How to Write a LinkedIn Recommendation (40+ Examples)

August 18, 2026
14
min read
Virgile Donadieu
Lead Growth @Taplio

Key Takeaways

  1. Recommendations rarely create opportunities, they close them: they matter late in a decision, when someone already interested is looking for a reason to be confident.
  2. Three specific ones beat twenty generic ones: the credibility of the writer and the specificity of the evidence decide the value, not the count.
  3. Use a five-step structure: concrete opening, the relationship, one named quality, hard evidence, and who the person would be right for. Keep it to 100 to 150 words.
  4. Match the evidence to the vantage point: a manager can speak to growth over time, a client can only speak to outcomes, and a peer sees how someone works day to day.
  5. Name the angle when you ask: a vague request gets postponed forever, and writing one first is the most effective opener there is.

Most LinkedIn recommendations are unreadable. Not badly written exactly, just interchangeable: pleasure to work with, great team player, would hire again. Nobody finishes them, including the person who requested one.

The ones that work do something specific. This guide covers what separates them, a five-step structure with examples at every step, 8 recommendations by relationship, 15 more by role, how to ask for one, and how to say no when you do not want to write it.

What is a LinkedIn recommendation?

A recommendation is a short written endorsement from someone in your network that appears in a dedicated section of your profile, under your name and their relationship to you.

It is not the same as a skill endorsement. Endorsements are one click and mean almost nothing. A recommendation is prose, it carries the writer's name and headline, and it is visible to anyone who scrolls that far. You have to approve it before it appears, and you can hide or remove it at any time.

Do LinkedIn recommendations actually help?

Ask this on Reddit and you will get a hundred people saying nobody reads them. They are half right, and the half they get wrong is the half that matters.

Here is the honest version. Recommendations almost never create an opportunity on their own. No recruiter searches LinkedIn by recommendation count, and nobody has ever been hired because a former colleague wrote three warm paragraphs. As a top-of-funnel signal they are close to worthless.

Where they do work is late in a decision, when someone has already decided they are interested and is looking for a reason to be confident. A hiring manager comparing two shortlisted candidates, a prospect deciding whether to reply, a client checking a freelancer before sending a brief: all three land on your profile with a specific doubt, and a specific recommendation from a credible person answers it in a way your own copy cannot. You wrote your About section. You did not write that.

Which leads to the practical rule. Three or four detailed recommendations from people whose names carry weight in your field will do more than twenty generic ones. Volume is not the metric; specificity and the credibility of the writer are. A recommendation from a client describing the actual result you delivered is worth more than five from colleagues saying you were pleasant.

And if you write them for other people, there is a second-order effect worth knowing: your name, headline and photo sit on their profile permanently, in front of everyone who visits it.

How to write a LinkedIn recommendation in 5 steps

The structure below takes about ten minutes and produces something the person will actually want on their profile.

Step 1: Open with something concrete

Skip the throat-clearing. No "I had the pleasure of working with". Start where the story starts.

  • "We inherited a support backlog of 4,000 tickets. Priya cleared it in seven weeks without hiring anyone."
  • "I have managed 30-odd engineers. Tom is the only one who ever talked me out of a bad decision twice in the same week."
  • "I hired Lena to fix one landing page. She rewrote our entire onboarding sequence because she noticed the real problem was upstream."

Step 2: State the relationship and the context

One sentence. The reader needs to know how you know them and whether your opinion counts.

  • "I was Priya's manager for two years at Northbeam, where she led a support team of six."
  • "Tom and I worked side by side on the platform team for three years."
  • "I was Lena's client for eighteen months while she consulted for us on lifecycle marketing."

Step 3: Name the specific thing they do well

One quality, not five. A list of adjectives reads as filler; a single well-chosen trait reads as observation.

  • "Her strength is triage. She can look at a mess and tell you in ten minutes which twenty per cent of it matters."
  • "Tom writes the clearest technical documents I have read. Non-engineers on our leadership team could follow his architecture proposals, which is not normal."
  • "Lena asks the uncomfortable question early. It saved us two quarters of building the wrong thing."

Step 4: Back it with evidence

A number, an outcome, or a moment. This is the step most people skip and it is the one that makes the recommendation credible.

