Introduction
Migrating from SurveyMonkey to Sprig is a rebuild, not an import. Sprig documents no import for SurveyMonkey survey designs or historical responses, so the move is four projects: audit what’s still fielding, export and archive every response before your contract lapses, rebuild each instrument from an annotated questionnaire, and run a bridge study on any tracked metric. Plan the whole migration around the contract date, not the rebuild.
Last updated November 2026.
What a Migration Actually Involves
A SurveyMonkey migration is four separate projects that share a deadline. Treating it as one export-and-import task is the most common reason these moves run late, because each project has a different owner, a different failure mode and a different point of no return.
The Four Projects
The work breaks down this way:
- The instrument rebuild: every survey still in use gets recreated in Sprig, question by question, with its logic.
- The archive: every response you might need again leaves SurveyMonkey in a format you can read without SurveyMonkey.
- The measurement transition: any metric you track over time gets a bridge study so the series survives the change.
- The access and integration cutover: users, single sign-on, provisioning and downstream data consumers move without a gap.
The rebuild is the most visible project and usually the least risky. The archive is the one with a contractual clock on it. The measurement transition is the one most teams skip, and it’s the one an executive notices six months later when a satisfaction score drops for reasons nobody can explain.
Why the Order Matters
The archive comes first because it’s the only project you can’t redo. Once a SurveyMonkey Enterprise account lapses, SurveyMonkey’s own documentation describes a 30-day pause, after which users are removed from the team and accounts drop to the free Basic plan, where exporting isn’t available. The rebuild can slip a month with no lasting harm. A missed export window can’t be recovered.
A Sequenced Migration Plan
The plan below sets phases, owners and dependencies. It deliberately carries no durations, because the length of each phase depends on your account size, your API limit and your bridge-study windows, and no published source gives a general figure.
| Phase | Owner | Depends on | Done when |
|:---:|:---:|:---:|:---:|
| 1. Audit | Research operations lead | Contract end date known | Every survey marked live, dormant or dead |
| 2. Archive | Analytics owner | Audit list | Every export verified against response counts |
| 3. Identity setup | IT admin | Sprig contract signed | SSO live, SCIM deprovisioning tested |
| 4. Rebuild | Survey owners | Audit, attributes loaded | Every live survey drafted and line-checked |
| 5. Bridge and overlap | Metric owners | Rebuild of tracked surveys | Written threshold met or break declared |
| 6. Cutover | Research operations lead | Validation approval | Every link points at Sprig |
| 7. Decommission | Procurement and privacy owner | Final exports verified | SurveyMonkey account closed |
The table assigns each phase one accountable owner: the research operations lead for the audit and cutover, the analytics owner for the archive, the IT admin for identity, survey owners for the rebuild, metric owners for the bridge studies, and procurement with the privacy owner for decommissioning.
Phases 2 and 3 can run alongside phase 1’s later steps. Phase 5 is the only one that can’t be compressed by adding people, because it runs on reporting cycles rather than effort.
What Does Not Port
Some of what a SurveyMonkey Enterprise team relies on has no documented equivalent in Sprig, and a few things port only with restrictions. This section lists them before the procedure, because a reader who finds a blocking gap in week six has been misled by the first five weeks. Each row names what to do instead.
The table below compares SurveyMonkey’s published help center against docs.sprig.com as of November 11, 2026. "Not documented" means the capability is absent from Sprig’s published documentation, which isn’t the same as a confirmed absence. Ask Sprig directly about any row that’s load-bearing for your program.
| What you have in SurveyMonkey | Does it port | What you do instead | Source |
|:---:|:---:|:---:|:---:|
| Survey design files | No documented import | Annotate the design PDF and rebuild through AI File Upload | Both help centers |
| Historical responses | No documented import | Archive CSV and SPSS exports outside Sprig | docs.sprig.com export and CSV import pages |
| Slider question | Not documented | Rebuild as a Rating Scale and bridge the item | Question type pages |
| Click map question | Not documented | Image response options, or a prototype test with Recorded Task | Question type pages |
| File upload question | Not documented | Collect files through a separate secure channel | Question type pages |
| Date / time question | Not documented | Validated or open text, or pass the date as a user attribute | Question type pages |
| Name, email, phone, address fields | Partial | Pass identity as attributes, or use Validated Text where you must ask | Question type pages |
| Text and image A/B tests | Not documented as question types | Blocks showing a random subset, link surveys only | Blocks page |
| Same-page logic | Not documented | Separate pages with skip logic, or display logic on a multi-question page, tested | Logic pages |
| Dynamic redirects | Not documented | Route respondents after the survey from your own system | Enterprise feature list |
| Respondent authentication by SSO | Not documented | Unique links carrying a user identifier | Enterprise feature list |
| Offline kiosk mode | Not documented | Keep SurveyMonkey for offline collection | Enterprise feature list |
| EU or Canadian data center | Not documented, Sprig hosts on AWS in the USA | Stop and resolve residency before migrating | Enterprise feature list, sprig.com/security-standards |
| SCIM provisioning | Partial, deprovisioning only | Invite users manually or through domain discovery | SCIM pages |
| Industry Benchmarks | Not documented | Keep historical benchmark exports with the archive | Enterprise feature list |
Question Types With No Documented Equivalent
Four SurveyMonkey question types have no documented counterpart in Sprig: Slider, Click map, File upload and Date / time. SurveyMonkey documents 23 question types on its help center, and Sprig documents 15 plus an image display mode. Most of the gap sits in these four.
