How to Find Your First 100 Users on X and LinkedIn: A 30-Day Plan

By The Odeist AI team · 2026-10-14 · 9 min read

Short answer: your first 100 users on X and LinkedIn come from three places, in this order: people who already know you, strangers posting about your problem right now, and people who watch you build. Spend days 1 to 3 messaging your network one person at a time, weeks 1 and 2 on a daily 30-minute search-and-reply routine, week 3 on posts made from what users tell you, and week 4 on cutting whatever brought nobody. Log every conversation in one sheet, so you know what actually worked.

Where do your first 100 users on X and LinkedIn come from?

A launch post reaches your followers, and on a young account those are mostly friends and other founders. Users come from three pools, and each needs a different move.

Work your target backwards. Say you sell a tool that writes CRM notes from sales calls, and you guess that one in four people who talk to you will try it. Then 100 users means about 400 real conversations, or roughly 15 a day for a month. Your ratio will be different. The point stands: you need a daily conversation habit, not a better launch post.

What should you set up before day 1?

One hour, the day before you launch:

1. Fix both profiles. One line on who you help and what changes for them, a real photo, and a link that goes straight to sign-up. Our playbook for getting customers from X replies covers the profile in detail.
2. Pin one post on X and add it to your LinkedIn Featured section. What it is, who it is for, one screenshot, how to try it. Everyone who clicks your name from a reply lands here.
3. Start a tracker. A spreadsheet with: name, network, source (network, a search phrase, or a post), first contact, replied, tried it, still using it after 7 days, and their exact words.
4. Write ten problem phrases the way a customer says them, not the way your landing page does. The phrase table in our LinkedIn and X leads guide is a good start.
5. Decide your offer: "I'll set it up with you in 15 minutes." It gives a maybe a reason to say yes.

Days 1–3: who should you message first?

Not your feed. Your list. Go through your LinkedIn connections and X mutuals and write down 50 people who either have the problem or work next to someone who does. Then message each one by name.

The strong one names a problem they have, asks for something small and specific, and offers your time instead of asking for theirs.

For people who are not buyers, ask a different question: "Who is the one person you know who still does this by hand?" An introduction from someone they trust is usually the warmest lead you will get all month.

A sample three days:

• Day 1: publish your launch post on X and LinkedIn, written separately for each. Send 20 personal messages. Reply to every comment.
• Day 2: 20 more messages. Set up everyone who said yes, on a call if possible. Log each one.
• Day 3: the last 10 messages, and one follow-up to anyone who said "later". Then start the search routine below.

Weeks 1–2: how do you find strangers who need it right now?

Your network runs out in days. Strangers do not. Give this 30 minutes a day, at the same time:

In launch weeks, the best searches are about switching and timing, not just the problem. For the CRM-notes example:

• "alternative to" [competitor] -filter:replies
• "switching from" [competitor]
• "is there a tool that" ("call notes" OR "sales calls")
• "just hired our first" (sales OR SDR)
• "how do you" "call notes" since:2026-11-01

On X, quotes match an exact phrase, OR matches either word, a minus sign excludes, and since: with last week's date keeps results recent. On LinkedIn, paste the same phrase, filter to Posts and read the comments too: the people writing "same problem here" are a second list.

How to write the reply is covered in the X replies playbook: show you read the post, give one useful thing, end with a question.

This routine is also the habit that tends to slip once early users start asking for fixes. That is the step Odeist AI is built for: its X and LinkedIn agents run your searches on a schedule and draft a reply to each match in your voice, for you to approve from your phone.

Week 2: how do you turn a reply into an active user?

A reply starts a conversation. A user needs a second step. When someone answers you, move to a private message and offer to do the work for them.

Paul Graham calls recruiting users by hand "the most common unscalable thing founders have to do at the start" in Do Things That Don't Scale. Setting people up yourself also gives you something no dashboard will: you watch where they get stuck.

Seven days after each sign-up, check the tracker. Anyone who has not come back gets one question: "What stopped you?" Their answer is either a bug, a missing feature, or a sign they were never the buyer. All three are useful.

Week 3: what should you post when nobody follows you yet?

By now you have something most launch posts lack: real conversations. Post from them, three times a week, while replies stay your main work.

Write each idea once and adapt it for each network, which we cover in how to schedule posts to X, LinkedIn and Instagram at once. Use your users' words, never their names, unless they say yes.

Week 4: what should you double down on?

Every Friday, sort the tracker by source and answer three questions:

1. Which sources brought people who are still using it? Not sign-ups. Users still there after a week.
2. Which search phrases brought nothing? Delete them. Replace them with phrases your users actually said on calls.
3. Who are your five happiest users? Ask each one: "Who else do you know with this problem?" and whether you can quote a sentence of theirs.

Then plan month two around the top two sources. Expect that to mean fewer, better searches and more time in the conversations they produce.

Which mistakes stall the first 100?

• Launching to a feed instead of to people. Likes on a launch post are not users.
• Searching for your category name. People describe their problem, not your product.
• Not logging the source. Without it, week 4 is guesswork.
• Stopping the routine when you get busy. The week you ship fixes is the week the pipeline empties.

How can Odeist AI run this playbook with you?

Odeist AI is a team of AI agents for startups and small teams. For the first 100 users, it takes on the parts that slip:

• It learns your business in about a minute. Paste your website and it writes six documents every agent works from: Product, Buyers, Competitors, Voice, Visual style and an Action plan, week by week. It can also learn your voice from your own posts.
• It finds the conversations. Your X and LinkedIn agents search for the keywords you give them, such as your problem phrases, and draft a reply to each match in your voice.
• You keep the final say. Each agent has one switch. "Ask me first" sends every draft to WhatsApp, Telegram or Slack with Approve, Edit and Skip. Once the drafts sound like you, switch to "On its own" and it posts by itself. Skips and edits become rules the agent follows. Our guide to what an AI marketing agent should do without you explains how to choose.
• It helps with launch day. The Launch kit writes an X thread, a LinkedIn post, a launch email, a press release, three taglines and a plan from a week before to a week after. The Hacker News agent drafts your Show HN, which you post yourself.
• It reports back. Every Monday the team reviews the week and says what it will change.

A reply uses 2 credits, and plans start at $29 a month; see pricing. New accounts get 7 days of Growth with 1,000 credits, and no card.

"When Chivvy was two people doing everything, Odeist AI was our social media team. It found people asking for what we’d built, and the Odeist AI team treated our launch like their own." — Chivvy

Start free, or get a free growth report to see what your marketing is missing, in about a minute.