Lead generation is broken.

Too many companies:

- Try to generate leads themselves, but fail
- Hire a lead gen agency with bloated retainers who can’t deliver

Lead generation as a whole is always changing, so unless you’ve got a dedicated team
on it (that knows what they’re doing) — you can expect to overpay for underwhelming
results.

That said, in 2025 — you don’t need to try to generate leads yourself, or pay an
outsized retainers for subpar work.

With Leadbird, you:

- Don’t pay for overpriced retainers
- Only pay per meeting-ready-lead we deliver
- Get the expertise of career outbound salespeople

We achieve this for you through:

- Targeted lead list building
- Domain + inbox set-up and warm-up
- Proven cold email copywriting principles
- Done-for-you cold email campaign management

Book a free demo from the link in my bio to join over 115 active clients getting
high-quality, meeting-ready-leads every single month.

Talk soon,
Nick

Nick Abraham

I recently visited a company’s site and got hit with a live cold call from one of their reps a few minutes later. I absolutely loved it.

Every bit of data we have shows that the faster you call your leads after they show any type of intent, the more likely they are to book into an appointment. 

So much so that we call leads that respond to our cold emails as fast as humanly possible to try to capitalize on this. 

But, frankly, I had never seen it done off of a website visitor play. In retrospect, it's quite obvious, though. 

1. Identify website visitor
2. Enrich and get email + phone
3. Assign to rep
4. Have rep call

There's clear alpha in this right now, and I'd love to see more companies take it on. 

The future of GTM belongs to those who pick up the phone fastest.

For the longest time while cold emailing local businesses, I removed “info”, “support”, and “help” @ [company name] emails. Wow, what a mistake that was.

I thought that those were generic emails and nobody checks them. So, I’d of course favor actual people’s (generally the owner) emails. Except over time, I realized how small this made our list.

It became clear that not all SMB business owners actually set up personal email accounts, and sometimes do operate out of the generic names (some friends in the space made this clear to me!).

And, it just…worked (see below for confirmation).

This changes a lot for us. It’s almost like we unlocked new TAM by confidently reaching out to addresses we never would have before, being that we now know they actually may be fruitful.

All that to say: if you’re cold emailing local businesses, don’t shy away from the generic addresses. They might just be exactly what will work.

I send 2M+ cold emails and 1M+ LinkedIn DMs per month for 1,000+ active clients across Leadbird and Cleverly

Mark Cuban says the best business opportunity right now is selling AI implementation to businesses. Here’s how I’d do it with cold email:

People that do it wrong will mass-send generic copy about “implementing AI” to a massive list. It will go nowhere, and they’ll claim this “doesn’t work”.

People that do it right will:

1. Find 1 problem that’s solvable with AI for 1 type of business (ex: a specific AI workflow for HVAC businesses' back office).

2. Solve it in a way you can productize/replicate (using AI).

3. Scrape a list of ONLY those types of businesses (smaller, but more targeted).

4. Email a subset of the list offering to do execute this for free in exchange for case studies/testimonals.

5. Run the playbook 3 times for free to land case studies (video customer interviews + written case studies on-site).

6. Email the rest of the list saying:

"{{name}} - want {{problem solved}}?

Asking because we did it for other {{company types}} like {{company 1}}, {{company 1}}, and {{company 3}}.

Worth a chat?"

This has really always been the playbook for getting services businesses off the ground. AI has just increased demand for certain services, and made it faster to deliver for the vendors.

If I were fresh out of college trying to start something, this would likely be it.

It feels like every week, we see people in outbound preach about a new data vendor, infrastructure provider, or even sending platform. 

Even within tools themselves, I feel like I can't scroll the timeline without seeing Smartlead, Clay, or others announce a big new feature.

It's both awesome and exhausting, sometimes.

At the same time, it's indicative of how much this space moves and how fast it does. Every outbound marketer needs to be on their toes constantly to ensure they're not missing something. 

And though I don't want to come across like I'm fear-mongering, it's true. I've seen this in the last year with competitively generated agency owners, unfortunately fizzling out because they were stuck in their ways and didn't adapt. 

You absolutely need to keep your finger on the pulse of all tools and strategies in this space or you will fall behind. 

I like to frame it mentally as being a student of the game. It is ever-evolving, meaning it is quite literally impossible to master it. 

That’s my Tuesday reminder.

I've sent over 3,000,000 cold emails since January 1st. Here are my 25 rules for deliverability every outbound marketer needs to know in 2026:

Infra:

1. Always use Google or Microsoft accounts, never SMTP. SMTP providers consistently fail after a few months when spam filters update.