Slider is the one that catches trackers. A slider and a five-point rating scale aren’t the same instrument, because respondents use a continuous scale differently from labeled points. Rebuilding a slider as a Rating Scale is the right move, and the item still needs a bridge study (see The Measurement Discontinuity) before its trend can continue.
Click map has no direct equivalent. If the question asks which part of an image draws attention, image response options can recover a coarse version by turning regions into choices. If it’s really a usability question, a prototype test with Sprig’s Recorded Task type, which is Enterprise-only and limited to prototype tests, is the better instrument.
File upload has no documented Sprig question type. Collect files through a secure channel your security team already approves, and link the submission to the respondent’s ID. Date / time questions rebuild as Validated Text or Open Text with a format instruction, but if the date is something your systems already know, like a signup or purchase date, pass it as a user attribute and stop asking.
Survey Flow Constructs
SurveyMonkey’s Advanced Branching is where enterprise instruments break. It evaluates conditions across answers, Contacts data, custom variables and survey language, then skips, shows, hides or invalidates questions with a custom error message. Sprig documents skip logic, display logic, response piping and attribute piping, with display logic and both kinds of piping on Enterprise plans only.
Three SurveyMonkey constructs have no documented Sprig path. Same-page logic, which reveals a follow-up on the same page, has none: Sprig’s skip logic can’t redirect from one question to another on the same page. Dynamic Redirects, which send respondents to different URLs based on branching rules, have none. Invalidating an answer with a custom error message isn’t documented either.
Partial Ports Still Need Checking
A question type that exists on both sides isn’t automatically available in your study. Sprig restricts several types by plan and by study type, so a mapping that marks them as a clean match is misleading:
- Rank Order is Enterprise-only, limited to in-product surveys and Surveys, and takes 2 to 10 options
- MaxDiff, the nearest match for SurveyMonkey’s Best worst scale, is Enterprise-only and limited to Surveys in Standard Format, with 4 to 24 items
- Multi-question single pages are available on Surveys only
- Image response options are available on Surveys only and need an updated SDK for in-product use
- Dropdowns don’t support an "Other" choice
- Blocks, which carry randomized subsets, run on link surveys only
Check every partial row against your instrument’s delivery channel and your Sprig plan before you count it as ported. A Rank Order question that worked in a SurveyMonkey email survey ports to a Sprig link survey on Enterprise, not to every channel on every plan.
What Ports Cleanly
The list above is long, so keep it in proportion. Multiple choice, checkboxes, comment boxes, single textboxes, matrix and rating scale grids, and Net Promoter Score (NPS) all have documented Sprig counterparts with no plan restriction. Question and response-option randomization, carry forward and skip logic between pages all have documented equivalents. For most customer satisfaction, onboarding and product surveys, the rebuild is mechanical.
The decision rule for this section: if a row marked Not documented sits in a survey you can’t redesign, resolve it with Sprig or keep that survey on SurveyMonkey before you plan the rest of the migration.
Who Should Not Migrate
Some SurveyMonkey Enterprise customers shouldn’t move to Sprig, and it’s better to find that out here than after the contract has lapsed. If any of the conditions below is a hard requirement for your program, stop and resolve it before you plan anything else.
Teams That Need Data Residency Outside the United States
If your data has to stay in the European Union or Canada, don’t migrate on current documentation. SurveyMonkey’s Enterprise plan documents an EU Data Center that stores survey and response data in Ireland and a Canadian Data Center that stores it in Canada.
Sprig documents that it uses Amazon Web Services facilities in the USA to host its software, and it documents no regional hosting option. For a team under a residency requirement, that’s disqualifying rather than inconvenient.
Teams That Collect Offline or by Native SMS
If field staff collect responses on tablets without a connection, SurveyMonkey’s Offline Kiosk Mode does something Sprig doesn’t document at all. Keep SurveyMonkey for that work or find another offline tool.
The same applies to native SMS sending: SurveyMonkey’s Enterprise plan documents sending surveys by text message, while Sprig documents no native SMS delivery. A Sprig link can still be sent through your own SMS tool, which works for some programs and not for others.
Teams Whose Reporting Depends on Built-In Significance Testing
If stakeholders read significance flags straight off SurveyMonkey’s Compare Rules, plan for that loss before you migrate. SurveyMonkey publishes its method: a 95% confidence level, with at least 30 responses per answer option in the comparison. Sprig documents no significance testing. Teams that keep the move usually rerun those tests in a stats tool or an AI client working on the exported data, and somebody has to own that step.
