Doug Camplejohn
(00:01)
Hello everyone. This is Doug Camplejohn and welcome to this week's episode of Revenue Renegades. This week we welcome Marc Andrusko, otherwise known as the other Marc A from Andreessen Horowitz, to the show. Marc, welcome.
Marc Andrusko
(00:14)
Thank you so much for having me. And yeah, I feel legally obligated to make that joke every time I meet someone in tech or venture.
Doug Camplejohn
(00:19)
Does the other Marc laugh at it now or is it just old hat?
Marc Andrusko
(00:25)
I think so. I remember my first week when I started, I got an email with a bunch of fund performance data, and I was like, do all new hires get this or did someone mess up my email address? Mine is unfortunately probably a lot more depressing than his, but that’s okay.
Doug Camplejohn
(00:36)
Here’s your bank statement, with a few zeros. That’s awesome.
Well, for anybody in the world who hasn’t heard of Andreessen Horowitz by this point—one of the most famous VC firms in the world—can you give us a little background on what the firm is about, what sets it apart, and what you focus on there?
Marc Andrusko
(01:03)
Yes. Andreessen Horowitz was originally started by Marc Andreessen and Ben Horowitz. Big presence in San Francisco and Menlo Park, classic Silicon Valley, but I work out of our New York office in Soho.
We’re looking to back generational companies across categories, stages, and sizes. We’ve been purpose-built to do any round at any time.
I’m an investor in our billion-dollar-plus AI apps fund, largely focused on the application layer. Colleagues of mine focus on infrastructure, American dynamism, crypto, bio and healthcare, gaming, and more.
What we’re probably most known for is introducing the platform model. Traditionally, VCs would back a technical founder zero-to-one, then parachute in a professional manager to run the company. Marc and Ben, having been entrepreneurs themselves, thought that was wrong. They believed founders are the best people to run their companies indefinitely. The firm was created to support them in perpetuity.
Now “platform VC” is common, but the scale of A16Z’s platform today is unmatched. We employ about 600 people, of which only 75 are on the investing side. That means 500+ employees are purely in service of portfolio companies. That gives founders huge leverage.
Doug Camplejohn
(03:19)
It is amazing. I remember when Andreessen Horowitz hit the scene, the notion of it’s not just a check, it’s infrastructure and services—that was revolutionary. Now other firms have copied it, but no one does it at the same scale.
Marc Andrusko
(03:45)
Exactly. If you try to do “platform” small, you end up with generalists juggling too much—helping companies sell, recruiting execs, recruiting engineers—without doing any one thing excellently. At our scale, we can have true domain specialists, which is a differentiator.
As an investor, you’re still the quarterback for the founder relationship. The platform just gives you steroids, not a replacement.
Doug Camplejohn
(04:40)
Another big innovation was being a multi-stage investor. Traditionally, firms were seed only, or Series A only, or late stage only. A16Z said: let’s invest at any stage, so we can back the winners, whenever we find them.
Marc Andrusko
(05:13)
Right. Venture is a power-law business. A small handful of companies generate the vast bulk of returns. You want the flexibility to catch those companies at seed or at later stage. Sometimes it’s clear early on, sometimes not until much later.
I focus mostly between seed and Series B, but friends in dedicated pre-seed or growth funds sometimes have it easier—you either get to conviction now, or you miss it forever. At A16Z we can keep tracking, which is both a blessing and a curse.
Doug Camplejohn
(08:08)
You and I met because of your article “Death of a Salesforce.” Catchy title. What motivated you to write that?
Marc Andrusko
(08:31)
I love a pun, so the title was meant to grab attention—Salesforce will be fine for a long time.
But I got interested because of my time at Plaid, where I worked on the Growth team. One of my projects was around NPS, customer satisfaction, and feeding that back into product roadmaps. Coordinating all the feedback across Qualtrics surveys, Salesforce, Zendesk, Slack, Reddit, Twitter, GitHub—it was a nightmare.
