The weekly top 10 for B2B tech operators · Every Friday
The 'payback' period is the nuance of why we measure CAC. How long until we break even? Benchmark-wise, the negative trough is way longer than you think, so take a seat! New B2B customers, on average, take 2 years and 2 months to become profitable. This really highlights a deepening dependency on access to capital to fund a SaaS company's growth through these SaaS Cash Flow Trough. BONUS: Here are last week's CAC Payback benchmarks.
According to this Reforge article, marketers often don't see the expected big returns from testing because they avoid major risks. Making bigger bets with strong business cases can lead to transformational success - it has some great IRL example bets from Groupon and Google, and there is also a "Big Bet Calculator" embedded in the article for you to use.
Here is Mark Suster's series on his Medium Blog covering StartUp Boards. With a follow-up article that shows a board structure by stage! He also provides a blog post AND a 43-slide deck.
The tech-ism cycle is generally always boom, bubble, bust, boom again (but a little more chill). Is AI going to be any different? Doubt it. Another good tech-ism is that we always overestimate short-term impact and underestimate long-term transformation. Crazy Stupid Tech breaks this down: hype comes first, usefulness comes later, and the real returns go to whoever survives the whipsawing.
Fast follow-up from above, and this was discussed a few weeks back - Stratechery now also makes the argument that, when it comes to AI, the real moats aren't models - they're distribution and compute control. As of today, Google wins with integration, Nvidia wins with infrastructure, and OpenAI wins with velocity (for now). The takeaway for startups: your moat won't be "better AI," - side quest - here is a list of AI Startups that have raised $100m plus this last year.
Hey - we all have it. Everyone knows they'll have to pay down tech debt sooner or later - Hyperact takes a Product perspective and claims that tech debt is a product choice, not an engineering mistake. Worth the read, as we all need to consider tech debt further up the decision tree.
I got this report from Stripe - Vertical SaaS is evolving and the category is maturing fast: 70% of companies now sell more than one product, fintech is the second product for nearly half of them, and AI adoption sits just over 50%.
Gartner says global IT spend will break $6T in 2026, with enterprise software jumping 15.2%. But here's the kicker: most of that growth isn't new logos or usage - it's price hikes and AI add-ons. Budgets are rising, but tolerance probably isn't, so if you're selling into the enterprise, expect harder ROI scrutiny (even if those wallets technically "grow.").
The end-of-year planning season is here, and SaaStr has a surprisingly useful AI benchmarking tool that builds a C60-style financial plan in seconds. Revenue, burn, CAC, payback, hiring - all benchmarked instantly against thousands of SaaS peers.
Super interesting article updating the classic startup playbook. MVP → PMF → scale is old school - unlocking hidden loops, data flywheels, distribution hacks, and agent-driven workflows are all part of the new playbook, along with design roles around outcomes and what it takes to be a leader in this new company design.
Get the top 10, every Friday
Curated SaaS and tech insight from around the web, repackaged for people to put to good use — free.