Teams Mid-Way Through a Tracked Series With a Fixed Target
If a satisfaction or NPS number is tied to a bonus, a contract threshold or a board target this year, the migration will move that number for reasons unrelated to customers. You can still migrate, but only with a bridge study and the target owner’s agreement in advance. If you can’t get either, wait until the target resets.
Teams That Need Full SCIM Lifecycle Management
SurveyMonkey’s Enterprise plan documents SCIM to link your identity provider and track accounts. Sprig documents SCIM for deprovisioning only: it soft-deletes a user when your identity provider sends an inactive status, but it doesn’t create users or sync groups. If your security policy requires identity-provider-driven account creation, that’s a gap to raise with Sprig before signing.
The Pre-Migration Audit
The audit decides the scope of everything after it. A long-lived SurveyMonkey account often holds far more surveys than are actually in use, and rebuilding dead instruments is the easiest way to waste a migration budget. The audit’s job is to find the live ones, their owners and their dependencies, and to set the export deadline.
What to Inventory
For every survey in the account, across every user and every division, capture these fields in one spreadsheet:
- Survey title and ID, and the owner’s name
- Whether a collector is open, and the date of the last response
- Delivery channel: web link, email invitation, website embed, SMS or kiosk
- Question types used, flagged against the does-not-port table above
- Logic used: skip logic, Advanced Branching, quotas, piping, custom variables
- Languages fielded
- Response count and whether the survey is part of a tracked series
- Every downstream consumer of its data: dashboards, Salesforce, Tableau, Power BI, scheduled exports
The Audit Checklist, in Order
Run the audit in this sequence, because each step narrows the next:
- Find the contract end date and the clause that governs data after termination. Name the person who owns it.
- Export the account-wide survey list through the API or by account, including every division’s surveys.
- Mark each survey live, dormant (no responses in 12 months) or dead.
- For each live survey, record channel, question types, logic and languages.
- Identify every tracked metric and its owner, and ask whether a target is attached.
- List every system that reads SurveyMonkey data, and the person who’d notice if it stopped.
- Flag every survey that hits a does-not-port row, and decide its remedy now.
- Set the export deadline early enough before the contract end date to redo any export that fails.
Who to Interview During the Audit
The account data tells you which surveys are live. It doesn’t tell you why they exist. Spend half an hour with each division’s admin and the owner of every tracked metric, and ask three questions: which surveys would you miss if they stopped tomorrow, which reports depend on them, and which survey changes are already overdue.
The third question matters because a migration is the cheapest moment to fix a survey, as long as the fix goes into the bridge plan rather than happening silently.
The Two Items Nobody Checks
The first is who owns the contract clause. The person running the migration usually isn’t the person who signed the renewal, and the export deadline depends on terms only procurement or legal can read. Find them in week one.
The second is downstream consumers. A SurveyMonkey integration with Salesforce or a scheduled export into a dashboard often outlives the person who set it up. Asking "who would notice if this data stopped arriving" finds consumers that an integration list misses.
A Worked Example of Audit Triage
Consider a hypothetical account. A software company’s SurveyMonkey Enterprise team holds 412 surveys across four divisions. The audit finds 37 with an open collector and a response in the last quarter, 61 dormant and 314 dead. Of the 37 live surveys, nine feed a tracked metric, six use a question type from the does-not-port table, and four send data into Salesforce.
That triage changes the plan. The rebuild scope is 37 surveys, not 412. The archive scope is 98, the live and dormant ones, with the dead surveys exported only if legal asks. The bridge-study scope is nine.
The six surveys with non-porting question types get their redesign decided now, and the four Salesforce feeds get named owners before anything else moves. These figures are illustrative, but the shape is typical: the live set is a small fraction of the account.
Sizing the Rebuild From the Audit
Size the rebuild from the live list, the logic column and the Partial and Not documented flags. A survey with only Yes rows and simple skip logic is a short task.
A survey with Advanced Branching on custom variables, a quota and a non-porting question type is a design project with its own review. Sort the live list by that complexity, and rebuild one simple survey first to calibrate the team’s pace before committing dates for the rest.
The decision rule for the audit: a survey with no responses in 12 months and no downstream consumer gets archived, not rebuilt.
Exporting from SurveyMonkey
SurveyMonkey gives you two separate things to export: the survey design and the response data. They come out in different formats with different gaps, and the design export is weaker than most teams expect.
The Design Export Leaves Out the Logic
SurveyMonkey’s survey design export is a PDF. Its help center describes downloading "a PDF of the survey design in your default survey language," and lists what the PDF won’t include: logic, the menu choices in Dropdown and Matrix of dropdown menus questions, languages other than the default, the footer and the navigation buttons. Design PDFs are labeled a paid feature.
