How We Started Using OpenClaw for Lead Gen and Why I Wish We'd Done It Sooner

I keep telling myself I'm gonna write this post and then I don't write it. It's been like three weeks. A buddy of mine in the industry has been on my case about sharing what I've been doing with OpenClaw because he thinks other production companies would want to know. He's probably right. He's been right about a lot of things lately which is honestly getting kind of annoying.

So here we go. I'm writing it. It's a Tuesday night and I should be colour grading a project I shot for a wellness clinic in Sherwood Park but instead I'm writing about lead generation. The client already emailed twice about the rough cut. Sorry Marcia...

Some Background First

I run a video production company in Edmonton. Corporate stuff, commercials, brand videos, some event work, drone, real estate, that kind of thing. Been at it for a while now. Before I found OpenClaw, my process for finding new clients was basically: sit at my desk every Monday morning for four or five hours and cold email people.

I'm not exaggerating about the four or five hours. That was a real number. I'd google things like "construction companies Edmonton" or "tech startups Alberta" and just open a million tabs and try to figure out who might want video. Then write emails. So many emails. Very "Dear Sir or Madam, we would love the opportunity to discuss your video needs" type stuff. Nobody replies to those emails. Or almost nobody. My reply rate was like 3%. I looked it up.

And the leads that did come through were often just bad. I still remember the guy from Mill Woods who wanted a "Netflix-quality documentary series" about his supplement brand for $400. Four hundred bucks. For a series. I think about that interaction more than I probably should.

I didn't get into video production to write cold emails all day. But that's what like half the job was. Mondays were brutal. I'd finish the prospecting and be too drained to do any actual creative work. It sucked. There's no other way to say it.

How I Found OpenClaw (After Ignoring It For Way Too Long)

So January 2026, I'm at a networking mixer downtown and I'm talking to this guy I know who does commercial work in Edmonton. We're a few beers in and he starts telling me about this open-source AI tool he's been using to find clients. OpenClaw. He's going on about "autonomous agents" and "automated pipelines" and I'm nodding along but honestly I'm mostly thinking about the Oilers game I'm missing.

I'd been burned before is the thing. Tried Instantly, tried some scraping tool I can't even remember now, spent $2,800 on Facebook ads that got me exactly two leads who both ghosted. So some guy at a bar telling me about another AI tool was just noise to me. I figured it was another shiny thing that wouldn't actually work.

He sent me a bunch of links the next day. Reddit threads, YouTube videos, a GitHub page. I didn't click any of them for like two weeks.

Then one Saturday I'm at home with nothing going on. It's snowing. I'm bored. I finally click one of the links and end up going down a rabbit hole for like three hours.

The thing that got me to actually try it was this: I gave OpenClaw a test task. Find manufacturing companies near Nisku that had been around for ten-plus years and had outdated websites. It found four in about an hour. But not just their names — it pulled the marketing contact at each company, links to their websites, recent news about them. One of them had just been in the Journal for expanding their facility. Another one had posted a job for a "communications coordinator" which if you know anything about this industry means they're about to start spending money on marketing.

That job posting thing is what got me. I never checked job boards when I was doing prospect research. Never. It wasn't part of my process. I'd set up the agent to flag marketing-related hiring as a buying signal and it just... did that. Automatically. While I was making a sandwich.

That same research would've taken me most of a Monday morning doing it manually. And I wouldn't have caught the job posting. I just wouldn't have.

What It Actually Is

OpenClaw is basically an AI that can go do things on the internet. Not just answer questions like ChatGPT but actually browse websites, check LinkedIn profiles, read pages, and put together reports based on what it finds. You give it a goal and it figures out how to do it. It runs in the background while you do other stuff.

You do need to be somewhat comfortable with command line stuff to set it up. I'm not gonna pretend I found that part easy. I don't have a CS degree or anything. But I'm the kind of person who'll sit with documentation for a full weekend if I think the payoff is worth it. Took me all of Saturday and Sunday for the initial setup and then a few evenings over the next couple weeks tweaking things. If you're not that person, there are people offering OpenClaw setup as a service now. Google around, you'll find them.

How I Use It Day to Day

OK so the most important thing I did, and I cannot stress this enough, was write a really specific description of who I wanted to find. Spent about two hours on this and it felt like a waste of time while I was doing it but it turned out to be the thing that made everything else work.

When I was vague about it OpenClaw returned garbage. First week I told it something like "find companies in Edmonton that might want video" and got back a pet groomer, two daycares, and a mud flap manufacturer. So. Be specific.

What I actually want: tech companies, manufacturers, energy companies (renewable and traditional — it's Alberta, oil and gas is still a massive part of the economy here), agriculture businesses with consumer-facing brands, and professional services firms with 50 to 500 employees. Under 50 they usually can't afford me. Over 500 they've got agencies or in-house teams.

Geography is Edmonton metro, Calgary metro, and anywhere along the QE2 if the project is right.

And then I told it what signals to look for. This is where it gets good. Websites with little or no video, or with that generic 2017-era drone-shot-of-a-building "welcome to our company" stuff that every production company was pumping out eight years ago. Job postings with "marketing" or "brand" or "communications" in the title. Recent funding rounds. News about expansion.

The agent runs overnight every night. Checks business directories, the Alberta company registry, industry association member lists, actual websites which it reads and evaluates. I built a little scoring thing where it rates each website from 1 to 10 based on how much video they have. A 2 means stock photos and text walls. Good prospect. An 8 or 9 means they've already got a YouTube channel and a produced hero video. Probably not my target.

It also checks LinkedIn through Sales Navigator. Picks up hiring signals, company milestones, headcount changes. I've found that companies that just hired their first marketing person are like 3 or 4x more likely to invest in video in the next six months. New marketing hires wanna make their mark. Video is usually on their list.