2. Domain endings don't matter much. Use .com when possible, but .org, .info, .co, and .biz work fine because they're cheaper.

3. Warm up domains for minimum 2 weeks, ideally a month for better performance and longevity.

4. Use an A/B rotation strategy. Buy two sets of infrastructure per client - use set A in month one, set B in month two, then rotate.

5. Keep emergency non-branded domains ready as backups when main infrastructure goes down.

6. Use proper naming conventions: Client_Campaign_Industry_LaunchDate_Group to identify performance issues.

7. All providers will eventually mess up DNS settings at scale. Always send yourself a test email and check headers before launching campaigns.

Campaign setup

8. Don't track opens. Open data is unreliable and hurts deliverability.

9. Plain text only. No fancy formatting.

10. Send Monday-Friday, 9-5, but randomize the exact time daily within a 30-minute window using Smartlead's API. Saturdays work well for SMBs and mid-market targets.Keep campaigns running on holidays because business owners check email every day.

Lead list management

11. Format company names properly, but ideally don't use them at all. Campaigns perform worse when they include company name or title.

12. When using titles, normalize them with AI (e.g., "CEO" not "Chief Executive Officer").

13. Remove all email security from your list. Use Google's DNS endpoint to identify anyone with Proofpoint, Barracuda, Mimecast.If removing security emails makes your list too small, put them at the end instead.

14. Don't isolate ISPs. Mix Google, Outlook, and SMTP recipients or your domain gets flagged.

15. Always validate with any tool right before launch, regardless of what data providers claim.

16. Separate catch-all campaigns from regular campaigns because catch-alls have higher bounce rates.

Copywriting

17. Spintax everything, but review all variants yourself. AI creates variants that sound unnatural.

18. Keep it simple. No images, PDFs, or links in the first email.

19. Two tests for copy: Can a 10-year-old understand what you're selling? Would you actually say this to someone in person?

20. Make copy conversational, not complicated with technical terms and corporate language.

21. Strong offers > personalization. Focus on making the offer better instead of building complex personalization workflows.

Continued in the comments.

Name drop campaigns make 100% automated emails look like warm referrals. But most people execute them wrong and kill credibility. Here's how to do it right:

The concept: You reference a colleague at the prospect's company in your cold email. This makes it look like you sat there, did your research, and wrote the email normally. When done right, these convert 2-3x better than standard cold outreach because you've lowered their automated email guard.

The mechanics:

- Pull colleague names by identifying their senior leadership (Find People → Department → Seniority) within the company.

- Reference the colleague in your first line ("not sure if you're the right person or if I should be speaking with [[ColleagueName]]")

- Optional: Add a PS that ties back to the colleague reference

- End with a soft CTA

The entire email should almost read like someone at their company gave you a warm intro.

Common execution mistakes:

- Generic opening lines that don't reference the colleague
- Formatting that exposes it as automated (inconsistent spacing, broken variables, obvious merge fields)
- Not cleaning your data before sending (colleague left the company, wrong person entirely)

The prospect should almost think a coworker mentioned them. When they open the email and see their colleague's name, their guard drops. That's when your message actually gets read instead of deleted.

We’re seeing incredible results with this right now…let me know if you want to hear more on it.

undefined undefined

Hey, thanks for replying. 

Here's the guide:

https://drive.google.com/file/d/1Ogdq83o7sRMARFopD_t_-nqHF0XIOPsD/view?usp=share_link

Leadbird

P.S. If you’re an established agency or SaaS with a GREAT offer and want leads — Leadbird can help

… and we’ll do it on a Pay-Per-Lead basis.

Book today: https://www.leadbird.io/book-a-call

Spintax Guide

Tool

Article

Message

yuri

For the first time ever, YC included agencies in their request for startups. I’d be lying if I said I was surprised.

Historically, Agencies were rarely considered in venture conversations because of the required human capital. Revenue scaled only with headcount. This made software-like margins impossible to replicate.

YC’s argument is that this is no longer the case, thanks to AI. In a roundabout way, they're saying you can get the same output with less humans, and that all of the salaries you aren't going to pay go right back to your bottom line. 

That's the semi-obvious stuff. The less obvious angle here is that YC realized how fundamentally great the business model is compared to what they typically invest in. 

The venture game is one of moonshots. You pour a bunch of money into an asset that will burn cash for 5-10 years and hope it either turns cash flow positive at scale, gets acquired, or IPO’s.