That omission list is the most important fact in this section. A PDF with no logic and no dropdown choices is an incomplete specification, and any rebuild working from it will miss branches and answer options without saying so. Before the PDF is useful, someone has to annotate it with the logic and the missing choices, working from the live survey’s Design Survey tab.
The API Route Gets the Full Structure
For accounts with many surveys, SurveyMonkey’s API is the more complete route. Its quick start describes an endpoint, /surveys/{id}/details, that "returns the survey’s design with all question ids and answer option ids." That JSON carries the answer options the PDF drops and gives you a machine-readable inventory.
Plan for the rate limit. SurveyMonkey’s API documentation says draft and private apps "will initially be subject to request limits of 120 calls/minute and starting at 500 calls/day." An account with 300 surveys and a bulk response export for each will take several days of calls at the starting limit, so either request a higher limit early or start the pull weeks before the deadline.
Bulk Responses by API
SurveyMonkey’s API also documents a bulk responses endpoint, /surveys/{survey_id}/responses/bulk, with a collector-level version. For accounts with dozens of live surveys, scripting the response pull against that endpoint is more reliable than clicking through exports one survey at a time, and it produces files you can check programmatically against each survey’s response count.
Keep the script with the archive so the export can be repeated if a file turns out to be incomplete.
Response Export Formats
SurveyMonkey’s export documentation lists individual responses in XLS, PDF, CSV and SPSS formats, and question summaries in PDF, PPT, XLS and CSV. The SPSS export is a .SAV file and is offered only for individual responses. Exports are a paid feature.
Export each live and dormant survey in two formats: CSV for portability and SPSS for preserved variable labels. The SPSS file keeps the question text and value labels attached to the data, which matters when someone opens the archive in three years without SurveyMonkey to explain what column Q14_3 meant.
Two Limits That Shape the Export Plan
SurveyMonkey states that when a survey has more than 100,000 responses, or when respondents multiplied by questions exceeds 100,000, "only CSV exports are available." A large tracker will hit that ceiling, and its SPSS labels have to be rebuilt from the design file. Completed exports stay under Previous exports for 14 days, so download each file when it’s ready.
Export the Context, Not Only the Data
A response file without its questionnaire is half an archive. For each survey, keep the design PDF, the API details JSON, the response exports, the collector settings and the custom variables used. Custom variables appear as columns in the individual responses export, but their meaning lives in someone’s memory unless you write it down.
The Contract-Clock Worksheet
Work every deadline backward from the contract end date, using only the windows SurveyMonkey documents. Fill in the dates for your account:
| Deadline | Set relative to | Why |
|:---:|:---:|:---:|
| Contract end date | Your order form | Starts SurveyMonkey's documented 30-day pause on non-renewal |
| Final export complete | Before the end date, with a margin you choose to redo failed files | The free plan can’t export after downgrade |
| Final export started | Before the completion date, by your API arithmetic | 500 calls a day is the documented starting limit |
| Parallel runs ended | Before the final export starts | The final export has to include the last overlap responses |
| Bridge studies started | Whole reporting cycles before the overlap ends | The bridge needs complete cycles |
| Rebuild of tracked surveys done | Before the bridge studies start | Bridges need the Sprig instrument |
| Audit complete | Before anything else | Sets every scope above |
The worksheet carries no durations on purpose. It holds the dependencies, and your account size and API limit supply the arithmetic. Download each completed export within SurveyMonkey’s 14-day Previous exports window.
The decision rule for exporting: if a survey’s data could be needed for a future comparison, export it in both CSV and SPSS with its design file, and verify the row count against SurveyMonkey’s response count before you close the task.
The Historical Data Decision
Your SurveyMonkey response history won’t move into Sprig, because Sprig documents no way to import historical survey responses. Decide where that history will live before the rebuild starts, because the decision changes what you export and how long you keep paying attention to the old account.
What the Documentation Shows
Sprig’s export documentation covers data leaving Sprig: CSV exports of up to 950,000 rows, delivered by an emailed link that expires after one week.
Sprig’s CSV import is documented for two purposes only: "Bulk uploading of user attributes" and "Bulk uploading of Users for creating Manual Groups." Neither brings in survey responses. Sprig’s API reference lists Retrieve Responses as a read endpoint, and no endpoint in the published index creates or imports responses.
That’s a documented absence as of November 11, 2026, not a promise it will stay that way. Check the current documentation before you plan around it, and ask Sprig directly if response import would change your design.
The Asymmetry, Stated Plainly
This cuts one way. Your SurveyMonkey history can’t come into Sprig, and your Sprig data starts on the day the first rebuilt survey launches. Any analysis that spans the switch has to join two exports outside both platforms. For a program that reports year-over-year trends, that join is a permanent part of the reporting workflow until the pre-migration years age out of the comparison window.
Where the History Should Live
The history needs a home that outlives the SurveyMonkey contract. Three options work, in order of preference:
- A data warehouse table per tracked series, with a platform column marking which rows came from which system
- A frozen export folder in governed storage, with CSV, SPSS and the design file for every survey
- A research archive in your existing document system, for one-off studies nobody will reanalyze
Trackers belong in option 1. One-off studies belong in option 2 or 3.