It made me realize how static, siloed, and retroactive CRM workflows feel. After spending a 12-hour day selling, the last thing you want to do is spend another hour updating data in Salesforce. That pain stuck with me.
Doug Camplejohn
(12:53)
Totally. CRMs don’t die. ACT! is still around. Siebel lasted years. Salesforce and HubSpot are dominant. But some argue that this AI cycle will be different—that incumbents will outlast the upstarts. Where do you land?
Marc Andrusko
(13:36)
Incumbents are sticky. Think about this: nearly 50% of IT workloads are still on-prem. That’s the staying power of systems of record.
But I think startups still have room:
- If you can capture early-stage companies (YC cohorts, for instance) and grow with them.
- Or if you can wedge into the underserved mid-market segment of HubSpot/Salesforce where customers overpay for features they don’t use.
CRM is a $90B+ category, growing 6% a year, with “shadow spend” in Excel, Notion, Airtable, etc. Huge market, lots of room.
Doug Camplejohn
(17:51)
Exactly. Salesforce came under Siebel, HubSpot came under Salesforce, and now others are coming under HubSpot. But a rip-and-replace doesn’t happen overnight—it might happen piece by piece (e.g., SDRs move first, then AEs, then CSMs).
Marc Andrusko
(19:02)
Right. And we’ve seen the wedge-product playbook: automate a messy workflow, sync it into Salesforce/Epic/NetSuite/etc., then over time own more of the workflow. That’s a backdoor into becoming the system of record.
We’re also now seeing founders start with the ambition to be the system of record on day one, not just a bolt-on. That’s a harder path, but some are going for it.
Doug Camplejohn
(21:28)
Exactly. If you didn’t build as a system of record from day one, it’s very tough to retrofit later. Just look at Gong, Clari, Outreach—they hit $100–$200M ARR and only then tried to expand beyond their wedge. Gong did well, others less so.
Rippling was smart—they started with system of record + integrations. And Salesforce’s real moat isn’t just CRM, it’s the AppExchange ecosystem. Recreating workflows and integrations is a huge lift.
Marc Andrusko
(23:19)
Yes, and ecosystem posture will matter a lot. Some incumbents are restricting API access to keep out startups, but that risks alienating customers. Look at what happened in banking with Plaid: when banks blocked integrations to Venmo/Robinhood, users blamed their banks, not Venmo.
If Salesforce or others do too much API-gating, customers will revolt.
Doug Camplejohn
(26:33)
Exactly. I spoke with Tamar Yehoshua (ex-CPO of Slack, now at Glean, also an advisor to Coffee), and she noted Salesforce has always been generous with CRM APIs, but with Slack we’re seeing some restrictions. This will be an interesting test of how far they’ll take it.
Marc Andrusko
(27:18)
Exactly. And meanwhile there’s a Palantir-like theme emerging: forward-deployed engineers solving workflows company by company. When costs are low enough, that works as a scalable business model. Klarna famously “churned” off Salesforce/Workday in favor of custom AI + homegrown systems—though they eventually walked it back.
Point is: customers have options, incumbents can’t overplay their hand.
Doug Camplejohn
(30:29)
Customization limits are evaporating quickly. Instead of “buy another app or live with vendor defaults,” companies now want toolkits they can adapt. Which brings us to silos—sales, marketing, CS. Rippling collapsed silos internally for HR, Finance, IT. Do you see GTM silos collapsing too?
Marc Andrusko
(31:29)
Absolutely. I think current GTM org design is not customer-centric. Marketing, Sales, Partnerships, CS fight over attribution instead of focusing on customer experience.
Tools mirror this silos problem. A typical growth-stage GTM stack is 20+ tools costing $2.5 million annually. That fragmentation is insane.
AI-native CRMs like Coffee aim to consolidate workflows across the lifecycle. If software can track an account seamlessly end-to-end, the org structure should evolve to match that.
