Buying Signals for a Two-Person B2B Startup: Free First, Tool Later

For a two-person B2B startup, the most affordable buying-signal tool is no tool at all, at least for the first 100 conversations. The signals that matter early are free or close to it: people who already know you, people who react to your posts, your competitor's followers, and companies hiring for the role you sell into. Pay for a tool once you know which of those signals produces replies and your list has outgrown what one person can check in a week.

Our own LinkedIn numbers show why targeting beats tooling at this stage. Between June and September 2026 we sent 836 invitations to people picked by signals, with no note on the invitation. 40% accepted, and 25% of the people we then messaged replied (HeyReach data, pulled 1 October 2026). The invitation itself said nothing, so the lift came from who we picked.

Why a two-person team should not start with a tool

Start with the arithmetic. One person doing product, support and sales signs maybe one or two clients a month. Plan around that number, or make one of the two people own revenue full time.

Then the hours. Your evenings can produce features. They cannot produce 25 conversations a day, and whoever owns revenue needs every morning for outreach.

The third reason matters most. The first 100 conversations are research more than sales: they tell you what to build and what to charge. No tool skips that part. A tool bought before those conversations automates a message nobody has tested, so you end up scaling a guess.

If you are not yet sure the idea is real, test it before you build. One landing page, one clear sentence, and 200 to 300 manual messages to the people you think need it. If nobody replies to a human, nobody is going to reply to your product.

The signals worth working by hand

Four of these five sources cost nothing but time. The fifth costs about $72. They are ordered warmest to coldest, so work them top down.

Your own post engagement is the most underrated list a small team has. These people chose your thinking on purpose, and the list refreshes every time you publish.

Competitor followers need a hard filter. A category leader in our space has 24,061 followers, and pulling all of them through a scraping service costs about $72. Keep buyer titles in your target geography, drop the vendor's own staff and every recruiter, and about 10% survive. That leaves roughly 2,400 people who follow your category by choice. The full play is in your competitor's followers are your warmest leads.

Hiring data is often free at the source. When we checked six enterprise accounts, four published their whole job board on the open web. Read the role, not just the headcount. A new sales role points to a sales problem, and that only matters if you solve it. More in how to turn job posts into pipeline.

Funding alerts and keyword likes sit below all five. They are the easiest signals to buy, and that is why they are crowded: dozens of vendors read the same announcement on the same morning. We covered that in signal saturation.

A weekly routine two people can keep

1. One person owns revenue. Helping with sales is not enough. Product startups and service firms tend to fail the same way, with nobody owning revenue full time.
2. Keep the list small. A few thousand people you would be happy to close beat fifty thousand you would not.
3. On Monday, read the signals. Go through the five sources for the accounts on your list and mark who moved.
4. Every day, write 25 messages across LinkedIn and email, yourself. It takes under two hours.
5. Log the signal behind every reply. After a month the log shows which signal is worth paying for, and that becomes the brief for your first tool.
6. Once a month, reopen the quiet ones. The person who said "later" in spring may have a new role, a new budget or a post from last week by autumn. Lists rot fast (the last vendor file we measured was 22% out of date), so check before you write.

The full stage-by-stage version is in signal-based selling, the complete system.

What our numbers say about targeting

Neither campaign put a note on the invitation. "Replied" counts any answer, positive or not, from the people we messaged after they accepted. Source: HeyReach, pulled 1 October 2026.

Without a note, acceptance comes down to who you invite and how your profile reads. Signal-picked strangers accepted at 40%. For context, Expandi's 2026 benchmark across 13.2 million connection requests puts average acceptance at 28.5% and message replies at 10.4% (source).

The warm audience accepted more often and then replied at the same rate. Warmth helped at the door. Inside the conversation, the message and the offer did the work, so the first 100 hand-written conversations are where you learn the part no signal can supply.

The filter matters as much as the signal. In production on 21 August 2026, our system evaluated 161,072 companies and 8,689 passed the ICP filter, about 5%. Most of what any signal feed shows you is not your buyer. More in more signals won't fix your pipeline.

When to pay for a signal tool, and which kind

Pay once three things are true: your reply log shows which signal works, your list has outgrown one person's weekly check, and you want the same accounts watched every month instead of pulled once. Then choose by your situation.

• Your buyers are local businesses on Google Maps. Skip signals. Scrape the businesses, pull the info@ address from each website (a small language model does it for about $0.50 per 500 websites), write to info@ first, then to employees, with LinkedIn in parallel where it exists. Repeat every three months with a short, no-brainer offer. Our campaigns to these lists average a 4 to 6% interested reply rate, mostly because those inboxes do not get hundreds of cold emails a day.
• You already have the list and only need to send more. A LinkedIn sending tool such as Waalaxy or HeyReach covers it. It sends messages. It does not decide who deserves one today.
• Someone on the team enjoys building tables. Build your signals in Clay. You pay in credits and in the hours it takes to keep the tables running.
• You want the people found, your accounts watched every month and a draft waiting for approval. That is the job B2B Signals does, either with you (you approve and send) or for you (we run the outreach and you take the calls). A person approves every message before it leaves.
• You are solo and have not had 100 conversations yet. Do not buy any of these, ours included. Work the free signals by hand first.

Mistakes small teams make with buying signals

Buying the coldest signal first. A funding round is public, so you are message 40 that week. If you use funding at all, wait 30 to 60 days, when the budget is being spent and the inbox has gone quiet.

Treating the signal as the message. "They started a new role" is a database row. The role has to connect to a problem you solve, or it is noise.

Treating a follow as fresh intent. A follow can be three years old. Use follower lists to build the audience, then act first on the people who also liked or commented in the last month.

Letting AI write the first 100 messages. Prospects feel it from the first word, and you skip the part where you learn what makes people reply. Write first, polish after.

Sending once and never looking again. The account that ignored you in March is a different account by September.

Frequently asked questions

What is an affordable buying-signal tool for a two-person B2B startup? For the first 100 conversations, you do not need one. The early signals, from people who know you to companies hiring for your buyer's role, are free or nearly free to work by hand. Buy a tool once you know which signal produces replies.

How many outreach messages should a small team send each day? About 25 a day across LinkedIn and email, written by hand. That takes under two hours and leaves time for product and calls.

Do I need a note on a LinkedIn connection request? Not to get accepted. Our 836 signal-picked invitations from June to September 2026 carried no note, and 40% were accepted.

When is a signal tool worth paying for? When your reply log shows which signal works, your list has outgrown one person's weekly check, and you want the same accounts watched every month.

Should a small startup buy intent data? Usually not at the start. Intent scores describe accounts, while a small team needs a named person with a reason to talk now. See intent data vs buying signals.