The Archive Is a Privacy Object
An archive of responses is personal data with a retention obligation, not just a file. Before you export, check what your privacy notice told respondents about retention, whether any survey collected protected health information under SurveyMonkey’s HIPAA terms, and who’ll own deletion requests for the archive after the SurveyMonkey account is gone.
If you collected health data, SurveyMonkey documents that non-renewing HIPAA accounts are suspended and their data is deleted when the suspension ends, so that export can’t wait.
Keeping the Two Series Joinable
The pre-migration and post-migration data only join cleanly if they share a schema. Before the first Sprig survey launches, write a mapping from each SurveyMonkey column to its Sprig export column, including how answer options are coded.
Sprig’s export lets you choose response text, numerical values or both, so export both from Sprig and keep SurveyMonkey’s numeric coding next to its labels. A join that matches on answer text breaks the first time someone fixes a typo in an option.
The EU Data Center Exception
If your SurveyMonkey account stores data in the EU Data Center, read this before the contract ends. SurveyMonkey documents that downgrading from Enterprise in that case requires a new US self-serve account and that "your data cannot be migrated." The export from the EU account is the only copy you’ll get.
What a Usable Archive Contains
An archive is worth keeping only if someone can use it later without SurveyMonkey. For each survey, it holds the individual responses in CSV and SPSS, the design PDF, the API details file with question and answer option IDs, a note defining each custom variable, the collector settings, the fielding dates, and the owner’s name.
For each tracked series, it also holds the bridge-study result that connects it to the Sprig series.
The decision rule: if any history might be needed after the contract ends, it leaves SurveyMonkey in readable form, with its questionnaire, early enough before the end date to repeat any export that fails.
Mapping Question Types
Every SurveyMonkey question type maps to one of three verdicts in Sprig: Yes, Partial or Not documented. A Partial verdict names its restriction, because a type that exists only on Enterprise plans or only on link surveys hasn’t ported for a team whose instrument runs somewhere else.
The Mapping Table
Plan gating and capabilities in this table were verified against each vendor’s published documentation on November 11, 2026.
| SurveyMonkey type | Sprig type | Verdict | Restriction or note |
|:---:|:---:|:---:|:---:|
| Multiple choice | Multiple Choice Single-Select | Yes | None documented |
| Checkboxes | Multiple Choice Multi-Select | Yes | "None of the Above" can't combine with Range or Maximum |
| Dropdown | Multiple choice in dropdown display | Partial | No "Other" choice |
| Matrix of dropdown menus | Matrix / Accordion | Partial | Dropdown cells not documented |
| Image choice | Multiple choice with images | Partial | Surveys only, image files up to 10MB |
| Star rating | Rating Scale | Partial | Icon style not documented, bridge if tracked |
| Slider | None | Not documented | Rebuild as Rating Scale |
| Ranking | Rank Order | Partial | Enterprise only, 2 to 10 options |
| Matrix / rating scale | Matrix / Accordion | Yes | Rows and columns locked after launch |
| Best worst scale | MaxDiff | Partial | Enterprise, Surveys in Standard Format, 4 to 24 items |
| Comment box | Open Text | Yes | AI theming available |
| Single textbox | Open Text or Validated Text | Yes | Validated Text not on Heatmap or Replay studies |
| Multiple textboxes | Multi-Question: Single Page | Partial | Surveys only |
| Name, Email, Phone, Address | Validated Text, or attributes | Partial | Prefer attributes for known identity |
| Date / time | None | Not documented | Validated Text or attribute |
| Net Promoter Score | NPS | Yes | None documented |
| Click map | None | Not documented | Image options or Recorded Task |
| File upload | None | Not documented | Separate secure channel |
| Text and Image A/B test | Blocks random subset | Partial | Link surveys only |
Rebuilding Logic and Survey Flow
Most SurveyMonkey skip logic rebuilds cleanly in Sprig. The trouble sits in Advanced Branching, same-page logic and anything that leans on custom variables, so rebuild the flow from a written map rather than from memory.
The Logic Mapping Table
| SurveyMonkey logic | Sprig equivalent | Verdict |
|:---:|:---:|:---:|
| Page skip logic | Skip Logic to a page | Yes |
| Question skip logic | Skip Logic to a question | Partial, not within the same page |
| Disqualification logic | Skip to survey end, or quota screen-out | Partial, custom page not documented |
| Advanced branching | Display Logic plus Skip Logic | Partial, display logic Enterprise only |
| Same-page logic | None | Not documented |
| Page, question and block randomization | Randomization and Blocks | Yes, Blocks on link surveys only |
| Carry forward responses | Carry Forward | Yes, not with image options |
| Quotas | Response-based quotas | Partial, Enterprise only |
| Custom variables | User attributes and attribute piping | Partial, piping Enterprise only |
| Question and answer piping | Response Piping | Partial, Enterprise only |
| Dynamic redirects | None | Not documented |
Translating Advanced Branching
Advanced Branching conditions fall into two kinds, and they translate differently. Conditions on earlier answers become Sprig display logic or skip logic on those answers. Conditions on Contacts data or custom variables become display logic on user attributes, which Sprig documents for Enterprise plans, provided the attribute values are in Sprig before the survey loads.