By no fault of YC, this leads to a lot of failed companies and a lot of burned investor money. 

To me, YC asking for agencies is almost a safer hedge to a lot of the moonshot venture plays they try. Sure, rarely do agencies get acquired for what SaaS companies do or even IPO…but they're much more predictable and easier to spin up and operate than a lot of other companies. 

I say this is someone who's automated and systemized a lot of our fulfillment. We're able to run an extremely lean team for a relatively large client base because of these systems. And I do think this will become the norm moving forward.

Cold email sending schedules should not be the hot topic that they are. Really, the rules for sending are simple (backed by 25M+ cold emails):

1. Send Monday-Friday, 9am-5pm. We use Smartlead’s API to randomize start and end time daily (+/- 30 mins of start and end time to increase variability).

2. Saturdays can, in fact, be a good day to hit SMBs or mid-market companies. 

3. Same goes for holidays.

It’s that simple. Stop reading in to data about which days and times perform better. You’ll drive yourself crazy.

There’s a massive business opportunity for someone that wants to use this cold email right now. Let’s walk through why:

I don't need to educate you on the AI opportunity right now for businesses that aren't using it at all. Businesses, especially local ones that operate way outside of our bubble, have arguably more upside than anyone in terms of properly using AI to boost efficiency.

There are huge consulting shops being spun up right now that solely implement this for customers.

And I think there's a huge play to do so locally with this cold email alone. Why?

1. The subject line is relevant enough to a business owner in a given suburb / area.

2. We layer on the local social proof in the first line with an outcome promise as well. 

3. We really play up the local angle by effectively saying “we are going to be in the area upcoming, so let us know if you are interested”. 

4. We of course end with a soft CTA. 

Cold email doesn't need to be that difficult. You just need to have a strong and unique offer presented well. This is a great example of that.

People are taking spintax too far and it's ruining their cold emails. They're trying to spin individual words to increase variance, but this often leads to certain sentences just not making sense or sounding weird when read out loud.

This is not the way. You clearly show your prospect that a human did not write the email.

The right way to run spintax:

1. Identify the right variables to do it with in the first place.

Certain ones are over the top and risk-prone for messing up the email. Especially with certain Clay workflows, people are going overboard.

Easy, less risk-prone ones:

- Case studies (switch out company name and result)
- CTA sentence (spin 4-5 different ways of saying "Can I send you more info?")

2. Instead of spinning specific words, do it with specific phrases all together that still mean the same thing.

This reduces the chance of the email sending platform creating a sentence that just looks weird or doesn't make sense.

This is “boring” stuff but it's important.

My agency has worked with 500+ clients. In the last 5 years, I've learned a TON about how to deal with unhappy customers. My top 3 learnings about doing it:

1. Product <> Fulfillment <> Sales must all be aligned.

All messaging has to be completely aligned at all times and updated frequently across the board if needed.

Customers can and will get confused by what you are saying, pitching, or delivering.

Example:

Our product team built a new way to launch campaigns, and our fulfillment team started leveraging it. But, our sales team is still pitching the prior way (the outcome was the same, but the workflow is different).

This ultimately led to confused customers and was an issue we had to mitigate.

Don't let that happen!

2. Your contract is, literally, the law.

This goes for any B2B company. Everything you put in there on how your service gets delivered is very important. So is the reverse (everything you don't do.)

This will help prevent scope creep and give you clear boundaries on what you do and do not do for clients.

3. Calling mechanisms are crucial.

Similarly, this goes for both lead generation agencies and other B2B companies. Do not let anyone on your team get comfortable with doing the back-and-forth email battle.

The much more efficient, personable, and higher-leverage play is simply calling the client and figuring out the problem.

Broadly, most people are reasonable and somewhat logical, so explaining your stance (as long as it's justified) should work just fine.

Ensure you have a cell phone data/calling mechanism for all clients.

I know this wasn't anything flashy, but I promise you this is what is important at scale.

Almost every day, I get a cold email referencing my engagement on a social post I didn't think twice about. Cold emailers are using this signal the wrong way, and it's ruining their reply rate. Here's how to fix it:

People engaging with content in your space are actively researching the problem you solve right now. These lists convert at 3-4x higher rates than standard cold outreach because of recency and intent.

The wrong way:

"Saw you engaging with [Creator's] content about cold email..."

This feels like surveillance, not relevance. 

The right way:

"Noticed you've been engaging with content about outbound lately..."

Reference the intent and topic, not the creator. The intent matters, but usually the specific creator doesn't.

