Tuning AI Comment Suggestions to Match Your Company's Voice

AI comment suggestions solve a real problem — the “I don’t know what to say” hesitation that stops people from engaging even when they’re willing to. But the suggestions are only as good as the tone they’re set to.

Setting the baseline tone

Admins can configure a company-wide tone — professional, casual, or thought-leader — that shapes how every suggestion reads by default. Pick the one that matches how your best natural communicators already write on LinkedIn, not an aspirational tone nobody on the team actually uses.

Why three options matter

Employees choosing between three AI-generated comment options, rather than one, changes the experience meaningfully. A single suggestion feels like being told what to say; three options feels like being offered a starting point, and gives each person a chance to pick the version that sounds most like them.

Editing is not optional — it’s the point

The platform never posts a comment automatically. Every suggestion is meant to be edited before publishing, even lightly. Encourage your team to treat suggestions as a first draft, not a final answer — a suggestion edited to include one specific, personal detail reads authentically; one posted verbatim usually doesn’t.

When to revisit the tone setting

If you’re getting feedback that suggestions feel generic or off-brand, that’s a signal to revisit the company-wide tone setting rather than asking individuals to simply “write it differently.” A five-minute adjustment to the baseline tone often fixes what feels like a widespread engagement problem.

Letting the system learn

Employees can rate suggestions with a thumbs up or down, which feeds back into future recommendations. Encourage your team to actually use this — the more feedback the system gets early on, the faster suggestions start to feel personalised rather than generic.

---