That last condition is the common surprise. A SurveyMonkey custom variable travels in the survey link, so it’s available the moment the respondent arrives. A Sprig attribute has to be stored first, through the SDK, the API or a CSV upload, and link surveys need unique links carrying a user identifier. If your branching depends on data passed at click time, plan the attribute load before you rebuild the survey.
A Worked Example of One Branching Rule
Take a SurveyMonkey Advanced Branching rule that reads: if the respondent’s custom variable plan equals "enterprise" and they answered 0 to 6 on the NPS question, show the churn-reason question. In Sprig, that becomes display logic on the churn-reason question with two conditions: the user attribute plan equals "enterprise," and the NPS answer is less than or equal to 6.
The rebuild has three tasks, not one. First, load plan as a user attribute through the SDK, the API or a CSV upload, because Sprig reads attributes stored before the survey loads rather than values in the link. Second, make the survey link unique per respondent so the attribute can be matched. Third, test both paths with a respondent on each plan.
Skip Logic Inside a Page
Sprig’s survey logic page states that skip logic can’t redirect respondents from one question to another on the same page. A SurveyMonkey page that skips between its own questions has to be split into separate pages, or rebuilt as a multi-question page with display logic. Test both versions, because splitting a page changes what the respondent sees at once, and that can move answers.
Quotas Move From One to Many
SurveyMonkey limits quotas to one per survey, in Simple or Combination form, and doesn’t allow them on surveys buying SurveyMonkey Audience responses. Sprig’s response-based quotas are Enterprise-only, carry no documented limit on their number, and use single-select multiple choice, rating scale or NPS questions as screeners.
Sprig’s documentation also says quotas can’t be deleted from an active survey and their rules can’t change after launch, so define them completely before you go live.
Carry Forward Has One Restriction
SurveyMonkey’s carry forward passes selected answer choices into later questions, and Sprig documents Carry Forward that does the same. Sprig’s question documentation notes one limit: image response options aren’t supported with carry forward, so a SurveyMonkey image choice question feeding a later question has to use text options in Sprig.
Randomization Is Close to Parity
SurveyMonkey offers page, question and block randomization. Sprig documents randomization of questions, blocks, block contents, random subsets and response options on all plans, and it respects skip logic, display logic, piping and carry-forward ordering. Blocks are available on link surveys only, so an in-product survey that relied on block randomization needs a check.
Disqualification Pages and End Screens
SurveyMonkey’s disqualification logic sends respondents to a custom disqualification page. In Sprig, route those respondents to the end of the survey with skip logic, or let a quota screener end the survey for anyone who doesn’t match an open quota. A custom disqualification page isn’t documented, so if the page carried instructions or a redirect, move that message into the survey’s closing text or handle it outside the survey.
The decision rule: write the logic map for each survey before the rebuild, and treat any condition on data passed at click time as an attribute-loading task first.
Rebuilding the Instrument in Sprig
The fastest documented rebuild route is Sprig’s AI File Upload, part of the Design Agent, which reads a questionnaire document and drafts a study from it. For teams that want the rebuild scripted, Sprig’s Study API and MCP connector create draft studies programmatically. Both still need a human check against the source.
The AI File Upload Route
Sprig’s AI File Upload accepts "your .pdf or .docx file," up to 25 MB, and it’s "Available on all enterprise plans." That pairs directly with SurveyMonkey’s design PDF, but only after the PDF is fixed. Because SurveyMonkey’s PDF leaves out logic and dropdown choices, uploading it as exported produces a draft missing exactly the parts that break instruments.
Convert the PDF to a Word document and annotate it in the format Sprig documents:
- Put the study title and objectives at the top
- Number every question (1., 2., 3.) and end each with a question mark
- Add REQUIRED:, SKIP LOGIC:, DISPLAY LOGIC: and RANDOMIZE: notes on the lines they govern
- Mark multi-question pages with [Page 1 Start] and [Page 1 End]
- Restore every dropdown and matrix-of-dropdown choice the PDF dropped
An Annotated Questionnaire, Line by Line
A rebuilt question in the annotated document looks like this, with the annotations on their own lines below the question:
- 4. How likely are you to recommend Acme to a colleague?
- REQUIRED: yes
- SKIP LOGIC: if 9 or 10, skip to question 6
- 5. What’s the main reason for your score?
- DISPLAY LOGIC: show only if plan attribute equals enterprise
The annotation is what turns SurveyMonkey’s logic-free PDF into a complete specification. Have the survey’s owner write it, since they know which branches matter, and keep the annotated file in the archive as the record of what the rebuild was meant to produce.
