How surveys become signals and decisions

Sending a survey from **Forms** (sidebar Surveys, `/forms`) is only half the story. Once a survey goes out, Verinode IQ keeps watching it in the background, on the same nightly sweep that runs ever…

10 min read·Updated July 13, 2026
On this page

What this article covers

Sending a survey from Forms (sidebar Surveys, /forms) is only half the story. Once a survey goes out, Verinode IQ keeps watching it in the background, on the same nightly sweep that runs every other detector, and turns what it sees into decisions: cards that show up in the Feed and in the Decisions log with a title, a plain-language read, and (where there is one) a one-tap action. This article is about that second half: how a survey going quiet, or filling up, or coming back with a low score, becomes something you can act on without opening Forms at all. For the survey-building and response-reading mechanics themselves, see the Forms overview and reading the responses feed.

Three things happen after you send a survey, and they are the three sections below:

  1. A lifecycle check on every open survey: has it stalled, or has it just filled up.
  2. A satisfaction check on the ratings that come back, comparing your team to you and to the industry.
  3. A one-tap reminder sitting right on the decision card, so nudging non-responders doesn't mean leaving the card to go find the survey.

Note

None of this needs the survey type to be a certain kind. Team surveys, vendor pulses, tool checks, supply-partner surveys, carrier and TPA surveys, and process surveys all run through the same lifecycle detector. Only vendor, tool, and supply surveys also get the satisfaction/gap/benchmark checks in section two, because those are the ones with a Verinode Score and an operator rating to compare against.

Section 1: the lifecycle detector (stalling and complete)

What it is. Every survey you send out to your team, a vendor, a carrier, or a TPA gets watched by a rule-based detector (no AI cost) that runs every night for every real operator. It reads two things per survey: how many people were invited, and how many have replied, and turns that into one of two decisions. Self-audits (Audits and Reviews, the other two Forms tabs) don't have outside recipients, so they can't stall or complete in this sense and are skipped.

When a survey is "stalling"

A survey is marked stalling once all of these are true: it has been sent to at least two people, it has been open a handful of days, fewer than half of the invitees have replied, and it hasn't been open so long that a low reply count reads as a dead survey rather than a slow one. A brand-new survey isn't stalling yet, it just hasn't had time. A survey that's been open a month with nobody answering isn't stalling either, at that point it's simply not going to fill up, and Verinode stops nagging about it.

What the card says. The title is stable for the life of the survey, for example:

Your Vendor Assessment is waiting on replies

(the exact wording uses the survey's subject name or title, falling back to "Your \[kind\] survey" if neither is set). The body carries the live counts and updates every night the survey stays open, something like:

3 of 8 people have answered, 5 days after you sent it. A quick reminder to the 5 who haven't usually lifts the response rate, and you need enough answers for the read to mean anything.

The recommendation underneath is direct: send a reminder to the people who haven't answered, or nudge them in person. The severity starts at Info and escalates to Warning once the survey has been open long enough that the slow pace is starting to look like a pattern rather than normal lag. A stalling card is filed under the same domain as the survey itself, team, vendor, carrier, or process, so it sits alongside your other decisions for that area.

What resolves it. The card clears itself the moment the survey fills up, closes, or ages past the point where Verinode stops tracking it, with one exception: if you've already marked the decision actioned, Verinode leaves the resolution to the outcome you record rather than silently closing it out from under you.

When a survey is "complete"

Once every invitee has answered, and that happened recently (within about a week, so a first run after a feature update doesn't flood your Feed with FYIs about surveys that quietly finished months ago), Verinode files a calm "results are in" decision:

Your Vendor Assessment results are in All 8 people answered. The average rating was 4.2 out of 5. Open Forms to read what they said.

If the average rating comes back low, and the survey type is a team, carrier, TPA, or process survey (vendor, tool, and supply surveys already get their own dedicated satisfaction findings, described below, so the completion note stays neutral for those to avoid saying the same thing twice), the severity steps up to Warning and the recommendation reads:

The scores came back low. Read the comments and pick one thing to change this week.

Otherwise it's an Info card telling you to skim the answers and note anything worth acting on. Either way, the card links straight into Forms at the specific survey (/forms?surveyId=...) so the full breakdown is one tap away.

Tip

Both decision titles stay fixed per survey ("...is waiting on replies" / "...results are in") even as the live numbers inside the body change night to night. That's deliberate: it's what lets Verinode update a running decision in place instead of creating a new one every time the count moves, and it's why dismissing or parking a stalling card sticks even as replies trickle in.

Section 2: satisfaction, gap, and benchmark signals

What it is. Once at least two responses have landed on a vendor, tool, or supply-partner survey, Verinode compares your team's average rating against two other numbers it already has on file: your own personal rating of that relationship, and the industry benchmark (the Verinode Score) for that vendor. Out of that comparison come up to four distinct findings, each with its own signal and its own card. Team, carrier, TPA, and process surveys don't carry a Verinode Score to compare against, so only the first check (below) can apply to them; the gap and benchmark checks are vendor-relationship specific.