  • "Response time went from 26 hours to under 4, and it stayed there after she moved teams."
  • "When the payment service failed on a Friday night, Tom had a fix in production before most of us had seen the alert."
  • "Trial-to-paid conversion went up by a third in the quarter after her changes shipped."

Step 5: Close with a clear recommendation

Say who they would be right for. A generic endorsement helps nobody; a targeted one tells the reader whether it applies to them.

  • "Any company scaling a support function past its first ten people should be talking to Priya."
  • "If you are hiring a senior backend engineer who can also carry a room, hire Tom."
  • "I would work with Lena again tomorrow, and I have recommended her to three other founders since."

Length: 100 to 150 words. Long recommendations do not get read, and the discipline of cutting forces you to keep the specific parts.

8 LinkedIn recommendation examples by relationship

The angle changes depending on who is writing. A manager can speak to growth over time; a client can only speak to outcomes. Match the evidence to the vantage point.

1. Manager recommending a direct report

"Sam joined my team as a junior analyst and left it running our forecasting model. I watched that happen over two years. What made the difference was not raw ability, it was that he asked for feedback constantly and actually changed his work after receiving it, which is rarer than it should be. By the end he was catching errors in the model I had written myself. Any analytics team would be lucky to have him, particularly one that wants someone who will grow into more than the job description."

2. Direct report recommending a manager

"Elena managed me for three years, through a reorganisation and two rounds of layoffs. She told us the truth in both, including the parts that made her look bad. She also protected the team's focus better than any manager I have had: in three years I was pulled into exactly one project that turned out to be a waste of time. If you are considering working for her, do it. If you are considering hiring her to lead a team, she will build one that stays."

3. Peer recommending a colleague (sample recommendation for a colleague)

"Marc and I sat on opposite sides of the same problem for two years, design and engineering, which is usually a recipe for friction. It never was. He would come to design reviews having already built a rough prototype, so we argued about something real instead of a mock-up. Our redesign shipped six weeks early, and most of that was down to how he ran the handoff. I would take him on any product team I join."

4. Client recommending a freelancer or consultant

"We hired Nadia to run a three-month content project. She delivered it, then told us the strategy behind it was wrong and showed us why, which is not what we paid her for and was worth more than what we did pay her for. Organic signups doubled over the following two quarters. She is direct, she meets deadlines, and she will tell you things you would rather not hear. I have rehired her twice."

5. Freelancer recommending a client

"I have worked with a lot of clients. Jonas is in the top handful, and the reason is unglamorous: he answers questions within the day, he gives feedback in one round rather than five, and he pays on time without being chased. The work was better because the process was not fighting me. If you are deciding whether to take a project with his team, take it."

6. Mentor recommending a mentee

"I mentored Aisha for a year while she moved from journalism into product marketing. Most people making that jump want a roadmap. She wanted the hardest question I could give her each month, and she came back having done more than I asked roughly every time. She landed a role at a company I had told her was out of reach. It was not. Whoever hires her next should give her more responsibility than feels comfortable."

7. Recommending a business owner or founder

"I was an early customer of Rui's company and later an advisor. What stood out was how he handled being wrong: when our use case turned out to be a bad fit, he told us so and helped us find an alternative rather than keeping the contract. That cost him revenue and earned him three referrals from me. If you are considering working with him in any capacity, he is straight with people."

8. Recommending an intern or junior hire

"Yuki interned with us for six months. On week three she noticed our reporting pipeline was silently dropping around 8% of events, something four full-time people had missed for a year. She then fixed it. Beyond the technical side, she asked good questions in meetings where more senior people stayed quiet. She will be an excellent hire for someone, and I suspect that someone will be lucky to get her before word spreads."

15 LinkedIn recommendation examples by role and industry

Shorter versions you can adapt. Each one names a specific behaviour rather than a personality trait, which is the whole trick.

Software engineer

"Diego rewrote our deployment pipeline and cut release time from four hours to twelve minutes. More usefully, he documented it so the rest of us could maintain it after he moved teams. He is the engineer other engineers ask before making a decision."

Product manager

"Hana killed two of her own projects in the year I worked with her, both after the data said they were not working. That is the hardest thing a PM does. The one she kept became our biggest revenue line."

Designer

"Omar presents design work in the language of the business, not the language of design. Our CFO understood why the checkout redesign mattered because Omar explained it in terms of abandoned carts. Conversion went up 18%."

