Issue 309 / 376 10 signals + pod 19 source links Markdown

The weekly top 10 for B2B tech operators · Every Friday

Top 10 in Tech - What to know for Week ending January 31 2025

Friday 09:00 NZT Curated by Jon Davies
Top 10 in Tech - What to know for Week ending January 31, 2025

SaaS METRIC OF THE WEEK

NRR is one metric that has become a new modern gold standard for SaaS growth (see Pod of the Week below). So much so that ChartMogul has a trends and benchmark report: industry average NRR ranges from 90% to 100%, and the top-performing SaaS companies achieve NRRs 120%+. It's an action-packed NRR report, though - lots of fantastic insight.

EXPANSION

Expanding from above (get it???) - NRR and expansion strategies are pragmatic and capital-efficient growth practices that any good SaaS company needs but are getting harder to come by. Top quartile companies have been hit hard, seeing NRR drop from 119% to 107%. But at scale, more growth is coming from expansion vs. new. The proportion of ARR gained from expansion has increased from 28.8% in 2020 to 32.3% now. In comparison, the proportion of ARR gained from the new business has fallen from 62.0% to 57.9%.

TRIGGER TECHNIQUE.

This is a new entry for the Tech Dictionaries; The Trigger Technique outlined in this articleshows how to leverage customer interviews to learn what "triggers" potential customers' buying decisions. This information helps design more targeted marketing campaigns and boost sales.

AI part 1

Unless you were living under an AI rock this week, you may have noticed that a new AI Model, DeepSeek, stole the show, showing us all a few things: First mover advantage usually doesn't play out too well, constraints are a ripe spot for innovation (in this case US sanctions on Chips) and that AI commoditization was inevitable. It also ripped deep into the already established AI markets, giving Nvidia the most unwanted crown of the "biggest market loss In history" (about $600B). Yay to everyone's Stock Portfolio!

AI part 2

Fast follow to #4 above; BenedictEvans breaks down the real issue with AI models: it's not enough just to be better. As AI commoditization accelerates, differentiation shifts from model capabilities to distribution, UI, and product integration. The AI gold rush isn't just about who builds the best model—it's about who owns the customer (and why first-mover advantage is generally problematic)

AI part 3

This is a repost from a newsletter section late last year, but very relevant to #4 and #5 above this week. Every year, Benedict Evans goes on an absolute blinder in PowerPoint, exploring macro and strategic trends in the tech industry. Thisyear's version (across a 90-page slide deck) is all about AI (again, as last year's presentation was "AI, and everything else"). This year's post is about the post-hype investment surge where proven market value hasn't yet been firmly established (by that, I mean investment vs. Value Creation), which Deep Seek may be close to cracking. He wrote this in November and noted back then (which seems to be like 5 AI years ago) that scale is hard and REALLY expensive, which limits competitiveness (until last week). Slide 59 is the best TL;DR: The future can take a long time - unless US sanctions provide the ripe, constrained environment for quick innovation.

FRICTION LOGS

Another addition to your tech dictionary. Software and product design are never truly finished. Friction logs track every pain point in a product experience, helping teams pinpoint issues and refine UX. Want to set one up? Here's a solid guide on how to do it right.

Get the top 10, every Friday

Curated SaaS and tech insight from around the web, repackaged for people to put to good use — free.