Team dissatisfaction. If your team's average rating on a vendor drops below 3 out of 5, Verinode raises a flag:

Team rates this vendor 2.4/5 Your team's average satisfaction is 2.4 from 6 responses. This is critically low, investigate immediately.

Below 2.0 the severity is Critical and the recommendation is to talk to your team about specific pain points and consider alternatives or a vendor improvement plan. Between 2.0 and 3.0 it's Warning, and the ask is a team check-in to find the friction points before it gets worse.

Owner-team perception gap. If you've rated the same relationship yourself, Verinode checks how far your number sits from your team's average. A gap of 1.5 points or more (on the 5-point scale) is worth surfacing:

Team rates 2.1 vs your 4.0, they see something you don't

When the team rates it lower than you do, that's a signal your team is living with day-to-day friction you aren't seeing from the owner's seat, and the recommendation is a direct conversation plus watching the day-to-day usage before making a call. When the team rates it higher than you do, the card flips the framing, your team values the relationship more than your own cost or contract concerns might suggest, and the recommendation is to weigh operational value, not just price, before changing anything. A gap of 2.5 points or more escalates to Critical; otherwise it's Warning.

Below or above the industry benchmark. Team ratings run 1 to 5; the Verinode Score for a vendor runs 1 to 10, so Verinode scales the team average onto the same 10-point line (multiplying by 2) before comparing. A gap of 2 or more points in either direction is worth a card:

Team satisfaction 3.2 points below industry benchmark Your team rates this vendor 2.8/5 (5.6 scaled), but the industry benchmark is 8.8/10. Peers rate this vendor higher, the issue may be specific to your setup.

Below benchmark reads as Warning (or Critical at a 4-plus point gap), with the recommendation pointing at training and configuration rather than assuming the vendor itself is the problem, peers are getting more out of the same vendor, so the gap is likely fixable on your side. Above benchmark by 2 or more points is an Info card that frames the relationship as one worth protecting: lock in pricing before renewal, consider expanding it, and use it as your own internal bar when evaluating alternatives elsewhere.

All four of these live under the vendor domain and dedupe on the specific vendor relationship, so a recurring low score updates the same card night over night instead of stacking duplicates.

Section 3: the one-tap reminder on a decision card

What it is. Open a stalling-survey decision (any card whose signal type is the one described in section one) and, right on the card, you'll find a small nudge panel:

Nudge the people who haven't answered Resends the survey to the 5 people still outstanding. [ Send reminder ]

This only renders on that specific kind of decision, on any other decision the panel doesn't appear at all. The count in the sub-line is read straight from the decision's own evidence, so it always matches what the card's body already told you.

  1. 1Open the stalling decision from the Feed or the Decisions log.
  2. 2Read the nudge panel: it names how many people haven't answered yet.
  3. 3Click Send reminder. The button shows "Sending…" while it resends the survey invite to everyone still outstanding, in place, no trip to Forms required.
  4. 4When it finishes, the panel replaces itself with a confirmation: "Reminder sent to 5 people." (If everyone had already answered by the time you clicked, it reads "Everyone has already answered, nothing to resend.")
  5. 5The page refreshes so the decision reflects the latest state.

If the resend fails for any reason, the panel shows the error in place rather than failing silently, so you know to try again or check the survey directly in Forms.

Heads up

The reminder resends the invite, it does not extend the survey's close date or change its recipient list. If a survey needs a longer window or a different set of people, make that change from Forms directly; the decision card's button is built for the single most common fix (a plain nudge), not for editing the survey.

Empty states

  • No surveys open at all. The lifecycle detector simply has nothing to check, no stalling or complete cards appear, and any previously open stalling cards resolve themselves once their survey closes or ages out.
  • A response count of one. A one-person survey never triggers a stalling card, there's no one to "nudge" when the sole invitee hasn't replied yet.
  • Fewer than two responses on a vendor survey. The satisfaction, gap, and benchmark checks in section two need at least two responses to mean anything and simply don't run below that.
  • No personal rating on file. The owner-team perception gap check is skipped entirely if you've never rated that vendor relationship yourself, there's nothing to compare the team average against.
  • No Verinode Score for a vendor. The below/above-benchmark checks are skipped for any vendor without a Verinode Score on file yet; the team-dissatisfaction check (section two, first check) can still fire on its own.

Best-practice example

Say you send a Vendor Assessment to your six-person crew about a materials supplier. Four days in, only two of six have replied, a stalling card appears in your Feed: "Your Vendor Assessment is waiting on replies," Info severity, with a one-tap Send reminder button naming the four outstanding people. You click it, the invite resends, and the card refreshes to confirm. Two days later the last four answer. Because they finished within the last week, the stalling card resolves and a fresh "results are in" card appears with the average rating and response count. If that average comes in more than two points below the supplier's Verinode Score, a separate below-benchmark card explains the gap in plain terms and points you at training and configuration instead of assuming the vendor itself is the issue, peers are getting more out of the same relationship, so the fix is probably on your side.

Data sources

Data sources

  1. 1.Your surveys, invites, and responses. Your business.
  2. 2.Your own satisfaction ratings on vendor relationships. Your business.
  3. 3.Verinode Score for the vendor (where available). Verinode reference data.
Was this helpful?