Data analyst

"Fatima found a tracking error that had been inflating our attributed revenue for eleven months. She raised it knowing it would make her own dashboards look worse. That instinct is worth more than any technical skill on her CV."

Sales representative

"Callum closed 140% of quota two years running, and the churn on his accounts was the lowest on the team. Those two things rarely go together. He qualifies hard and does not sell people things they will regret."

Marketing manager

"Ines took a channel we had written off and turned it into a third of our pipeline within two quarters. She works from evidence: every campaign she ran had a hypothesis attached, and she killed the ones that failed without defending them."

Content writer

"Theo writes fast and edits himself ruthlessly, which is a combination I have almost never seen. He also learned our product properly before writing a word, so nothing needed fact-checking by the team. Organic traffic tripled over the year."

Recruiter

"Ana filled seven roles for us in a market where we had failed for months. What made the difference was that she pushed back on our brief and told us two of the roles were unhireable as written. She was right, we rewrote them, and they filled in three weeks."

HR and people operations

"Marta rebuilt our onboarding after our 90-day attrition hit an uncomfortable number. It halved within two quarters. She is unusually good at getting managers to actually do the thing rather than agree to do the thing."

Finance and accounting

"Pieter closes the month in three days and explains variances in language non-finance people follow. When we were fundraising, the diligence process took half the time it should have because his records were in order before anyone asked."

Operations manager

"Grace mapped a fulfilment process nobody fully understood, then removed four steps from it. Cost per order dropped 22% and complaints fell with it. She does the unglamorous work of finding out how things actually run."

Customer success manager

"Luis kept net revenue retention above 110% across a portfolio nobody wanted. He does it by getting in front of problems: half his renewals were secured months early because he had already fixed whatever would have come up."

Teacher or trainer

"Sophie trained our entire support team on the new platform in two weeks. She rebuilt the material after the first session because it was not landing, which most trainers would not have bothered doing. Ticket escalations dropped by a third afterwards."

Project manager

"Adam ran a migration across four teams and two time zones and it shipped on the date he gave us at the start. He does this by being unpopular early: he raises risks when they are still cheap rather than when they are visible."

Executive or department head

"Rebecca took over a division losing money and had it profitable within five quarters, without the layoffs everyone expected. She did it by cutting the product line rather than the people. The team she inherited is still largely intact, which tells you most of what you need to know."

How to request a LinkedIn recommendation

Asking is the part people avoid, and the reason it feels awkward is usually that the request is vague. "Would you mind writing me a recommendation?" makes the other person do all the work, so they postpone it indefinitely.

Fix it by naming the angle. Three messages that get answered:

Asking a former manager

"Hi [Name], I am putting my profile in order for a move into [type of role]. Would you be willing to write a short LinkedIn recommendation? If it helps, the part most relevant is probably [specific project], and what you said at the time about [specific thing] stuck with me. Happy to write one for you as well."

Asking a client

"Hi [Name], I am asking a few clients for LinkedIn recommendations. Would you be up for one covering the [project] work? A few lines is plenty, and if it is easier I can send a draft you can edit or rewrite entirely."

Asking a colleague, with an offer first

"Hi [Name], I wrote you a recommendation this morning, no strings attached, I just thought someone should say it publicly. If you ever feel like returning the favour, the angle I am working on is [thing]. No pressure either way."

Two notes. Offering a draft is not cheating; most people are relieved. And writing one first is by far the most effective opener, because it removes the transactional feel. If you want the longer version of this, we have a full guide to requesting recommendations covering timing and follow-ups.

Mistakes to avoid

  • Writing in adjectives. "Hardworking, dedicated, reliable" describes nobody. One specific behaviour beats five traits.
  • Recycling the same text. Recommendations are public. Two people receiving near-identical paragraphs from you is visible and embarrassing.
  • Making it about you. "In my twenty years leading teams" is a recommendation for yourself.
  • Vague timing. Say when and where you worked together. Without it, readers cannot weigh the opinion.
  • Overclaiming. Calling someone the best you have ever worked with, twice, in different recommendations, devalues both.
  • Collecting instead of curating. Ten generic recommendations on a profile look like a favour exchange. Three specific ones look like evidence.
  • Leaving the rest of the profile weak. Recommendations sit near the bottom of the page. If the top of your profile does not earn the scroll, nobody reaches them.