Doug Camplejohn
(34:52)
Yes! The walls between functions are ridiculous. I’d get multiple cold calls from Salesforce AEs with zero context of who I am or what my company is doing—despite all that data sitting somewhere inside Salesforce. Same thing with AE to CSM handoffs: verbal dumps, no structured record.
Marc Andrusko
(36:07)
Exactly. The silos are artificial, and ripe for change.
Doug Camplejohn
(36:09)
Let’s pivot to pricing. People say the per-seat SaaS model is dead. Everyone is talking usage or outcome-based pricing. What are you seeing?
Marc Andrusko
(37:07)
It’s early days, but outcome-based pricing makes sense. Right now most AI apps sit in the “labor replacement” bucket: replacing human tasks at a fraction of cost. Pricing is something like: human cost = $100K, app does it for $20K × N. That works.
Where it gets exciting is “found money”—producing revenue that wouldn’t have existed otherwise. If Coffee books you a net-new customer, why wouldn’t you pay a rev-share? That’s powerful but only possible once the tech is bulletproof.
Legal tech is instructive too: my portfolio company Eve helps plaintiff litigators handle 5–10× more cases. Since those lawyers earn contingency fees, Eve directly grows their revenue, which justifies far higher willingness to pay.
Doug Camplejohn
(42:11)
Great example. Now let’s touch on benchmarks. Some dev tool AI startups are hitting crazy ARR very early. Have benchmarks shifted?
Marc Andrusko
(42:51)
Yes. Things are working faster than ever—founders can build and sell faster. Boards are pushing CEOs to “buy the AI,” so procurement cycles collapse. Six-person teams are getting seven-figure ACVs within months.
That acceleration is real, not a bubble. But it also means investors have to update their definitions of “good, great, excellent traction”—while still doing the qualitative diligence on people and market.
Doug Camplejohn
(46:20)
If startups can hit such numbers fast, what happens to late-stage VC?
Marc Andrusko
(46:38)
We’ll still see winners raise mega-rounds. Venture is a power law, and preferential resource attachment is real: raise from top VCs, hire the best people, win the market. Not every founder wants that path, but ambitious ones still will.
Doug Camplejohn
(49:20)
Very true. Trillion-dollar outcomes are the new aspiration.
Marc Andrusko
(49:31)
Exactly. The ambition level is unprecedented. The other week Meta and Microsoft added nearly half a trillion in market cap in one day just on AI earnings commentary—that shows the scale of belief.
Doug Camplejohn
(50:36)
Love it. Let’s close with a personal lightning round. Tell listeners something surprising about you.
Marc Andrusko
(50:47)
I got a black belt in karate at age 11. People don’t expect that. Luckily I haven’t had to deploy it!
Doug Camplejohn
(51:06)
Ha! What do you do for fun outside of work?
Marc Andrusko
(51:23)
Mostly hang with my wife, toddler son, and our big rescue dog. I also still play competitive tennis when I can. Always open to hitting with folks in tech.
Doug Camplejohn
(51:53)
So not full pickleball convert yet?
Marc Andrusko
(52:02)
I dabble when visiting my parents in Florida, where it’s everywhere!
Doug Camplejohn
(52:16)
Nice. Okay, last one: favorite personal product?
Marc Andrusko
(52:35)
Eight Sleep. I run hot, my wife runs cold. It literally saved our nights. I keep it at -9, she’s at +2. Now I can’t sleep without it.
Doug Camplejohn
(54:21)
Love it. Same here. Okay, Marc, how can listeners stay in touch?
Marc Andrusko
(54:31)
I share thoughts on software and AI regularly, either on the Andreessen website, or on Twitter and LinkedIn. On Twitter I’m @mandrusko1.
Doug Camplejohn
(54:58)
Perfect. Marc, thank you so much—this was a fantastic conversation.
Marc Andrusko
(55:03)
Thank you, Doug. My pleasure.
