Founder-led marketing on LinkedIn and X: the cadence, formats and what to measure

Founder-led marketing means the founder becomes the company's main publisher, posting on LinkedIn and X under their own name, on a schedule, about the problem the product solves. It suits business-to-business (B2B) startups because 95% of business clients say they are not actively seeking goods or services at any given moment, and more than 40% of B2B deals stall on internal misalignment. Start with 2 LinkedIn posts a week.

What is founder-led marketing and does it work for B2B startups

Founder-led marketing means the founder becomes the company's main publisher. The founder posts on LinkedIn and X under their own name, on a schedule, about the problem the product solves. The company page plays a supporting role. For a business-to-business (B2B) startup selling to a buying group, this approach works when the posts reach the people who are not yet shopping and the people sales never meets.

The buyer numbers behind it

Most of your market is not buying today. The 2024 Edelman and LinkedIn report found that 95% of business clients say they are not actively seeking goods or services at any given moment. As an example, not a benchmark: if your target list holds 100 accounts and 95% are out of market, about 5 are buying now. Posting keeps you known to the other 95 until their moment comes.

Deals also stall inside the customer. The 2025 Edelman and LinkedIn report, based on a survey of 1,934 global business executives, stated that more than 40% of B2B deals stall due to internal misalignment within buying groups. It named hidden buyers as a major cause: internal stakeholders who significantly influence business purchases even if they are not the primary users of a product or service.

Buyer type	What the 2025 report found	What founder posts do (our reading)
Hidden buyers	71% have relatively little or no interaction with sales	Reach people a sales call never will
Hidden buyers	55% use thought leadership to vet vendors	Give them material to judge you by
Target buyers	56% use thought leadership to vet vendors	Back up what the champion says internally

The report also found that hidden buyers consume just as much thought leadership content as target buyers. That reading time is your audience.

Why the founder, not a hire

Early growth is a learning problem. Matt Lerner put it plainly in First Round Review: "Ultimately, founders need to be the ones that figure out how their business is going to grow". The same piece made the trade-off clear: strategizing, hiring and delegating are great ways to execute, but they are very slow ways to learn.

In our reading, publishing fits that logic. Each post tests a claim about the customer's problem in public, and the replies tell you which framing lands. A first marketing hire can later scale what the founder has proven, the way a growth loop scales a step that already converts.

This week: pick one problem your buyers argue about internally, write one post a day on it for five days, and log which posts draw replies from finance or operations titles.

How often should a founder post on LinkedIn and X

The published guidance does not agree, and the gap matters most on LinkedIn. Buffer's study is the one with a large dataset behind it: its data scientist analyzed more than 2 million LinkedIn posts from 94,000+ accounts and compared each account's busy weeks with its quiet ones. Hootsuite's ranges, by its own account, come from a theoretical approach built on best practices and conversations with social media managers.

Buffer measured the added impressions per LinkedIn post at each weekly band, from 1,182 at 2 to 5 posts to 16,946 at 11 or more.

Extra LinkedIn impressions per post compared with posting once a weekimpressions per post

Extra LinkedIn impressions per post compared with posting once a week

Item	Value, impressions per post
2 to 5 posts a week	1,182 impressions per post
6 to 10 posts a week	5,001 impressions per post
11+ posts a week	16,946 impressions per post

Source	LinkedIn	X	Basis
Buffer	2 to 5 posts per week	three to four times every day	Per-post study (LinkedIn); guidance (X)
Hootsuite	1 to 2 times per day	2 to 3 times per day	Best practices and expert interviews
Hootsuite, what businesses did	5.5 times per week	18.1 times per week	Measured in Q1 of 2025

On LinkedIn, Buffer found that moving from 1 post a week to 2 to 5 is like flipping a switch. Each higher band still added reach per post, but the improvement between 6 to 10 posts and 11+ posts is even smaller than the leap from 1 post a week to 2 to 5. The condition is quality: Low-quality content posted frequently won’t yield great results.

On X, Buffer's guide calls three to five posts per day, spread out over time a sustainable range, and says a mix of original posts, replies, and reshares keeps your account visible. We read that as permission to count replies toward the daily number.