How to run this properly:

1. Identify 5-10 creators with your ICP audience (look for competitive overlap and high engagement).

2. Build a DNC list before scraping (generate competitor domains and exclude them so you don't email creators directly)

3. Scrape engagement weekly with Apify (pull last 7 days of engagers to keep lists fresh)

4. Deduplicate against a master database (dedupe by LinkedIn URL first, then at ingestion and pre-campaign)

5. Enrich and route channels (email first, if open profile → send InMail, then connection requests)

6. Use intent-based copy, not creator name drops

Common mistakes that kill these campaigns:

- Scraping once instead of weekly (make the campaign evergreen)
- Not excluding competitor domains (you email the creator and other competitors)
- Name dropping creators in copy (feels invasive)
- Not deduplicating (hitting same people repeatedly)

Quantity of prospects usually isn’t high (especially after filtering down), but the quality is solid if hit right. Strongly recommend!

The problem with cold email is that once people figure out what’s working, they overuse it and run it in to the ground…until it no longer works. Let’s fix that.

The perfect example of this is the time-tested campaign copy:

“If we could help generate 1[result] with our [unique mechanism], would you be interested in speaking?”

It worked for a while because it wasn’t intrusive, was relatively low-risk, and promised an outcome (those principles remain, by the way). But, it got overplayed.

The fix to this is not simply different copy, but offering something different altogether.

By this, I mean:

1. Identifying a small set of deliverables everyone in your ICP ideally needs, or could benefit from.

2. Anchoring your cold email around offering that for free.

3. Doing a great job at it, and then opening up the conversation to more services.

There’s a great example of this in the graphic below. Yes, you’ll do some free work. But you have the right to qualify leads up front and only do it if it makes sense.

Like it or not, this will help a lot of outbound teams “win” cold email coming up.

Last week, I posted about being cold called by a rep after I visited their site. I said I loved it, and people were…not happy, to say the least.

I saw tons of comments saying that this was “creepy” and bordering on an “invasion of privacy”.

I’m not here to debate the efficacy of the play. If your company books demos from it, do it. If you don’t, then don’t.

But this privacy point is hilarious to me.

People wrote comments about a company using their online behavior to give them targeted messages at a specific time.

…and then opened Instagram and saw 3 retargeting ads from sites they visited in the last week, and didn’t bat an eye.

My point is that companies do this all day long. We wouldn’t have some of the social media products we do if advertisers couldn’t pay for this ability.

It’s not novel. It’s just being adopted in B2B.

Adding prospects to a Google Doc or sending random calendar invites to stand out in the inbox isn’t new.

Docs, Loom, Canva - people have used these for years because their domain reputation helps deliverability. And yes, you’ll likely land in primary.

But at scale, it’s not the unlock people think it is for two reasons:

1. Enterprises run heavy phishing training. If they don’t recognize the sender, they’re told not to click. That kills the play fast.

2. It carries the same negative sentiment as cold email. Some people will think it’s clever. Others will hate it.

Sorry to burst your bubble.

I used Notion Agents to build a lead gen competitor intelligence agent in maybe 60 minutes. How it works:

1. Discolike grabs the largest B2B lead gen agencies (our competitors).

2. An agent runs via Firecrawl MCP that finds updates on them (new case study, site update, etc.)

3. Sends me a monthly update of all new competitor activity.

No, this isn’t going to change my life and suddenly add $100K of revenue to my agency. But, it’s a dead-simple way to keep me in the loop on our competitors without doing a bunch of research or paying a VA to do it.

Not to mention, there are obvious applications to use this data in a lead gen setting. Ex: Monitor prospects’ competitors’ activity > reference their activity in an email to a prospect.

I don’t think Notion Agents is getting enough love right now, so I thought I’d share.

"We don't chase customers if they don't get back to us after the first reply, mark them as closed lost." That’s a real quote I heard from someone trying to get leads from outbound.

Short take: Some people just don't want it.

Long take: Outbound marketers everywhere are complaining that reply rates are down, deliverability is getting harder, and meetings are scarce, to say the least.

Prospects are fatigued from being hit with the same cold email template variants and are only replying to the emails that are strong, relevant, and have good offers.

So to effectively “waste” a good lead because you had to follow-up with them more than once is simply irresponsible.

In fact, I think the winning outbound teams of the next few years will:

1. Get back to positive replies borderline instantly.
2. Be extremely regimented about maintaining their CRM.
3. Prioritize strong, tasteful, relevant follow-ups until Closed Won or Closed Lost.

