Skip to content

Preview Links and Open Links

Two links let someone with no Askalot account open a survey. They look similar and they are not the same thing, so the first section is about telling them apart.

Preview link Open link
Points at a questionnaire a campaign
For a reviewer reading your draft a volunteer taking part
Produces a preview — nobody was interviewed a real survey, marked volunteer
Reaches your data never yes, through that campaign
Expires always no — it is standing
Stopped by expiry, or the Stop button the Stop button, the cap, or the campaign closing

The difference that matters is the last-but-two row. A preview cannot enter a dataset — not "is excluded by default", but cannot: the extraction step refuses previews outright and there is no setting that admits them. A volunteer's answers are real data and are meant to be analysed, flagged so you can tell them from the people your sample design selected.

You have a questionnaire in progress and you want a colleague, a client, or an ethics reviewer to walk through it as a respondent would. They have no account and you do not want to create one for them.

Publishing one

  1. Open the questionnaire in Targetor
  2. In the Preview Links panel, set:
    • Maximum runs — how many separate walk-throughs this link may produce
    • Expires in (hours) — how long it stays live. Leave it blank for the default of three days
  3. Click Create link and copy the address you are given

Each person who opens the link gets their own run. Two reviewers opening the same link do not share a session and cannot overwrite each other's answers — the maximum counts runs, so a link capped at three is three walk-throughs, not three page loads by one person.

What the reviewer sees

The instrument exactly as it stands right now, including edits you make while the link is live. That is deliberate — a preview link is for reviewing work in progress — but it is also why the link always expires: a link you forget about keeps showing whatever the questionnaire becomes.

Finding the runs afterwards

Every preview appears on the Surveys page in Targetor, labelled preview, with the path the reviewer walked and the answers they gave. That is the point of keeping them: you can see where a reviewer hesitated or dropped out. They will never appear in a Bronze extraction, a Silver analysis base, or an export.

Stopping one

Stop revokes the link immediately; anyone still holding it is turned away. Runs already taken are kept and stay readable. If you need a link with a higher cap, use Rotate: it publishes a new link and stops the current one in the same act, so you are never left with two live links you have to remember.

An open link is a campaign's standing public door: post it, mail it, print it on a flyer. Anyone who opens it can complete the survey with no interviewer and no account, and their answers join that campaign's data marked as a volunteer.

Before you can publish one

The campaign must accept walk-ins. That is set on the campaign in Targetor and it has no default — you are asked to choose when you create the campaign, because it decides who is allowed into your study. If it is off, the Open Link panel says so and offers no publish control: a link into a campaign that refuses volunteers is a door that opens on a wall.

Publishing one

  1. Open the campaign in Targetor
  2. In the Open Link panel, set Maximum arrivals — or leave it blank to take the campaign's sampling strategy target size
  3. Click Publish and copy the address

The maximum is required and cannot be raised. A public link anyone can forward, with no ceiling, is a link nobody is counting; if your campaign has no sampling strategy to borrow a number from, you have to name one yourself. To raise it, Rotate — a new link with a fresh count, and the old one stopped in the same act.

The panel shows how much of the bound is spent as people arrive, so you can see a link approaching its ceiling before it closes.

What a volunteer is asked

The fields the campaign's sampling strategy actually measures — if it stratifies on age group and region, that is what they are asked for. Before any of it, the page states who is collecting the data and on what basis.

Name and email are optional, and an address given here is treated as something somebody typed, not as an identity. Two people may enter the same address and both take part; nothing is verified, nothing is matched to an existing respondent, and no account is created. That is why your analysis surfaces report how much of a base arrived unverified alongside the drawn/volunteer split.

Two things that close the door

A published link admits someone only while both hold:

  • the campaign still accepts walk-ins
  • the campaign is active — a draft is still being set up, and a paused one was paused to stop collecting

So pausing a campaign stops arrivals without touching the link, and the link starts working again when you resume. The panel tells you when a published link is currently turning people away for this reason.

Stopping one

Stop revokes the link and leaves the campaign alone — it keeps collecting from everyone else, from invitations, interviewers and walk-ins. You never have to pause data collection to close a public door.

What this means for your analysis

Volunteers reach your Bronze dataset with every other completed survey from that campaign, carrying an arrival of volunteer rather than drawn. Nothing is hidden from you and nothing is silently dropped:

  • The quality story reports the drawn/volunteer split of the base it graded, and how much of it arrived with unverified identity
  • The Bundle has a per-analysis choice of whether volunteers stay in the Silver analysis base. It cuts at Silver, not at extraction, so the split stays reportable whichever way you decide — and changing your mind re-derives Silver and Gold without touching Bronze

A volunteer carries no design weight and was drawn against no calibration target, which is exactly why the distinction is recorded rather than smoothed over. A dataset that cannot separate them cannot honestly weight either.