A floor a founder can keep

In our reading, a founder who also runs sales and hiring should commit to the bottom of Buffer's LinkedIn band and grow from there:

1. LinkedIn: 2 posts a week, every week. Pick two fixed days. As an example, 2 posts a week for 10 weeks is 20 posts, enough to compare against your own early weeks.
2. X: one original post a day plus replies to people your buyers follow. This sits below the published ranges. We think consistency at a lower number beats a higher target you drop in week three.
3. Move up one band only after four steady weeks, and only if drafts are not getting thinner.
4. Judge per-post impressions against your own baseline, the way Buffer compared each account with itself. Treat each cadence change as a test, using the steps in how to run a growth experiment.

What the LinkedIn feed rewards and how that shapes each post

LinkedIn does not show posts in the order they were published. By default the feed is sorted by "Top Updates", and those updates are picked from each member's own activity: the accounts they like, share and comment on. Every post a founder writes starts with the reader's history.

Sprout Social describes four factors behind what gets shown: post quality, post testing, member activity and relevancy to users. The quality check screens out spam such as "Chain letter-type content that requests likes, reactions, and shares." A post that passes is tested on parts of your network, with reach expanding or contracting over time based on engagement quality, relevance and sustained interaction. On relevancy, first-degree connections see your content first, and the connections you regularly interact with are also prioritized.

Factor	What it means for a founder's post (our reading)
Post quality	No engagement bait. Ask for an opinion, never for a like.
Post testing	The first readers decide how far the post travels, so write for your existing network.
Member activity	Readers who already engage with your topic are the ones the feed picks.
Relevancy	Name the topic, company or person plainly so the post is matched to the right readers.

Format follows the reader

Sprout Social says the algorithm favors posts that align with how members engage rather than prioritizing a specific post format. Buffer's frequency study covered text, image, video and carousels or document PDFs without breaking them out individually. In our reading, a founder should pick the format their buyers already respond to and test it, not chase a format rumored to be favored.

The first lines

Questions are often used as a hook to get followers to read a longer post. A template you can use this week:

1. Line one: a specific claim from your own work.
2. Line two: the tension or cost behind it.
3. Last line: one question your buyer can answer from experience.

Replying to comments

Because regular interaction raises your priority with a connection, we treat replies as part of the post. Block a fixed window after publishing, reply to every comment with a full sentence, and ask a follow-up where the commenter raised a real point.

Measure it with engagement rate per post, which Buffer defines as the percentage of viewers who interacted. As an example, not a benchmark: a post with 1000 impressions and 20 interactions has an engagement rate of 2%. Log it per post, next to whether you replied to every comment. For a way to compare the two in a structured test, see how to run a growth experiment.

What should a founder write about

Most people who read a founder's post are not shopping that week. The 2024 Edelman and LinkedIn report, which surveyed nearly 3,500 management-level professionals across seven countries, found that 95% of business clients say they are not actively seeking goods or services at any given moment. So a post has a different job. It should change how a reader thinks about a problem before they ever need a vendor.

That same report says good thought leadership makes buyers reexamine their assumptions. The 2025 edition found that buyers gravitate to content that questions assumptions and invites new ways of thinking. It also found that, just like target buyers, hidden buyers use thought leadership to evaluate vendors. Keep the pitch out of the feed.

Three pillars to rotate

Pillar	Purpose (our reading)	Example topic	Source material
Opinions backed by evidence	Challenge an assumption your buyer holds	Why annual contracts slow our smallest customers' rollout	Customer calls, support tickets, your own churn data
Build-in-public numbers	Show how decisions get made, with real figures	What changed when we moved our trial from 14 to 7 days	Your dashboard, experiment logs, board updates
Teardowns of the market	Explain how the category works, with no pitch	How three competitors price their entry tier	Public pricing pages, changelogs, job posts

For the third pillar, our pricing page teardowns show the format.

Where the material comes from

Close's co-founder Steli Efti described how his early content drew on experience and expertise by running this sort of outsourced sales agency on demand for tech startups. His topics often came straight from customer conversations: a lot of the topics that I talked about literally were videos that I recorded right after a call with a founder. Writing was his bottleneck. He fixed it when a collaborator told him to hit record on your phone and talk me through the draft.

That gives a weekly routine you can start now:

1. Record a two-minute voice note after every customer or prospect call. Name the problem they described and what you told them.
2. On Friday, sort the notes into the three pillars.
3. Pick the strongest note in each pillar and hand it to whoever edits, or edit it yourself.

As an example, not a benchmark: 10 calls a week gives you 10 notes. If 3 of them hold up, that is a week of posts from conversations you were already having.