What AI File Upload Won’t Do
Sprig documents that MaxDiff questions and prototype links aren’t supported through AI File Upload and have to be added by hand in the survey builder. A SurveyMonkey Best worst scale question therefore gets rebuilt manually. Consent, Text/URL Prompt and Video/Voice questions can’t sit inside multi-question pages, so a SurveyMonkey page mixing those with other questions needs splitting.
The Scripted Route
Sprig’s Study API documents operations to create a draft study, create questions, create blocks and validate a study, and its MCP connector exposes the same workflow to 11 AI clients, including Claude, ChatGPT and Cursor.
Combined with SurveyMonkey’s /surveys/{id}/details endpoint, a team can map SurveyMonkey’s JSON to Sprig draft studies in a script. Studies created through MCP stay in draft until someone launches them, and each carries an AI-generated watermark.
This is a rebuild surface, not a converter. Nothing in either vendor’s documentation maps one schema onto the other, so the mapping logic is yours to write and test.
The Order That Works
Rebuild in this order: confirm the plan and channel restrictions for every Partial question, load any attributes the logic depends on, upload or script the draft, add the excluded questions by hand, then compare the draft line by line with the source design. That comparison is the step teams skip, and it’s where a dropped answer option gets caught before it breaks a trend line.
Rebuilding Multilingual Surveys
SurveyMonkey’s design PDF covers the default language only, so translations don’t travel with it. Sprig documents AI Translations that translate a whole survey, every question, choice and button label, with side-by-side review and editing.
Sprig also warns that regenerating translations overwrites manual edits, so finalize the source survey before generating them. For tracked surveys, reuse your approved SurveyMonkey translations rather than machine output, because a changed translation is a changed question.
The decision rule: no rebuilt survey launches until someone other than the rebuilder has checked every question, option and condition against the SurveyMonkey source.
The Measurement Discontinuity
Changing survey platforms changes the numbers. When a tracker moves from SurveyMonkey to Sprig, its rendering, its question presentation and often its delivery channel change at once, and the tracked metric moves for reasons unrelated to what customers think. Reporting that movement as a finding is the most expensive mistake in this guide.
Two Effects Get Confounded
A platform switch mixes two effects that survey methodology keeps separate. A measurement effect is a change in how people answer the same question because of how it’s presented. A selection effect is a change in who answers, because a different channel reaches a different population.
Moving from an emailed SurveyMonkey link to a Sprig in-product survey changes both. An in-product survey reaches people who are signed in and active right now. An email list reaches dormant users and recently churned customers as well. No published study estimates that difference for your product, and the only way to separate the two effects is a comparable dataset collected alongside the old one.
How Large Mode Effects Can Be
Pew Research Center measured mode effects directly in a 2015 study, "From Telephone to the Web." It randomly assigned 3,003 respondents to phone or web between July 7 and August 4, 2014, and compared 60 questions. Pew reported "a mean difference of 5.5 percentage points and a median difference of five points across the 60 questions."
Two cautions apply. First, Pew’s comparison was phone against web, and a SurveyMonkey-to-Sprig move is usually web against web, so treat 5.5 points as evidence that presentation moves answers, not as a forecast. Second, Pew randomized the samples, which holds the population constant. Your migration doesn’t.
The Items That Move Most Are Satisfaction Items
Pew’s largest differences landed on quality-of-life items. On satisfaction with family life, 62% of phone respondents said they were "very satisfied" against 44% of web respondents, an 18-point gap on one item. That’s the family of questions customer satisfaction, NPS and brand trackers live in, which makes them the items most exposed to a change in presentation.
Pew is direct about what that gap means: "there is no way to determine whether the telephone or the Web responses are more accurate." A post-migration number isn’t automatically the wrong one. It’s a different measurement, and without an overlap you can’t tell which way the truth lies.
Which of Your Metrics Are Exposed
Sort your tracked metrics by exposure before planning any overlap. The most exposed are subjective evaluations reported as a level: satisfaction scores, NPS, brand favorability and anything with a target attached. Less exposed are behavioral or factual items, like which features someone used, and metrics read only for direction month over month. A metric that’s both subjective and tied to a target goes to the top of the bridge-study list.
House Effects Are the Closer Analogue
Diana Schumann, Hawal Shamon and Jürgen-Friedrich Hake, writing in the International Journal of Public Opinion Research (doi 10.1093/ijpor/edz039), describe house effects: systematic differences tied to which organization fielded a survey. Those differences can appear "even if the survey agencies use identical question wordings or target populations." Same questions, same people, different fielding infrastructure, different numbers. That’s the closest published description of a platform switch.
Two Precedents, Two Shapes
Ahead of the 1994 redesign of the Current Population Survey, the US Bureau of Labor Statistics ran a Parallel Survey from July 1992 to December 1993, on a monthly sample about one-fifth the size of the regular survey.
