How to Ask for a LinkedIn Recommendation Without Writing It for Them

PBL DECISION FRAMEWORK

The PBL Context-to-Recommendation Brief

Create an ethical, useful request through five decisions—without borrowing the recommender's voice.

A recommendation should be independent evidence, not delegated copy

A LinkedIn recommendation can help a reader understand how another professional experienced your work. Its value comes from the recommender's relationship to the work, the specificity of what they observed and the freedom to describe it honestly. A polished paragraph that you wrote for someone else may look efficient, but it removes the independent judgement that makes the proof useful.

Treat the request as a memory brief. Your job is to identify the relevant witness, make the shared context easy to retrieve and protect factual boundaries. Their job is to decide whether to recommend you, what they genuinely believe and how to express it. The result may be less controlled than a self-written endorsement—and more credible because of that.

Choose someone who observed the contribution

LinkedIn currently allows recommendation requests to first-degree connections you work with or have worked with. Connection status is only the platform requirement; it does not establish evidentiary fit. A senior person who barely saw the work may produce a broad statement, while a peer, direct report, client-side collaborator or cross-functional partner may be able to describe your decisions and behaviour precisely.

Start with the claim your profile needs to support. If you want to be known for leading complex regional change, choose someone who saw you make trade-offs across markets—not someone who only knows your conference presence. Check organisational policy, client terms and the person's authority before requesting public commentary. Do not ask someone to imply a client, employer or board endorsement they cannot give.

Ask near a real moment, not from a generic campaign

The best time to ask is when both people can still identify the work: after a project closes, a role changes, a handover finishes or a useful retrospective takes place. You do not need to ask immediately, but name enough detail to restore the shared memory. A batch message sent to a long list of contacts signals that volume matters more than relevance.

Use one request for one relationship and one position. LinkedIn says you can request only one recommendation per position from a given member, even though it does not set a total limit. Choose the position and relationship carefully before sending because the relationship cannot be changed after the request; LinkedIn says a new request is required if that field is wrong.

Give context and prompts—not a finished recommendation

A good request answers four questions: why this person, why this work, why now and what kind of observation would be most useful. State the project or period, your role at the time and the capability you hope the recommendation might illuminate. Then provide optional prompts such as: What decision did you see me improve? How did I behave when constraints changed? What made our collaboration useful?

Prompts should help memory, not prescribe approval. Do not supply superlatives, numerical outcomes the person cannot verify or sentences ready to paste. If a result belonged to a team, make that clear. If a number is confidential, leave it out. The recommender should be able to write a narrower, different or even more qualified account than the one you imagined.

Use a request that makes no easy—and still gives enough detail

A practical message can be short: ‘Hi Sam—would you be comfortable writing an honest LinkedIn recommendation about our work on the regional service redesign last year? You directly saw how I structured the country trade-offs and facilitated the operating review. If useful, the moments most relevant to my current work are the decision workshop and the handover to local leads. Please use your own words, include only what you can verify and feel free to decline if the timing or public context is not right.’

This is a pattern, not copy for every relationship. Replace every fact, remove anything the person did not observe and adjust the tone to the actual relationship. The phrase ‘feel free to decline’ matters only if your behaviour supports it: do not chase repeatedly, make reciprocity a condition or use organisational seniority to pressure a response.

Protect consent, confidentiality and attribution

A recommendation appears on a public-facing professional profile for signed-in network members, not inside a private performance review. Ask the recommender to exclude restricted client names, confidential results, personal information and claims outside their direct knowledge. If the relationship has regulatory, legal or professional disclosure rules, follow those rules before requesting or publishing anything.

Read the final recommendation as evidence. Check names, role, dates, ownership and the boundary of any outcome. Do not silently reinterpret a team result as your individual achievement. LinkedIn lets recipients accept, dismiss or request a revision before the recommendation is displayed. Ask for a revision only to correct a material error, remove sensitive information or clarify a genuinely misleading statement—not to make the praise stronger.

Give recommendations independently, not transactionally

You can offer to write a recommendation for someone whose work you directly know, but do not make it a trade. A reciprocal pair written at the same time can still be honest; the risk appears when approval is expected, wording is coordinated or declining would damage the relationship. Separate the decisions: recommend their work only if you can stand behind the account, whether or not they write one for you.