People who are disciplined about chasing positive replies will run circles around those that don’t follow up more than once because they “don’t chase leads”.

I can’t believe this is even a take.

The easiest and fastest way to write more relevant, personalized email copy with 0 extra work is to let your list write the copy for you. Let me explain:

I recorded a podcast episode with 🦾Eric Nowoslawski a while back, where he explained this perfectly. He explains that marketers often go too broad with their targeting, making relevance much harder to achieve.

He goes on to share that you can simply tighten up your list with certain filters that make it easier to write copy that’s relevant to everyone on that list.

Perfect example: Let’s say you’re a bookkeeping service.

You can reach out to CEOs of companies in the USA with <20 people, and say something like:

“We help small businesses in the USA close their books within 5 days at the end of each month.”

…OR you can reach out to:

- CEOs of companies in the USA with <20 people
- Who founded their company less than 2 years ago
- Are a first-time founder
- Don’t have a finance and accounting person on their team

And say something like:

“I saw that you guys were recently founded, and congrats, it's your first time running a business like this. 

How are you thinking about your finance operation? Asking because we help…”

Do you see how simply filtering down the list makes it easier to write more relevant copy with no extra work? You of course give up some volume in doing so, but the tradeoff is often there.

I'm not exaggerating when I say that every single company should make a practice of doing this. It's a low-effort, high-reward way to get meetings on the board from outbound.

I'm talking about simply reactivating old leads that never closed or fizzled out from a time in the past. These are people that showed a level of interest in what you were offering, and they were more familiar with you than a true cold lead. 

Not to mention, especially if you're kicking off an outbound motion, these emails are more prone to replies. This will build trust signals early with Google and Microsoft, and overall improve deliverability. 

This is important enough that at onboarding, we ask clients for lists of older leads they want to reactivate. By reaching out to them, we're usually able to get 1-3 wins on the board in the first 7 days of a campaign going live. 

The messaging is fairly simple as well:

"Hey {{name}}, we spoke in the past about {{what you do}}.

Wanted to touch base again because we {{did something new / have new method, etc.}}.

Would it make sense for us to chat later this week?"

If you need any help with this, comment and I can answer as best as possible. I strongly recommend you get it up and running if you haven't already.

I’m convinced every B2B service provider should reach out to churned clients at the 30-day post-churn mark to win them back. As unconventional as it sounds, it’s been working well for us. My thesis:

Clients churn because they think the grass is greener on the other side (another vendor that’s cheaper, running the process in-house, etc.). In our experience (including vendors we’ve churned from) this often isn’t the case. 

Especially with regards to lead generation agencies: if an agency can get you some* leads (even if it’s less than you anticipated), that little bit is better than zero leadflow.

All that to say:

We’ve learned that clients aren’t always happier post-churn. Because they’re familiar with you and your process, you’re ripe to reach out again (unless you ended on bad terms, of course).

There are simple, non-intrusive ways to do so for every B2B service business. For a lead generation agency like us, for example, the messaging might say:

“Hey - noticed you never asked for your domains back. We could spin up infra today and relaunch a campaign if you’re still open to it.”

I’m not saying this has a 100% success rate. I am saying it’s worth doing.

Founder @ TrueAdvertize | B2B Outbound partner & Nice Guy | High Performing Swimming Athlete

Hey! This is Samuel. 

My day starts swimming as a high-performing athlete and it ends helping to scale ambitious B2B Companies 

I am a founder of a narrow-but-refined Claygency with expertise in B2B outbound strategies and mechanisms that drive business growth.

What I do, it's simple: 

I help B2B Companies grow by generating more qualified prospects through outbound strategies at scale using Clay!

And what I mean by "growth" is:

→ Generate a predictable flow of high-quality prospects consistently.
→ Close more sales opportunities, therefore increasing top-line revenue.
→ Fill your pipeline with dream prospects.
→ Embrace AI, automation, and streamlined client acquisition channels.
→ Don't stress about lead generation again.

That's what I do.

If you are interested in seeing how this can work for you, don't hesitate to message me and I'd be more than happy to help you.

P.S. I am quite interested in pushing myself beyond my limits too. For that reason, I became World Junior Champion 2021 in Finswimming.

If you'd like to talk about growth, apply to work with me (or chat sports), feel free to shoot me a message, and let's connect. 

Cheers 😘

Samuel Roa Cárdenas

how I'm feeling lately. 


Claude Remotion Control is gonna help me deploy GTM systems for our clients while I'm playing golf hehe.

