, and hiring ARE interesting for sure. OpenAI is still the top dog, although devs seem to be getting Gemini and Claude curious as well.
How well to execs understand and use AI and data inside their orgs? This survey takes a look.