When writing, use the same standard you hope to receive. Explain the relationship, identify a concrete contribution or behaviour, bound the context and avoid claims you cannot substantiate. LinkedIn allows a recommendation giver to revise or delete what they wrote later, and the recipient must review a revised version before it appears again.

Worked example: turn a vague request into a useful brief

Fictional teaching example — Arun is an operations director updating his profile for broader transformation roles. He first considers asking a former chief executive: ‘Could you say I am a world-class transformation leader who delivered exceptional results?’ The executive sponsored the programme but did not observe Arun's day-to-day work, and the language supplies both the judgement and conclusion.

Arun instead asks a former cross-functional peer who attended the decision reviews. He names their shared programme, the position he held and two moments the peer directly saw: how he surfaced market constraints before the rollout decision and how he clarified ownership during the handover. He asks for an honest account in the peer's own words, makes declining easy and excludes confidential figures. The example is hypothetical, not a PBL client story or performance claim.

Review recommendations as a proof portfolio

More recommendations are not automatically better. Review whether each one supports the professional association you want a relevant reader to understand, brings a distinct observer perspective and remains accurate. A detailed recommendation from an earlier role can still be useful when the capability transfers; label the historical context through the associated position rather than presenting it as current work.

LinkedIn allows accepted recommendations to be hidden. Hide a recommendation when it creates confusion, exposes a changed boundary or no longer supports the profile's job; do not edit history to pretend the relationship never existed. If the proof gap is about current work, seek current evidence instead of stretching an old endorsement. Pair recommendations with accurate Experience entries and work samples so a reader can triangulate the claim.

Measure usefulness, not recommendation count

The success measure is whether a relevant reader understands and trusts an important contribution more easily. Ask a few intended readers what the recommendation helped them believe, what still feels unproven and which context was missing. Notice whether qualified conversations reference the behaviour or judgement described—not merely that a recommendation exists.

If the recommendation is warm but vague, improve future requests with a clearer shared moment. If it is specific but irrelevant, revisit the profile position and the evidence job. If it is accurate and relevant but invisible, check whether you accepted it and whether its associated position supports the current profile story. Do not turn the process into a quota; one precise account can do a different job from several interchangeable compliments.

DECISION TABLE

Compare the choices clearly

ACTION CHECKLIST

Use this before making the next move

• Define the current profile claim the recommendation should help a reader assess.
• Choose a first-degree connection who directly observed that contribution.
• Confirm that a public recommendation is permitted by relevant policies and agreements.
• Name the correct relationship and your position at the time.
• Reference one specific shared project, period or decision.
• Offer two or three factual memory prompts rather than drafted praise.
• Separate your contribution from team or organisation outcomes.
• Exclude confidential names, figures and restricted information.
• Ask for an honest account in the recommender's own words.
• Make declining easy and avoid repeated follow-up.
• Review the received text for factual accuracy, permission and current relevance.
• Accept, request a necessary revision, dismiss or hide the recommendation deliberately.

COMMON QUESTIONS

Frequently asked questions

AUTHOR & METHOD

Resources by Personal Brand Lab

This article was developed from the PBL Resources methodology for turning personal-brand questions into practical decisions. It distinguishes operating recommendations from claims about guaranteed performance; no client result or market statistic has been invented.

Sources checked

• LinkedIn Help — Request a Recommendation — Checked on 11 September 2026 to verify the current desktop and mobile request paths, required relationship and position fields, personalised message option, primary-language rule and request limits.
• LinkedIn Help — Recommendations on LinkedIn — Checked on 11 September 2026 to verify first-degree connection eligibility, acceptance and revision choices, hiding, position-level limits and what happens when a connection is removed.
• LinkedIn Help — Request a Revision of a Recommendation — Checked on 11 September 2026 to verify that recipients can ask the recommender for specific changes through a personalised revision request.
• LinkedIn Help — Revise a Recommendation You've Given — Checked on 11 September 2026 to verify the giver's revision workflow and that the recipient must review a revised recommendation before it reappears.

Published and reviewed 11 Sep 2026.