By 7 AM there's new leads in a Google Sheet. Some days five, some days twelve.

Then it qualifies them. Used to take me 20 or 30 minutes per lead to check their website, find the right contact, assess whether they were worth emailing. OpenClaw does all of that now. Checks if the site is mobile responsive, looks at when it was last updated, scans for a careers page, checks what the company's competitors are doing with video, audits their social media. Then it scores everything and pushes the good ones into HubSpot.

Above 8 is high priority, I look at those first thing in the morning. 5 to 7 gets reviewed midweek. Below 5 I mostly skip.

The Part I Care About Most

I don't let OpenClaw email anyone.

I need to say this clearly. I know there's a temptation to automate the whole thing end to end and I think that would be a really really bad idea. People can tell when they get an automated email. They can always tell. And if the first contact someone has with your company feels like a robot wrote it, you're done. They're not gonna hire you. Especially not for something as personal as video production where the whole value prop is basically "trust me with your brand."

So what happens is I open HubSpot in the morning, look at the new high-priority leads, and I write each email myself. By hand. Every time.

But because OpenClaw already did all the research, I know a ton about each company before I write a single word. I know their recent news. I know what their website looks like. I know who the decision maker is. I know if they just hired someone. So instead of writing some generic "I'd love to work with you" thing, I can write something like:

Hey Marcus,

Saw you broke ground on the new facility near Nisku, congrats. Also noticed you brought on a comms manager a few months back so it seems like you're putting real investment into marketing right now.

I run a production company in Edmonton. Had a couple ideas about how a short video could help with recruiting for the new site. 15 minutes on the phone if you're free this week, no pressure.

That takes me like five minutes to write. But it reads like I spent an hour looking into their company. Because I kind of did. OpenClaw just did it at 3 AM while I was sleeping.

My reply rate on these emails is around 34 to 35 percent. I track it obsessively. The average for cold email is somewhere between 1 and 5 percent. That entire difference comes down to one thing: I know more about each person before I reach out. That's it. That's the whole thing.

Results After About Six Months

I don't wanna be one of those people who posts numbers without context and makes it sound like everything is perfect. It's not. My render PC crashes constantly. A client last week told me to "make it feel more alive" with zero additional explanation. I got in an argument with my colourist about crushed blacks on Thursday. He was right. I just didn't want to admit it.

The lead gen stuff though, that's legit different now.

I got back 10 to 15 hours a week. I tracked it for two months to make sure I wasn't just imagining it. Those hours go to actual production work or, sometimes, just going home at a normal time. Which felt weird at first.

About 70% of high-priority leads turn into at least a conversation. Not a contract, a conversation. Of those, maybe one in four becomes a real project. Compare that to the old days when I'd email 50 people to get one meeting.

Revenue is up around 40% year over year. Not all of that is OpenClaw. I got some referral work and one inbound lead from Google. But at least half the new clients in the last six months came through the OpenClaw pipeline.

And I don't dread Mondays anymore. That's the part that doesn't fit in a spreadsheet but matters the most to me. There's actual space now to think about creative work, to develop new skills, to actually pursue that documentary idea I've been kicking around for two years. That wasn't possible before because I was always worried about where the next project was coming from.

Quick Notes If You're Thinking About Trying This

If you're a freelancer just starting out, skip this. Go shoot stuff, meet people, hang out at Remedy on 124 Street or Rosso in Kensington, go to AMPIA events. You need relationships right now, not automation.

If you've been at it for a few years and you've got a small team and you're just tired of prospecting, then yeah. Worth looking into.

Setup took me a full weekend plus some evenings. You need to be comfortable with command line and API keys. If that's not you, hire someone. I've seen people offering OpenClaw setup as a service, probably $500 to $1,000.

Cost: OpenClaw is free (open-source). I pay about $150 to $200 a month in API fees. $100 for Sales Navigator. HubSpot free tier. So like $250 to $300 Canadian total. A part-time VA doing the same research would cost ten times that. Easy.

I'm currently experimenting with using it for competitive monitoring too. Keeping tabs on what other video companies in Edmonton and Calgary are posting, what work they're promoting. I also want to try having it draft proposals but I'm not sold on that yet. Maybe as a rough starting point I'd completely rewrite. Anything going out to a client needs to sound like me.

Bigger picture: this stuff is gonna be normal in a year or two. Most production companies here aren't doing it yet which is nice for me but that won't last. The shops that figure out how to use AI for the boring stuff while keeping the actual creative work human are gonna be fine. The ones that try to automate the creative parts too, I dunno. I don't think that goes well. But I also thought OpenClaw was a waste of time six months ago so what do I know.

If you're in production or any creative business in Alberta and wanna talk about this, hit me up. Not selling anything. I make videos, not software. But someone helped me out when I was starting and I try to return the favour when I can. Coffee's on me.

FAQ

How technical do I need to be?

If you've ever used the terminal for anything you can probably handle it. I'm a communications grad, not a programmer. I just have patience for documentation. If that's not you, hire it out.

Other CRMs besides HubSpot?

Yeah. Pipedrive, Salesforce, probably others. I've heard Pipedrive is the easiest to connect. If your CRM has an API it should work.

Isn't it creepy to know that much about someone?

Everything it finds is public info. Websites, LinkedIn, news articles. A good salesperson looks at the same stuff, they just spend way longer on it. Just don't be weird about it in the email. Stick to professional stuff.

Monthly cost?

$250 to $300 Canadian. API fees plus Sales Navigator plus whatever CRM you use.

Can it send emails?

It can. Don't let it. Write them yourself. Trust me on this one.