I've been in B2B lead generation for about four years, and over the last six months i systematically automated every piece of my pipeline. not the "buy a list and blast it" nonsense. real automation where the machine does the heavy lifting and i do the strategic thinking.

here's what actually moved the needle and what turned out to be a complete waste of time.

What actually worked

Intent-based targeting instead of lists - i stopped buying lead lists entirely. instead, i set up triggers around real buying behaviour: a company just raised a round, posted a new job for a role i solve for, their CTO tweeted about a problem in my space. that single shift took my cold reply rate from around 2% to 11%. The list is smaller, but everyone on its warm.
Multi-channel rhythm with real spacing - Email, then LinkedIn DM, then email again. Not two days apart - three days. For some reason that extra day dramatically cuts down the "stop spamming me" responses. the sweet spot was five touchpoints across three channels over fourteen days.
Follow-ups that don't sound automated - Everyone automates the first email. almost nobody does follow-up two and three well. i built sequences that pull in dynamic details about their company - a recent hire, a blog post they published, a product update. Not "just circling back" but actual new value in every message.
Killed cold calling entirely - For my ICP (B2B SaaS companies between 10 and 200 employees), email plus LinkedIn plus occasional direct mail outperformed phone calls by roughly four times in meetings booked per hour of effort.

What was a waste of time

Video outreach tools - Everyone told me video prospecting was the future. my response rates with personalised videos were lower than plain text emails in every segment i tested. i think people see a video thumbnail and assume it's a pitch before they even open it.
Buying intent data from the big vendors - Services like SixFigureData and similar charge thousands a month, and their "high intent" signals were wrong more than half the time. Building your own triggers from publicly available signals (Crunchbase, LinkedIn job posts, Twitter chatter) was more accurate and cost nothing except setup time.
AI-generated first drafts with no editing - The tech isn't there yet. Every AI-written email i sent without heavy editing got flagged as spam or ignored. AI as an assistant works, AI as the writer does not.

Current results

i spend about six hours a week on outreach total and consistently book between 8 and 12 qualified meetings. the pipeline mostly runs itself. when i get lazy and stop feeding it, it still produces for roughly two weeks before drying up.

Curious what tools and processes you all are using that actually move the needle. What's the one thing in your pipeline you wish you could automate but haven't figured out yet?