Building in public still works if you change what you share

Building in public used to be a revenue story. One founder wrote that publishing Stripe-verified monthly recurring revenue (MRR) figures attracted financial and acquisition interest that ultimately led to the sale of my previous software business, FeedbackPanda. The same founder now calls the practice dangerous. The reason given is that AI coding agents let a business-minded person turn your posts and a free trial into a rival product within days or weeks.

The old rule was a ceiling. Founders stopped reporting their financial metrics and sharing specifics about what exactly they were building around $20,000 to $30,000 in monthly recurring revenue. That founder's view now is that the threshold has effectively collapsed to zero. A February 6, 2025 post had already set a lower line: stop sharing MRR over ~$10K/month.

What replaced the revenue screenshot

One 2026 guide frames the change as a fork: the 2019 performance-in-public version is dead; the 2026 workflow-in-public version compounds. The workflow version includes a weekly 30-60 second screen recording of one new thing shipped that week and a monthly retro summarizing what shipped, what you learned, what is coming.

The founder who sold on public MRR now shares tripwires, meaning things I didn’t expect, and what happened when they occurred. The list also includes operational experiences, like an interesting customer service conversation or fixing a particularly gnarly bug.

The line sits at detail. Saying customers like a new integration is fine. Tying that integration to a count of new customers on a named plan tells any reader, human or machine, how the business makes money.

The share-or-keep-private list

This is our sorting of the points above. Use it as a checklist before each post this week.

Area	Share (our reading)	Keep private (our reading)
Metrics	What surprised you and what you changed	Revenue, customer counts, expenses
Customers	An anonymized support conversation and its lesson	Names, plan mix, which segment pays
Product and roadmap	A short demo of a feature already shipped	Upcoming features in buildable detail
How you work	Process, prompts, time spent	System architecture and data pipelines

The founder's own list put system architecture among the topics no longer worth sharing.

Run one test on every draft. If a stranger with a free trial and this post could estimate your revenue or rebuild the feature, cut the number or the specifics and keep the lesson. For an internal check on whether a shared experiment worked, measure it against your own retention data rather than posting the result.

A weekly cadence that fits in about two hours, with one long piece turned into many

Founders rarely run out of ideas. They run out of time. The build-in-public guide puts it plainly: without the stack, the same output costs 2-3 hours per day, and it ties that cost to founders quitting by week 6. Our answer is to batch the work into one recording per week and cut that single piece into everything else.

The weekly template

The minutes below are an example, not a benchmark. Adjust them to your own week.

Block	When	Example minutes (our example)	What you end up with
Capture	Daily, after calls	10	A running list of questions, objections and things that just happened
Record the long piece	Monday	20	One voice memo or video talking through a single idea from the list
Edit into a long draft	Monday	20	A newsletter issue or LinkedIn article built from the transcript
Cut into posts	Tuesday	20	Short LinkedIn posts and X posts, each carrying one point from the draft
Reply to comments	Same day as each post	20	Answers that often seed next week's capture list
Review	Friday	10	A note on which post drew replies and which idea goes next

In this example the week costs 10 + 20 + 20 + 20 + 20 + 10 = 100 minutes, which keeps it under two hours.

The capture block can be tiny. The guide describes a journaling habit of 2 minutes per day, captures the lessons that fuel monthly retros. A notes app is enough. For the newsletter cut, our newsletter growth playbook covers how to turn readers into subscribers.

Mistakes that break the habit

• Starting from a blank page. Steli Efti of Close described scheduling a 30 minute first draft of a blog post and seeing it slip "until like a week past or two." Talking first and editing second avoids the stall.
• Becoming the editor's bottleneck. Handing drafts to a helper only works if the input arrives. At Close, the helper sometimes waited for two or three weeks to get the draft. Fix the recording slot on the calendar and send the raw file the same day.
• Judging the system too early. For its weekly demo rhythm, the guide says the compounding starts at week 8 and becomes visible at week 12. We think the same patience applies here, so commit to a full quarter before changing the template.
• Batching replies days later. In our view, comments are where the next topics come from. Answer them on the day you post, while the thread is still active.
• Writing a new piece for every channel. One idea, cut several ways, is the whole point. If a post needs fresh research, it belongs on next week's capture list.

How to measure founder-led content

Track three numbers each week: profile visits, inbound conversations and signups attributed by asking. Profile visits show that a post made someone curious about the founder. Inbound conversations count direct messages, replies and meeting requests that a post started. Signups attributed by asking connect posts to pipeline, so that is the number to get right.