If you don't know what is the thing I'm talking about here. Claude what? 

Check this: https://lnkd.in/eCekMUZ6

Founder @ TrueAdvertize | B2B GTM partner & Nice Guy | High Performing Athlete | I use Clay for a living

Stop looking for the perfect signal. It doesn't exist.

Every GTM team I work with wants to find the ONE data point that predicts buying intent, the signal that tells you exactly which company is ready to buy.

That's not how it works.

Some of the most useful data is private. 

Revenue is a good example, it's estimated from employee count, it's often wrong, and you can't verify it, but that doesn't mean you ignore it. 

It means you build a system around it that gets closer to truth.

You take what's public. 

Job postings. 
Org changes. 
Technology adoption. 
Content they're publishing. 
Vendor shifts.

Real-world behavioral data that anyone can see.

Then you match those patterns against what you already know about companies that bought from you and then you can answer the following questions: 

What were they going through when they signed? 
What was happening publicly before the deal closed?

That's where Clay and Claude Code come in, to basically orchestrate dozens of signals together with the context of what actually made similar companies buy.

The signal doesn't close the deal, the context you give the system to interpret that signal does.

Any data point can be useful. What makes it powerful is knowing what to point it at.

There's this voice in my head every time I hand something off to my team.

"You could do it faster yourself."

And honestly? It's probably true. 

I could jump into Clay and build worklflows from the ground.

I could open Claude Code and set up the systems to generate programatic account research myself.

But that's the trap.

I'm still building my agency right now and not as you may think, like me being in the woods or somethig like that. No. 

I'm working on an amazing project that later down the line I'm going to talk about here and it requires 90% of my attention every single day.

So the hardest lesson isn't technical. 

It's not about Claude Code or Clay or which system to build next. It's my own ego getting in the way.

The goal isn't to do more work. It's to build systems the team can scale without me.

I build the new thing. The team runs it. I build the next new thing. They run that too.

Last week I shipped a content pipeline that integrates a scraping system, and a scoring workflow across 5 clients aside the current GTM work, each one with its own taste, and the hardest part wasn't building any of it because they were new to me.

It was walking away after.

Every time I step back, there's this pull. "Just let me make sure about that one thing." But if I get into that one thing, I'm the bottleneck again. And the whole point was to stop being the bottleneck.

The founders who scale in 2026 aren't the ones who work the hardest. 

They're the ones who build systems that run without them in the room.

I'm still learning how to be that founder. Every single day.

I keep watching founders demo new tools on LinkedIn like it's a personality trait.

Meanwhile, the ones actually closing deals aren't talking about models.

They're too busy building systems around what they already know about their customers.

I run Claude Code + Clay (no, I don't replace Claude code with Clay, I enhance the power of Clay using claude code) across every client project. It's fast. It's excellent. 

But Claude Code or any other tool has no idea why my clients' customers buy. 

It doesn't know which CRM signals are real and which ones are noise. 

That part is mine. That part is the whole game.

The competitive edge in 2026 isn't the tool.

 It's the operator. 

Your understanding of the problem. Your context. Your ability to structure what you actually want before you touch anything.

AI without context is just expensive autocomplete. 

Systems thinkers are going to run circles around tool collectors this year and most people won't notice until it is too late.

Now when we onboard a new client, a system triggers automatically to build the context that we need to start working systematically on the client.

No guessing, just pure structure. 

Here's what we actually build:

⤷ Claude Code pulls their CRM closed-won data and finds the real buying triggers in the web.
⤷ Clay enriches the next target list using those triggers as filters and we push the list automatically to Clay.
⤷ Claude Code generates three message variations matched to what actually made similar companies sign - we use PVP and PQS depending on the data that is available in the closed won deals that sits on the CRM.
⤷ First 200 sends go out, we usually test small and then scale if it makes sense.
⤷ Reply data feeds back into Claude Code
⤷ Next round is tighter. Every time.

The big difference here is that the old way treats outbound like a broadcast. This new system that we built treats it like a conversation that gets smarter with every send.

That's compounding context... a concept that I'm sure you're gonna hear a lot from now on.

So the key is step one. If you skip the CRM analysis, the rest of the loop runs on assumptions. And assumptions don't compound.

We're testing this version of this right now where the feedback loop runs weekly instead of after each campaign. That makes more sense now, I'll share the results here once we have done enough testing.

I deleted a database because I told Claude Code "just delete it."

No way to recover it.

Claude Code is the most powerful software tool I've ever used and I'm designing my entire business around it + Clay and Sales/GTM thinking, but the context can get away from you, and that can't happen.

