How to find LinkedIn post ideas (no more blank page)

The blank page on LinkedIn isn't real. You already have dozens of post ideas — you just don't know where to look yet. Here's a three-step method, grounded in data on what actually performs.

In short

→ The method has three parts: content pillars, structured watch, systematic recycling. → The topics most correlated with engagement in our data: AI, personal branding, technology — with a median like count 2 to 4 times above the corpus average. → Your watch isn't for copying — it's for spotting angles that engage, then reframing them through your own experience. → One good post = at least 3 possible variations by format (text, carousel, long-form). → LinkPost's strategy agent and viral inspirations automate the watch and angle-generation parts.

The blank page is a method problem, not an inspiration problem

The blank page rarely happens because you have nothing to say. It happens because you open the editor without having decided in advance what you want to talk about.

The fix: never sit down in front of an empty post without a plan. Decide on your pillars once and for all, feed an idea bank continuously, and only start writing when you already have an angle in mind.

How to build your content pillars

A pillar is a broad topic where you're credible and your audience cares. For an HR consultant: recruiting, company culture, career paths. For an independent developer: tech, freelancing, tools. For a SaaS founder: product, growth, building in public.

Practical rule: 3 pillars maximum. More than that, you dilute your topical authority and the algorithm has a harder time categorizing you.

For each pillar, list 10 different angles. Example with the "AI" pillar:

→ concrete use cases in your day-to-day work → mistakes to avoid with LLMs → comparison of tools you've actually tested → experience report after X weeks of use → a polarizing opinion on the impact on jobs → step-by-step tutorial on a specific workflow → an open question to your community → a quantified result from an experiment → debunking a common misconception → a prediction on where this is headed in 12 months

10 angles × 3 pillars = 30 starting posts. And each post can spin out into multiple formats. That's months of content.

The watch: observe without copying

Your watch is about seeing what engages in your niche and asking: "What would I have to say about this, from my experience?"

Three-move method:

→ Follow 15 to 20 creators in your niche. Not to passively like, but to analyze what hooks: the openers, the formats, the questions people ask in comments. → Note the angles that land. Not the exact content — the angle, the structure, the type of proof they use. → Reframe with your real experience. Direct copying is obvious, performs poorly, and builds zero authority.

LinkPost's Viral Inspirations feature shows you posts performing in your niche, filtered by topic and format. You come in with clear signals instead of drifting through your feed.

To understand which topics perform across the network as a whole, the analysis of 516,000 LinkedIn posts gives a broad view that helps orient your watch.

Which topics actually perform on LinkedIn

This is where data helps you prioritize. Based on our corpus of 516,144 posts (30,843 profiles, data from 2020-2026), the topics most correlated with engagement are:

| Topic | Median likes | |-------|-------------| | AI / AI agents | 82 | | AI tools | 80 | | Personal branding | 64 | | Technology | 53 | | Community | 52 | | Personal growth | 43 |

These numbers are correlations, not guarantees. Creators posting about AI often already have an audience engaged on those topics. The overall corpus median sits around 18 likes — so AI topics perform 3 to 4 times above average, which is a strong signal even accounting for sample bias.

The key takeaway: a niche topic you genuinely master and ground in real experience almost always beats a generic one. Personal branding at 64 outperforms broader themes because creators who excel at it know exactly who they're talking to.

The detailed breakdown of topics by engagement level adds nuance here — separating volume effects (lots of creators on a topic) from actual engagement effects.

Observational data from LinkPost (corpus 2020-2026, 62% French-language content). Correlation is not causation. Methodology and biases detailed in the full analysis.

Recycling: making 10 posts out of one idea

One of your old posts performed well? Don't repost it as-is. Recycle it intelligently:

→ Turn the text into a carousel — our study measures 2.3× more median impressions for carousel format → Pick one single point and develop it into a 1,500+ character long-form → Flip the angle: from "here's what works" to "here's what I did wrong for two years" → Update it with a new data point or fresh field observation → Pose the same question to your community as a poll

A good post has a much longer life than its first run. And 80% of your future followers haven't seen your old content.

What to do, concretely

1. Define your 3 pillars this week. Write 10 angles per pillar in a doc or a note.
2. Set up your watch. 15 creators to follow actively, with a weekly review of what performed.
3. Keep an idea bank. Every observation, anecdote, result, or question deserves a line. Post ideas come from there — not from spontaneous inspiration.
4. After every strong post, immediately note 3 possible variations while the topic is still fresh.
5. To automate angle generation and niche watch, LinkPost's strategy agent suggests ideas calibrated to your style and pillars.

And if you want to turn your ideas into leads directly, LinkMagnet automatically delivers a lead magnet via DM to people who comment on your posts — particularly effective on "comment to receive" posts.

The starting idea matters. The method behind it matters just as much. Try LinkPost to structure your content strategy and predict the impact of your posts before you publish.