Verified against Claude · 2026-08-02
Write a LinkedIn recommendation that would actually help someone get their next role
Turn specific, real memories of working with someone into a recommendation structured around a hire-relevant claim and one honest specific detail, instead of a string of interchangeable adjectives that could describe almost anyone on the platform.
The prompt
Ready to copy — highlighted parts are example details you can swap.
WHO I'M WRITING THIS FOR AND OUR WORKING RELATIONSHIP
A designer I managed for 14 months on the growth team
WHAT I ACTUALLY SAW THEM DO — specific projects, situations, decisions
She redesigned our onboarding flow after independently noticing drop-off data nobody had asked her to look at, and pushed back on a launch date when usability testing surfaced a real problem, which delayed us two weeks but avoided a much worse post-launch fix
THE ROLE OR DIRECTION THEY'RE LIKELY GOING FOR NEXT, IF I KNOW
Senior product designer roles at a larger B2B SaaS company
ONE THING THEY'D WANT EMPHASIZED, IF THEY TOLD ME
Their ability to work independently without close direction
HOW WELL I ACTUALLY KNOW THEM
Direct manager for 14 months, worked closely on 3 major projects
WHY GENERIC PRAISE ACTIVELY UNDERMINES CREDIBILITY
A LinkedIn recommendation sits permanently on the recipient's profile next to their role history, and a hiring manager or recruiter skimming a profile reads it as third-party evidence — but only if it reads as specific. A recommendation built entirely from adjectives ("hardworking," "great team player," "a pleasure to work with") is recognized instantly as a template and actually undermines credibility because it signals the writer either didn't know the person well or didn't think enough to write something real.
RULES
1. OPEN with one sentence naming the actual working relationship and duration — how you know them, not a generic "I had the pleasure."
2. LEAD the body with the single most relevant claim to Senior product designer roles at a larger B2B SaaS company if known, otherwise the single strongest thing you actually witnessed — not a list of five traits given equal weight.
3. BACK the claim with one specific example from She redesigned our onboarding flow after independently noticing drop-off data nobody had asked her to look at, and pushed back on a launch date when usability testing surfaced a real problem, which delayed us two weeks but avoided a much worse post-launch fix — a real project, a real decision, a moment, not a restated trait ("they are very organized, for example they are organized").
4. INCLUDE exactly one honest specific detail that isn't generic praise — something only someone who actually worked with them would know or say (a role they stepped into unasked, a hard moment they handled a specific way).
5. CLOSE with one sentence stating plainly what kind of team or role they'd be strong in next, tied to the evidence above, not a generic "any team would be lucky to have them."
6. KEEP it under 1,900 characters — LinkedIn recommendations have no strict limit, but readability drops hard past this length.
HONESTY CONSTRAINT
Do not claim a closeness or scope of work you weren't given in Direct manager for 14 months, worked closely on 3 major projects — if you only worked with them for a few months on one project, the recommendation should read like that, not like a five-year mentor relationship. A recommendation that overstates the relationship is checkable against the same profile's dates and titles, and a mismatch there costs the writer's credibility, not just the request's usefulness.
OUTPUT
The recommendation text, its character count, plus one line noting which specific example carried the most weight and why you led with it over the others given.Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
Recommendations sit permanently attached to a specific role in a profile's history, next to real names, which is precisely why generic-adjective ones are so easy for a reader to spot and discount — a reader skimming five recommendations quickly learns to differentiate the two that mention an actual project from the three that could be pasted onto any employee at any company.
Leading with the claim most relevant to the person's likely next role, rather than listing traits in whatever order they come to mind, matches how a hiring manager actually reads a profile: they are scanning for evidence relevant to a specific decision, not building a rounded personality profile, so a recommendation structured around "why this decision" outperforms one structured around "who this person is in general," even when both versions contain the same underlying facts about the person.
The honesty constraint about scope of relationship protects against the second most damaging failure mode after genericness: an inflated claim of closeness or authority ("I mentored them for years") that a reader can quietly cross-check against the stated job titles and dates on the same profile. A mismatch there costs the writer's own credibility on every other claim in the recommendation, not just the inflated one, because a reader who catches one overstatement reasonably discounts the rest of the text too.
Requiring exactly one honest specific detail beyond the main evidence-backed claim targets a real gap in most recommendations: even a well-structured one built around a single strong claim can still read as constructed if every sentence in it serves the argument too neatly, whereas one small, specific, slightly unpolished detail — the kind only an actual colleague would think to mention — is what signals the whole thing was written by someone who was really there, not assembled from a template with the right shape.
Verified against
Claude Sonnet 5 · 2026-08-02
ChatGPT GPT-5.1 · 2026-08-03
Changelog
- 2026-08-03 — Initial publish for the rewritten linkedin category, verified against Claude (Sonnet 5) and ChatGPT (GPT-5.1).
Need this built into your business?
If a prompt isn't enough — Google Ads management, built and maintained for you — that's Scult's day job.
EXPLORE GOOGLE ADS MANAGEMENT
