If you're sitting on a database of 30,000 contacts and still struggling to fill your pipeline, the problem isn't that you don't have enough names. The problem is that none of those names feels like a clear match. You built a list for everyone, and now no one on it feels worth calling. That's a specific kind of stall, and I see it happen to business owners all the time.
Sarah and I got into this on Episode 63, starting with a story from a recent networking meeting that stopped the room cold. You'll come away with a clear test for whether your ICP is too broad, a way to pressure test it before you build a single list, and a concrete reason why a smaller, tighter database will actually move your pipeline faster.
This post is based on Episode 63 of Revenue Rewired | Don't Overlook the Basics: The ICP Fundamentals That Are Quietly Costing You Revenue.
If you'd rather listen than read, find the full episode on Apple Podcasts, YouTube, Spotify, or Amazon. It's worth your time.
What a Networking Meeting Revealed About B2B ICPs
I'm part of a networking group, and recently they tried a format called "Give and Get." In the first half, you explain what you do, who you serve, and what your ideal client looks like. In the second half, you have to rebuild the whole thing through LinkedIn Sales Navigator, naming specific industries, titles, geographies, and headcounts.
The room handled the first half without much trouble. The second half exposed everything. Business owners who talked fluently about their company fell apart when they had to fill out actual filter fields. The ones who were vague in part one were completely stuck in part two.
One person listed their ideal client roles as owner, CEO, CTO, product developer, marketing manager, and office manager. I asked: if you walked into a room with all of those people at the same time, would you talk to the product developer the same way you'd talk to the owner of a $25 million company? They said, "Of course not." That's the whole point. If you wouldn't approach them the same way, you're not defining one ICP. You're defining six.
Why a Bigger Contact List Is Actually Your Biggest Problem
There's a version of this mistake I've made myself. Earlier in StringCan's history, I'd put "5 million to 250 million" into Sales Navigator and feel good about the result. I'd get enterprise companies that looked impressive on paper. But they weren't really who we were built to serve, and I was spending energy on a pursuit that wasn't right for us.
The pull to go bigger and give the tool every possible filter comes from FOMO. You don't want to miss anyone. But here's what actually happens: you end up in a metaphorical room with 35,000 people wearing name tags, and you have nothing specific to say to any of them. You feel paralyzed, not energized.
I spoke with a client recently who said he only needs one new client a month to hit his goals. I asked him why he thought he needed a database of 312,433 people. He didn't have a good answer. If your conversion rate is decent, a list of a thousand people sharing similar psychographics, the same motivations, pain points, and role type, will carry you for a year or two of proactive outreach. And you'll know exactly what to say to them.
What a Dating App Has to Do With Your Sales Pipeline
LinkedIn Sales Navigator, Zoom Info, and any prospecting database are matching tools. They tell you if someone looks like a fit on paper. They're not fit tools. They don't tell you what happens after you connect. That gap is where pipelines break down.
Sarah made an analogy in the episode that nails it: a dating app. You match on the criteria, connect, and then you show up and realize you didn't ask the right questions. The profile said one thing, the reality is something else. That's what happens when your ICP is built on form fields rather than on the actual characteristics of your best clients.
Sarah also flagged something I see often: owners using AI to develop their ICP, and getting back the same generic output their competitors are getting, because AI draws on the same public data everyone uses. AI can help you sharpen language and pressure test a pitch. It can't tell you what makes your best clients your best clients. That part has to come from looking at your last three favorite deals and finding the honest common thread.
Why This Engineer's 16-Step Plan Was Missing the First Step
I had a call with someone who was trying to figure out why people were visiting his site but not converting. He came in with a 16-point tracking strategy: pixels, funnel events, attribution models, the works. Then I asked if he had basic analytics installed. He didn't.
That's not unusual. Business owners with a technical background tend to engineer a complete solution before they fully understand the problem. The impulse makes sense, the capability is there, so why not use it? But if you don't have any data yet, you don't know what question you're trying to answer.
My suggestion: install the basics first, then track two things, where someone enters and where they exit. Give it 60 to 90 days. What you'll often find is that the issue isn't a missing tracking layer. It's that the content isn't connecting, or the messaging on the entry page isn't landing for the ICP you thought you were targeting. Fix that first, and a lot of the more complex architecture might not even be necessary.
FAQ
Q: How do I know if my ICP is too broad?
A: Go into LinkedIn Sales Navigator or even free LinkedIn and build a list using your current ICP parameters. If you end up with more than 10,000 results, or if you can't get it under 5,000 without removing companies you'd genuinely want to work with, that's a sign it's too broad. The goal is a list you could actually work through.
Q: How many contacts do I actually need in my database?
A: For B2B businesses, 500 to a few thousand highly matched contacts will outperform a list of 30,000. It depends on how many clients you actually need and what your realistic conversion rate is. If you need two or three new clients a quarter, you don't need hundreds of thousands of names.
Q: Does AI help or hurt with ICP development?
A: It can help you refine and pressure-test language. Training a GPT on your ICP draft and asking it to poke holes is a legitimate use. Where it falls short is originality: asking AI to build your ICP from scratch gives you what every competitor in your space is also getting. The specificity has to come from you.
Q: What does it mean to pressure test your ICP?
A: It means running your assumptions past people who know your business from different angles, current clients, past clients, and your team. You're checking whether how you describe your ideal client actually matches where your wins have come from, and identifying where you're filling gaps with assumptions rather than data.
Q: Where do I start with analytics if I have almost no data?
A: Install something basic, like Google Analytics, before anything else. Then identify one clear entry point and one clear exit point you care about, and watch what the data tells you over 60 to 90 days. Let the question you're trying to answer drive the tracking, not the other way around.
Ready to Pressure Test Your ICP?
At StringCan, we've spent 16-plus years helping B2B companies get clear about who they actually serve best. The ICP work isn't glamorous, but it shows up in every piece of marketing you produce, every sales conversation you have, and every campaign that does or doesn't convert. Getting it right is the foundation that everything else is built on.
If you want to talk through yours, reach out. Jay does this kind of pressure testing with business owners regularly, including in real time at networking events. Find Episode 63 of Revenue Rewired here, and if this topic connects, subscribe so you don't miss what's coming.