You give claude code a prompt and it goes 1,000 miles a minute, and then you have to go track down everything it did to understand if it's right, the dependecies, the QA checks.

It's quite exhausting and it's worth to a certain point.

I had a CRM ingestion workflow running perfectly, clean data, correct formatting, everything mapped to the right fields. 

Then I kept building on top of it, and because Claude Code kept going, it lost that context and overwrote the whole thing.

That's the nature of it, you can do a lot, but when it makes a mistake, it can make a really big one, like deleting your entire project database because you got sloppy with one sentence.

Now Clay, Clay forces you to think differently.

What format do I want? 
What happens in this exact situation? 

If you're writing data to 200,000 CRM rows, you can't vibe code that easily* 

You need to see every error and correct it yourself or set up great underlying processes that catches those things for you to review, again making sure it has the correct context. 

At the end of the day, setting things like that depends on your OWN situation, not on what people tell you to do without your context.

But what I can tell you right now is that voice isn't precise, Clicking is precise.

So the way I think about it now: Claude Code is for building, Clay is for running.

One is creative, the other is reliable, and your business probably needs both, because the moment you confuse them, you can lose something you can't get back.

When you think you have seen it all when it comes to outreach.

Then you got another surprise. 

Literally my headline says that I use Clay for a living, like if even if he didn’t automate this he just coulda checked my profile and understand that. 

And why is he asking me if I need a list of SaaS companies with the head of growth emails? 

If you think about that, the main reason why you use “clay” is to avoid things like that, it help you segment, qualify and in this specific situation DISQUALIFY.

The list is the message. 

And to make it even funnier, he said SaaS - I work mainly with SaaS so I’m really good getting high quality segments even for myself. 

I’m just surprised that in just one attempt to reach me out he completely disqualified his credentials by not doing what he actually need to do when working in this space. 

Think about it.

Anyway. 

Happy Sunday! 

Hopefully this post may help you to think twice about the outreach campaigns you run and more importantly how you make the other party feel.

Your CRM already knows why people buy. You're just not reading it correctly.

I'm running Claude Code across three client CRMs while my team pulls data through Clay on their closed-won deals. 

Not deal size or sales cycle length. The actual transactional trigger. 

What made someone go from "interested" to "yes, let's do this."

The reasons people buy almost never match what the sales team thinks. 

In the last CRM we pulled apart, marketing said the product demo closed the deal. 

The data said something else: 6 out of 8 closed-won deals had a failed vendor switch in the 90 days before signing. 

That wasn't in any sales deck.

Once you find that pattern, outbound changes completely.

You stop writing messages based on what you think prospects care about.

You start writing based on what actually made similar companies sign. 

That's not personalization. That's pattern recognition.

The data is already sitting in your CRM. Nobody's connecting it to outbound.

This is the unsexy work. Period. 

No Claude Code screenshots. 
No Clay workflow GIFs. 
No viral automation thread. 

Just reading your own data so every new campaign starts from proof, not guesses. That's real revenue engineering.

The pattern I keep seeing with early-stage founders looks something like this.

They hire before they have a playbook.

Series closes or something happen, so the board or just them say grow the team, so they hire SDRs, a marketing lead, and maybe a VP of Sales.

Six months later, pipeline hasn't moved.

Not because the people are bad, most of the time they are great professionals, but because nobody built the system they were supposed to run.

The SDRs don't know which accounts to target because nobody analyzed the closed-won data. 

The marketing manager is producing content with no feedback loop. 

The VP of Sales is managing a process that doesn't exist yet.

I keep seeing this, more often than I'd like, including struggles in the leadership when it comes to decide how the GTM team will be deploy. 

(tools & people)

So the fix isn't hiring less, It's sequencing it differently.

Before you start hiring a lot to increase your revenue based on more headcount, you need to know actually who actually buys. 

Why they buy
What signals show they're ready. 

That's a system problem --> not a headcount problem.

Build the system first with the data you already have to back it up, then hire someone to operate it.

Maybe you'll need to hire someone who understand what you're going through and get your GTM systems in place.

Or maybe not, it all depends on how well structured your data is and what hwo you're using it to increase revenue

So what I know is that the founders who will scale faster and cleaner this 2026 aren't the ones with the biggest team. 

They're usually the ones who built the system before they built the team.

There's one question I ask on every first client call and I keep asking after to our clients

The answer tells me everything I need to know about whether their GTM can be deployed.