How to politely decline a recommendation request

Sometimes you get asked by someone whose work you cannot honestly praise, or whom you barely worked with. Writing a lukewarm recommendation is worse for both of you than declining: readers notice faint praise immediately.

A version that keeps the relationship intact:

"Hi [Name], thanks for asking. I do not think I am the right person for this one, mostly because we overlapped on [limited scope] and I would not be able to speak to the work you are moving toward. [Other person] saw much more of it and would write something far more useful. Happy to make the introduction if that helps."

If there is no honest version of "I am the wrong person", declining without a reason is still better than writing something hollow. "I am keeping my recommendations to people I have managed directly" is a policy, not a judgement, and it is easy to say.

Troubleshooting: recommendations not showing up

Two problems come up constantly, and both have dull answers.

The recommendation you were sent is not on your profile. Almost always because you have not approved it. New recommendations arrive as a pending request and stay invisible until you accept them, which is easy to miss if your LinkedIn notifications are noisy. Open the Recommendations section of your profile, switch to the received tab, and look for anything waiting. The other common cause is that the section itself is hidden: if your Recommendations section is switched off in your public profile settings, logged-out visitors will not see it even though you can.

Nobody saw the recommendation you wrote. Recommendations do not reliably appear in the news feed. LinkedIn sometimes surfaces them as an activity item to close connections and often does not, so treat the feed as a bonus rather than the point. If you want people to know you wrote one, the reliable route is to mention it yourself in a post or a message. The permanent value is that your name and headline now sit on that person's profile, in front of everyone who visits it, for as long as they keep it.

One more thing worth checking once a year: recommendations you gave to people who have since changed roles. They stay live, they still carry your name, and a glowing recommendation attached to a context that no longer exists is worth a quick reread.

Frequently asked questions

What makes a good LinkedIn recommendation?

One specific behaviour, backed by a concrete result, from someone whose relationship to you is clear. The best ones are 100 to 150 words, name a single quality rather than a list of adjectives, and close by saying who the person would be right for.

How does a recommendation work on LinkedIn?

Someone writes it from your profile or in response to your request, you receive it for approval, and it only appears once you accept. You can hide or delete it later, and you can ask the writer to revise it before it goes live.

How many LinkedIn recommendations should I have?

Three to five detailed ones from credible people beats twenty generic ones. Aim for coverage of different vantage points: a manager, a client or customer, and a peer say more together than three managers.

How can I see my recommendations on LinkedIn?

They sit in the Recommendations section of your profile, below Skills. From there you can switch between recommendations you have received and ones you have given, and manage which received ones are visible.

Are LinkedIn recommendations public?

Yes, once you approve them. They appear on your profile with the writer's name, photo and headline, and they are visible to anyone who can view your profile, including logged-out visitors if your profile is public.

Do recruiters actually read LinkedIn recommendations?

Rarely at the screening stage, often at the decision stage. They matter most when someone is already interested and is looking for a reason to be confident, which is why a specific recommendation from a credible writer outperforms a pile of vague ones.

What is the 3/2/1 rule on LinkedIn?

It is an informal engagement habit, not a LinkedIn feature: three meaningful comments, two shared or resurfaced posts, one original post per day. It has nothing to do with recommendations, but it comes up in the same searches because both are about building credibility. The recommendation equivalent would be three specific ones from three different vantage points.

Do LinkedIn recommendations show up on the news feed?

Sometimes, and unreliably. LinkedIn may surface a recommendation as an activity item to close connections, but there is no guarantee. The durable value is that your name sits on the recipient's profile permanently, not the momentary feed exposure.

Can I edit or remove a LinkedIn recommendation after it is published?

You can hide or delete any recommendation you received at any time. If you wrote one, you can edit or withdraw it, and the person who received it is notified that it has changed.

Start growing on Linkedin

SAVE TIME. DRAFT LESS. POST MORE
👉  try Taplio for free
AVG. VIEWS
2,979
-1.5%
ENGAGEMENT
3.2%
+0.8%
Josquin
Free Monthly Report

How does your LinkedIn compare?

Get free monthly benchmarks on reach, engagement, and content format performance.

Get the Benchmark
200,000+ posts analyzed every month