AI Sales Tools Have a LinkedIn Problem
Scraper wars, trade wars and finding inspiration in IAM.
I’m writing this as I head to SFO for a few days of meetings and events in the Bay Area. I’ve lived in Austin since 2018, but before that I spent 8 years in San Francisco. For me personally, San Francisco is a strange combination of things: the birth of our daughter, beautiful weekend runs to the Golden Gate bridge and Warriors championships1—all set against the backdrop of the most grueling startup years I’ve ever experienced. I’ve actually only been back once since then so I’m eager to see how the city continues to evolve.
One obvious change: the Bay Area is much more spread out than it was pre-2018. My meetings are taking me places I never used to go. Back in the day it was unlikely you’d find people in the software industry living north of Marin or well east of Oakland. Not so much now. (The Peninsula and the South Bay, of course, have always been their own thing.) Clearly COVID drove SF denizens all across the Bay in addition to points east like Austin and Miami.
Ok, enough with the San Francisco rumination. This isn’t a recipe site2 where I share my life story before getting to the reason you’re actually reading.
This week, I thought it would be a good time to share a few things I’ve been paying attention to lately. They’re relevant to CROs, but may not warrant the full article treatment just yet.
Let’s dig in.
What CROs should know about LinkedIn data
For many CROs, data is a RevOps problem—its RevOps’ job to make sure you’ve got phone numbers and firmographics. That may be true, but a sales team without prospect data doesn’t function. What’s also true is that many of today’s fancy automated AI GTM plays rely on signals from LinkedIn to trigger outreach and tailor emails.3
Guess what? All of that is in flux as of last week. And much of your outbound pipeline might be affected.
Last week, LinkedIn summarily terminated the company pages for well-known sales intelligence vendors Apollo and Seamless.AI. Word on the street is that it’s the result of them scraping LinkedIn. It’s long been an open secret that these (and other companies in the space) are scraping LinkedIn on a massive scale.
LinkedIn has a long history of fighting against scrapers with policy (their terms of service explicitly forbid it), technology (they have automated systems in place), and legal action.
The most consequential legal action involved a little-known company called hiQ Labs. I’m not a lawyer, but here’s how I understand it. In 2017, LinkedIn sent them a cease and desist for a massive scraping operation that had started as early as 2012. hiQ sued, everyone made lawyerly arguments4, the Supreme Court got involved and they settled in 2022.
The reason this matters is it established that, while scraping public LinkedIn content isn’t against the law, it’s a violation of contract (LinkedIn’s user agreement). Since then, most of these companies have been operating in a gray area of not doing anything illegal but being subject to LinkedIn’s enforcement whims.
This all comes against the backdrop of larger controversies around the ethics and business models of AI and scraping. Large language models from OpenAI and others were built on massive amounts of data scraped from the web. Some organizations are fighting this (e.g. the New York Times) while others (e.g. Reddit) are licensing the data to make bank. These folks were fine with Google and Bing because they brought them traffic they could monetize. Not so much with companies sucking down their data for free and turning it into commercial products.
This reminds me of my days in social media analytics. From 2010 to 2018 I ran a social listening and analytics company (Union Metrics, now part of Brandwatch which is itself part of Cision—it’s a long story). What began as a wild west of web scraping and free APIs eventually evolved into a semi-lucrative data licensing businesses for social networks. LinkedIn, however, never really participated in any of this.
Over time, data breaches, reputational risk and questionable economics led to those lines of business being drastically curtailed. The number of products with reliable access to social data dwindled, competition went down and innovation dried up.
So, why does this matter to CROs?
Your team is probably using tools right now that scrape LinkedIn data and they’re probably running plays that require it. While LinkedIn almost certainly isn’t going to sue you, they may shut down your vendor. When you’re purchasing that fancy AI tool, make sure you get a straight answer about where their data comes from and what risk there is if LinkedIn comes after them. You and your ops team should have a plan for what to do if that data suddenly disappears so your pipeline doesn’t disappear with it.
Trade wars come for software
Like many of you, I’ve been avoiding looking at my portfolio recently. The S&P 500 is flirting with correction territory; the Nasdaq’s already there. There may be a handful of contributing factors, but the biggest one sure seems to be that we’re in a very unpredictable trade war with countries we’re all used to doing predictable business with. While most tariffs are on physical goods (e.g. aluminum), the uncertainty and anger are having an impact on software sales as well.
I spoke to a B2B tech CEO I know who recently lost a deal with a Canadian company. The prospect’s explicit reason was a preference to “buy Canadian” because of “unjustified tariffs”. This CEO isn’t alone; I’ve heard it from others as well.
Right now, that longstanding NAM theater that treats all of North America essentially the same is starting to look a little hard to manage. And it looks like your EMEA theater’s up next for a fight.
Everything is changing fast, so all this may be obsolete by the time I hit send on this newsletter, but it sure looks like we’re in for a challenging ride. Good luck out there. May cooler heads prevail.
GTM lessons from IAM
Speaking of the market (and Canada—more on that in a moment), 2025 was supposed to be a big year for IPOs. While it may not turn out that way, one that did happen was SailPoint’s re-IPO last month. One that hasn’t happened yet, but could in the relatively near future, is 1Password (which happens to be Canadian).
What’s the common thread here? They’re both part of the Identity and Access Management (IAM) space. It’s a $15B+ market growing 15% annually that’s clearly big enough to support multiple IPOs.5 There are lots of companies in the space, but SailPoint and 1Password represent very different ends of the GTM spectrum. SailPoint has transitioned from a classic enterprise on-prem business to an enterprise SaaS company. 1Password, on the other hand, has its roots in consumer PLG but has started selling their SaaS products to businesses.
And that brings me to why I find the space interesting and why I think you should too. CROs should be students of the game—keeping an eye out for everything from new cold call tactics all the way up to the structures and strategies that shape entire industries. It’s why I wrote about Vertical SaaS not too long ago. We all need to get our heads out of our own businesses sometimes so we can learn from others.
So, I asked the Gradient Works team to do something different this week and put together an in-depth report on the GTM patterns and strategies of companies in the IAM space.
The result is a completely ungated 20+ page download with no strings attached. Go ahead, get it for yourself.
It’s got specific company highlights:
1Password's shift from consumer to also serving 100k businesses on the way to $250M+ ARR
BeyondTrust’s platform-based cross-sell and upsell which drove 110% NRR and put them on the good side of the rule of 40
Keyfactor’s partner strategy which lead to 526% ARR growth
And more lessons from Teleport, AppViewX, CyberArk, Ping Identity plus SailPoint
And GTM takeaways you might find useful:
Hybrid GTM plays - how these companies balance PLG, inbound, outbound and channel
Segmentation approaches - how many of these companies serve consumer, SMB, mid-market and enterprise all at once
Benchmark & metrics - y/y growth, GRR/NRR, rule of 40 and more
Efficiency tactics - how these companies maximize efficiency post-ZIRP
Take a look and let me know what you think. If you all find these interesting, we’ll try to do more of them.
Rumor has it they also blew a 3-1 lead but we don’t talk about that.
Though I’m always looking for new content ideas…
How else would we get a hundred emails a day from friendly robots who “saw” our posts?
I learned the term “estoppel” doing research for this article.