"With perfect information, unlimited resources, and unlimited time - what data would you try to know about a company you want to sell to?"

Most clients freeze, and they should of course.

Not because they don't know, because they've never been asked to think about their market that way. 

They've been so busy running GTM motions that they never stopped to define what perfect targeting actually looks like.

When they answer, three things become clear.

1. Whether they have real product-market fit, you can hear it in how specific their answer is. 

 Clients who know their buyer describe exact scenarios, clients who don't describe firmographics.

2. Which data they're missing. 

If they say:

"I'd want to know which companies are actively evaluating competitors"  

That's a signal we can build a Clay enrichment around right now. 

If they say:

"I'd want more leads" 

We have deeper work to do first.

3. Whether their CRM is actually tracking the right things. 

Usually it's not, the data they need is sitting in their sales director's head, even in their sales team's head, not in their CRM.

That one question turns a vague "we need more pipeline" conversation into a buildable system that allow us deploy GTM teams correctly.

I've been running enterprise ABM for about 5 months now.

The single hardest thing to shift, everything I knew about SMB go to market.

SMB is greenfield, you start from zero if the client doesn't know very well their target market.

We build the list, write the messages based on the data, set up Clay, hit send and then we run multiple tests, shipping fast, until something sticks. 

(Again, this is only if we don't know what we want to know, the companies that actually would buy)

The whole system is yours.

Enterprise is the opposite.

They already have sales reps, marketing teams, content teams, CRM workflows, a decade of institutional knowledge sitting in people's heads that never made it into any system.

Our job isn't to replace any of that, it's to enhance what they already have.

When we walk in with Clay and Claude Code, the goal isn't to build something new.

It's to take what the sales team already knows and make those signals scalable.

Which accounts are investing in adjacent technology.
Which ones just restructured their sales org. 
Which ones are buying based on patterns the CRM was never set up to track.

Enterprise clients are comfortable spending $10K on a single discovery call, and it is a reality.

So the sales cycle isn't the problem, the problem is knowing which accounts are worth that call in the first place.

That answer doesn't come from a better tool, It comes from listening to the people who've been closing deals there for 15 years and turning what they know into a system that runs at scale.

Last week I was inside a Clay workflow, building enrichment logic for a new account list. The kind of thing that normally eats a full afternoon. Prompt testing, filter tweaking, reformatting output until it matches what the sales team needs.

This time I open Claude Code first.

I've been training a system that understands how Clay works. It generates prompts that match Clay's logic. Spins up workflows faster. And when the data comes back, it loops into the next iteration automatically.

So instead of 3 hours refining one enrichment table, the setup is done and I'm focused on the part that actually matters. 

Which accounts to target. 
Which signals are real. 
What message gets a great replies.

That's the thing people keep getting wrong. 

They think it's Clay or custom infrastructure.

 One or the other. Pick a side.

The answer is both. Clay handles the enrichment. Claude Code handles the thinking around the enrichment. 

The bottleneck was never the data. 

It was knowing what to do with it.

The competitive edge isn't which model you pick. 

Every founder has access to the same tools. 

The edge is knowing what you want and using Claude Code to get there faster.

The difference between a company that scales and a company that just gets bigger is one thing.

What they build before they hire.

Companies that scale build systems. 

Repeatable systems that produce consistent results no matter who operates them.

Companies that just get bigger add headcount. 

More salespeople doing the same unscalable thing, more managers managing more people doing more of the same.

I see it in almost every GTM team I work with, the instinct is always to hire. 

The board expects headcount growth. "We need 5 more SDRs" sounds like progress. 

But 5 SDRs running the same manual process doesn't give you 5x the output, you just get the same playbook at a higher burn rate.

One Fantastic SDR with the right systems in place can become a GTM lead that produces 10x better outputs compared to the team of 5. 

I've seen it, I've done it.

So the companies that actually compound their results are the ones that before hiring the next person, let's make sure that the right GTM infrastructure is in place, not just a simple workflow someone has to babysit.

Systems that learn from their own output and get better every cycle. That's the model. That's what we're building as well.

More people is a linear equation. Better systems is a compounding one.

taplio

The Battle

Nick Abraham

Samuel Roa Cárdenas

Followers

Posts

Engagements

Likes

Performance ScoreNick Abraham's LinkedIn Performance

Performance ScoreSamuel Roa Cárdenas's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Nick Abraham

Samuel Roa Cárdenas

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Nick Abraham's LinkedIn Performance

Performance ScoreSamuel Roa Cárdenas's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreNick Abraham's LinkedIn Performance