In a recent training webinar, our product team walked through three real, in-production forms used by our own internal teams. Each one solves a different problem, and each one stitches together features that any FormAssembly admin can reach for. Below is a tour of all three, with notes on what makes each one tick, and what you can reuse across your own builds.
A CSAT form for one-off product trainings
The problem behind this form is one most teams will recognize: when a trainer runs a one-off live session that doesn’t go through the full LMS, the team still wants standardized feedback on the session and a way to track which training events generated which scores over time. Buying a separate CSAT tool to capture that data feels like overkill. So our product trainer built it inside FormAssembly instead.
The form pulls together a handful of features:
- A prefilled training event name field, so the respondent doesn’t have to type the session title.
- A radio-button overall rating, styled as a star rating with a bit of custom code.
- A conditional optional field that appears only when someone rates the training three stars or less, asking for context on the low score.
- A matrix of agree/disagree questions that map to a CSAT score behind the scenes.
- A calculated CSAT percentage that uses variable assignments on the matrix responses, sums them, divides by the total possible points, multiplies by 100, and rounds to two decimal places.
- A Google Sheets workflow connector that pushes the data to two tabs in the same spreadsheet — one curated subset of fields for managers, and one with the full response set for deeper analysis.
The result is a flexible CSAT process that doesn’t require a separate CSAT product, doesn’t require manual aggregation across responses, and gives managers a tab they can filter by event, employee, or score range.
Self-grading knowledge checks (no manual grading required)
When our product training team needs to push a quick knowledge check to internal teammates — after a meeting, after a training, or as a Slack pop quiz — they don’t have the bandwidth to manually grade dozens of submissions. So the form does it for them.
The build relies on a few interconnected features:
- A short text intro that sets context before the quiz begins.
- Multiple-choice questions (radio buttons and checkboxes), so all options are visible at once.
- Variables assigned to each question, with a value of 0 for incorrect choices and 1 for correct ones.
- A flexible “select all that apply” pattern: for partial credit, the 1 value is evenly divided across correct choices. For all-or-nothing scoring, a calculated field returns 1 only if every correct choice was selected.
- A total-right calculated field that sums the variables and converts the count to a percentage.
- A dynamic thank you page that shows each question, the respondent’s answer, and a “you are correct” or “not quite there” verdict, driven by a formula on the thank you page itself.
The respondent gets instant feedback. The trainer gets zero grading overhead. And because the thank you page is fully customizable, past versions of these knowledge checks have linked respondents directly to documentation when they answered a question incorrectly, turning the quiz itself into a learning loop.
A Salesforce-connected customer feedback form
The most ambitious of the three builds — and the one that drew the most audience questions — was a Salesforce-connected customer feedback form.
Let’s set the stage.
The problem
- Customer-facing teams hear valuable feedback during conversations that customers wouldn’t necessarily submit through a public-facing channel.
- A CSM picks something up on a call.
- A support case email has a buried product insight.
- A sales conversation surfaces a feature gap.
- The team needed a way to capture all of that in context, route it cleanly into Salesforce, and report on it over time.
- The form is built around four load-bearing design choices.
A Salesforce custom button plus URL prefill.
On Account, Contact, and Case records, a custom “Feedback” button opens the form in a new tab and passes the Salesforce record ID as a URL parameter. From there, a prefill connector looks up the rest of the related record information — account, contact, case, user — automatically. The button only needs to pass one ID; the connector handles the rest. As a bonus, the form also captures the Salesforce user ID of whoever clicked the button, partially anonymizing the submission while still allowing reporting on who submitted what.
A dynamic picklist for feedback topics.
Topics live as custom records in Salesforce, and the form pulls them in fresh every time it opens. New topics added in Salesforce appear in the form automatically, without anyone editing the form itself. A status field (active vs. deactivated) on each topic record controls whether the topic still appears in the picklist, which preserves historical reporting on retired topics without polluting the live form.
A “submit additional feedback” loop using a redirect formula, not repeatable sections.
Instead of using repeatable sections to collect multiple pieces of feedback in one go, the form uses a redirect formula on the thank you page that reopens a fresh version of the form, carrying over the case ID and the requester ID. This was a deliberate data quality choice — repeatable sections were producing inconsistent feedback entries (too many, duplicated, or split across rows). One submission per piece of feedback turned out to be cleaner to report on.
A junction object structure in Salesforce.
The submitted feedback record connects via lookup relationships (not master-detail) to the contact, account, case, user, and topic records. Lookups mean that if a topic record is later removed or changed, the feedback record isn’t lost, which protects long-term data integrity in a product that’s still evolving.
The result is a feedback collection process that scales with the product, surfaces trends at both the topic and category levels, and lets the team report at any altitude using a single form.
What this means for FormAssembly builders
Three patterns travel across all three use cases.
First, use your own forms internally before you buy something separate. Every team has a CSAT, a knowledge check, or a feedback-collection problem somewhere. Those problems are almost always solvable with features you already pay for in FormAssembly.
Second, design for the report, not just the response. Each of these forms was built with the downstream data structure in mind — which fields would feed which reports, which IDs would map to which Salesforce objects, and which calculations would produce the metrics teams actually wanted to see. The forms that lasted weren’t the ones that captured the most data; they were the ones whose data was easiest to use later.
Third, dynamic features pay for themselves. Dynamic picklists that pull from Salesforce, calculated fields that score quizzes automatically, redirect formulas that reset forms cleanly — each one removes manual work that would otherwise compound over time. The upfront build is slightly more involved; the long-term operations cost drops by an order of magnitude.
Watch the full webinar
The full session includes the live builds of all three forms, the variable assignment and calculation logic behind the CSAT and quiz scoring, the Salesforce custom button setup for the feedback form, and the audience Q&A on dynamic picklist filtering, response anonymization, and the upcoming workflow templates feature.
