Public Link vs. Org-Gated vs. Invite-Only: A Real Decision Guide for Survey Distribution

Most survey advice talks about questions — scale types, wording, order effects — and skips right past distribution, as if how a survey reaches people is a solved, boring detail. It isn't. Who can see the link, and what it takes for them to respond, changes the kind of answers you get as much as the questions themselves do. SparkyLoop gives you three distribution modes, and each one is actually suited to a different kind of ask.

Public link — the default, and the right one more often than you'd think

A public link is exactly what it sounds like: one URL, anyone who has it can respond, no login required. This is the right mode whenever the honest answer to "who should be able to respond?" is "anyone I share this with," full stop — a feedback form linked from your product's footer, a survey dropped into a newsletter, a link posted after a public event.

The tradeoff is exactly the openness that makes it convenient: you can't guarantee who's actually answering, and you can't stop someone from responding twice unless you're relying on their goodwill. For low-stakes, high-reach feedback where you want breadth over certainty about who answered, that tradeoff is fine. For something where the respondent's identity or eligibility actually matters, it isn't.

Org-gated — for internal-only asks that have no business being public

Gating a survey to your own organization's members (available on plans that include gated distribution) requires the respondent to be logged in as part of your org before they can answer. This is the mode for the surveys that are fundamentally internal: a retro after a project wraps, feedback on an internal tool, a pulse check that should never be answerable by someone outside the company.

The reason this is a genuinely different tool from invite-only, not just a stricter version of it, is who it's for. Org-gated says "anyone on our team can answer this" — you're not tracking which specific person you expect to respond, you're restricting the pool to "people who work here." That's a much looser targeting than invite-only, but a much tighter one than public.

Invite-only — when you need to know exactly who you asked

Invite-only distribution sends individually tracked links by email to specific people. This is the mode to reach for whenever you actually care about who has and hasn't responded — a customer research round with a named list of accounts, a 360-degree review where every rater needs to be someone specific, an employee survey where you want to see the response rate against a known headcount.

For a 360-style review specifically — any survey built around a matrix/competency question — invite-only isn't optional, it's structural. Each invite carries both a subject (who's being rated) and a relationship (self, manager, peer, or direct report), and that pairing is what lets SparkyLoop group all the responses about one person into a single combined competency report afterward. Without individually tracked invites, there'd be no way to know whose rating belongs to whom.

The cost of invite-only is friction: you have to actually build the list and send the invites before you can publish, and you can't get a stray extra response from someone you didn't think to invite. For the kind of survey where that precision is the entire point, it's friction worth paying for.

A quick way to decide

If you don't care who answers and just want the broadest reach: public link. If you care that it's your own people and nobody outside the org, but don't need to track individuals: org-gated. If you need to know exactly who was asked, who answered, and — for 360s — in what relationship to the subject: invite-only.

It's worth deciding this deliberately rather than defaulting to whatever distribution mode your last survey used. A customer NPS survey and an internal 360 review have almost nothing in common in terms of who should be able to respond, and picking the wrong mode either leaks a survey wider than you meant or adds invite-management overhead to something that never needed it.

One practical note either way: decide on anonymity at the same time you pick a distribution mode. Anonymous responses and org-gated or invite-only distribution aren't contradictory — you can restrict who can respond while still not attaching identity to their answer — but it's a separate setting worth thinking through on purpose rather than leaving on the default.

How to do this in SparkyLoop

While a survey is still a draft, open its distribution settings and choose public, org-gated, or invite-only, along with whether responses are anonymous. For invite-only surveys, send invites by email before publishing — 360-style surveys also need a subject and relationship per invite. See all three modes on SparkyLoop's features page.