Fix the attribution question before you trust it

Self-reported attribution has two problems. Answers default to whatever is easiest to click, and answers come back too thin to use. On first-option bias, Ruler Analytics found 47% of the leads selecting the first option in the list were inaccurately attributed for one client. On thin answers, it tested over 350 leads of its own and found 72% of our total inbound demos either provided vague details or didn’t provide any detail at all.

A drop-down undercounts founder content because its options name channels, not moments. Someone who saw a post forwarded in a private chat has no option that fits, so they pick the nearest or the first. Ruler groups that kind of traffic with word of mouth, dark social etc.

Four changes address both problems:

1. Open text first. Ask "How did you hear about us?" as a free-text field, so nothing is pre-filled for the first-option habit to grab.
2. A shuffled list only as a fallback. If you must use a list, randomize the order for each visitor and include an option for a post by the founder.
3. A second ask on the first call. Ruler reports that the majority of responses were one word answers, so ask again live and write down the full answer.

Self-report still works alongside other data. Ruler reports that combining its tracking tool with the form field gave it an 81% accuracy on lead tracking.

The weekly sheet

Column	What to enter
Week and topic	The main theme posted that week
Profile visits	From the platform's analytics
Inbound conversations	Count of messages and meeting requests started by a post
Founder-post signups	Signups coded founder post
Good week? (our rule)	Yes if a measure beat the founder's own trailing median

When to keep or cut a topic

Judge each topic against the founder's own baseline, not anyone else's numbers. We keep a topic when it produces a good week on inbound conversations or founder-post signups in two of its first 3 posts. We cut it after 3 posts with no lift on either. A rise in profile visits alone justifies one more post on the topic, not a permanent slot.

The 90-day founder content playbook with a worked example

Matt Lerner, co-founder of SYSTM, puts the starting point plainly in First Round Review: "Ultimately, founders need to be the ones that figure out how their business is going to grow." He adds that strategizing, hiring and delegating "are very slow ways to learn" (First Round Review). The playbook below is our own method for doing that learning in public over 90 days.

The steps, in order

Step	What to do	Who	Output	Signal it worked (our reading)
Set the baseline	Record current profile visits, inbound messages and signups that cite the founder	Founder	One baseline row in a sheet	The row exists before the first post
Fix the profile	Rewrite the headline and featured section around one buyer problem	Founder	A profile that says who it is for	Visitors reply with the problem in their own words
Run the weekly cadence	Use the two-hour routine from earlier in this piece	Founder, with a helper for formatting	A steady run of posts	No week missed
Open conversations	Reply to every comment and send a direct message to engaged buyers	Founder	Booked calls	Calls come from people who match the customer profile
Ask how they heard	Add the self-reported attribution question to signup	Whoever owns the form	A weekly count of attributed signups	Answers name the founder or a post
Review monthly	Keep topics that drew conversations, drop the rest	Founder	A shorter topic list	Conversations rise month on month

A worked example

Take a seed-stage B2B founder selling to finance teams. These are example inputs, not benchmarks.

• First month: 10 posts, 100 profile visits, 3 conversations, no attributed signups yet.
• Second month: 10 posts, 1000 profile visits after one post about a month-end close problem travels, 10 conversations, 3 attributed signups.
• Third month: 10 posts, 1000 profile visits, 20 conversations, 5 attributed signups.

The output stayed flat. What moved was the topic mix after the monthly review. In the third month, 20 conversations from 1000 profile visits is a 2% visit-to-conversation rate, and that is the number we would track next quarter.

Read the signups through the caveats in the measurement section. Self-reported answers undercount, so treat 5 as a floor.

When to add a second voice or a hire

We would add a second voice when conversations outrun the founder's calendar. In this example, that point is around 20 a month, when replies start to slip past a day. The second voice is usually a cofounder or an early engineer who can post on the product side.

Hire only once the topic list has stopped changing for a full month. At that point the founder knows what drives replies and can hand over a working system rather than a guess. Hand the hire the sheet, the topic list and the attribution question, and keep the founder posting.

What this page does not cover

This page does not cover paid promotion, channels other than LinkedIn and X, or how to run the company page. The share-or-keep-private list is our editorial reading, not legal advice on what a startup may disclose publicly.

Related on Race to Scale: How to Grow a Newsletter: Recommendations, Swaps, Paid Ads, Referral and What Each Subscriber Costs; Referral program examples that worked, and the mechanics behind them; Waitlist launch strategy: how the referral queue and the gated waitlist work, what converts, and when a waitlist backf.