Its authors estimated the redesign raised the measured unemployment rate by .56 percentage point. A national tracker moved by more than half a point, and it took 18 months of parallel collection to show it.
The University of Michigan’s Surveys of Consumers moved from phone to web in 2024 after "over seven years of running a parallel data collection via the web alongside the official phone data collection." The crossover was phased: 25% web in April, half in May, 75% in June and all web from July.
Michigan reported a "remarkably strong 0.97 time series correlation" for its Index of Consumer Sentiment, and also that web estimates for the Current Conditions Index ran "consistently lower than (though parallel to) phone estimates."
What the Precedents Teach
A high correlation and a level shift aren’t in conflict. Correlation tracks whether the series moves together, and it says nothing about whether one runs below the other. So an overlap doesn’t prevent a break. It lets you measure the break and explain it.
The Trade-Off, Stated Plainly
The methodologically correct answer is to run SurveyMonkey and Sprig in parallel on every tracked metric. You’re migrating partly to stop paying for and running two platforms. Both things are true, and pretending otherwise produces a plan nobody follows.
There’s also no published rule for how long the overlap should be. The US Office of Management and Budget’s Standards and Guidelines for Statistical Surveys tells agencies to "Develop adjustment methods, such as crosswalks and bridge studies that will be used to preserve trend analyses and inform users about the effects of changes," and names no duration.
The precedents above range from 18 months to seven years. Anyone quoting a standard overlap of a few months is quoting a convention, not a source.
The decision rule: if anything downstream depends on the level of a metric rather than its direction, that metric needs an overlap, and its owner needs to agree to the plan before the switch.
Designing the Bridge Study
A bridge study runs the old and new instruments side by side so you can measure the gap between them before the old one stops. It’s the smallest version of the overlap that still lets you defend a tracked number after the move.
What to Hold Constant
Keep everything you can identical across both platforms:
- Question wording, answer options and order
- Scale length and labels, including the endpoints
- Sampling frame and channel, where the business allows it
- Fielding window, so both instruments see the same weeks
- Incentives and invitation copy
What to Vary
Vary only the platform. If the business also wants to change the channel, for example from email to in-product, change it in a second, later step. Two changes at once leave you with one gap and no way to split it.
The Bridge-Study Specification
Fill in one specification per tracked metric:
| Field | What to record |
|:---:|:---:|
| Metric and owner | The number, who reports it, and to whom |
| Level or direction | Whether anyone acts on the level, or only on change |
| Split | How respondents are assigned to each platform, randomized if possible |
| Sample per arm | Sized from your own power calculation for the smallest gap you'd act on |
| Window | Whole reporting cycles, never a partial one |
| Comparison | Mean or top-box difference with a confidence interval, per item |
| Decision threshold | The gap below which you'll declare the series continuous |
| Fallback | What you publish if the gap exceeds the threshold |
The Frame Problem
The hardest variable to hold constant is the sampling frame. If the SurveyMonkey version went to an email list and the Sprig version will run in-product, the two arms reach different people, and no amount of matching on wording removes that.
Where the business can tolerate it, run the bridge on the old channel first, through Sprig email to the same list, and move the channel in a second step with its own comparison.
Sizing It
Size each arm from a power calculation for the smallest difference your stakeholders would act on, run in a stats tool against your own historical variance. No published figure replaces that calculation, and borrowing another program’s sample size borrows its variance too.
The Decision Rule Nobody Writes Down
Write the threshold down before the data arrives. If the gap between platforms falls inside the threshold for every tracked item, declare the series continuous and say so in the next report. If it doesn’t, start a new series in Sprig, report both with a marked break, and tell stakeholders why the number moved.
A Worked Bridge Result
Here’s a hypothetical result to show the mechanics. A team splits one month’s NPS sample between SurveyMonkey and Sprig, about 1,200 respondents per arm. SurveyMonkey returns 38 and Sprig returns 35, with a 95% confidence interval on the difference that spans zero.
The team’s written threshold was a difference under five points with an interval spanning zero, so it declares the series continuous and notes the bridge in the next report. Had Sprig returned 30, the team would have started a new series. The numbers are invented for illustration, and the threshold has to come from your own decision context.
When the Bridge Shows a Break
A break isn’t a failure. Report the pre-migration and post-migration series separately, label the break in every chart, and publish the measured gap with the bridge study’s dates. If a target was attached, rebase it on the Sprig series with the target owner’s agreement.
The decision rule: no tracked metric switches platforms without a written threshold and a named owner who’s agreed to it.
Users, Roles, SSO and Provisioning
Single sign-on ports cleanly from SurveyMonkey to Sprig, and full provisioning doesn’t. Plan the identity cutover around that difference, because it decides how new joiners get access after the switch.
SCIM Is Deprovisioning Only
Sprig’s SCIM documentation is specific. SCIM "lets your identity provider deprovision people who already have a Sprig account," and "Sprig does not create users through SCIM." Group import isn’t supported, passwords don’t sync and name updates aren’t available yet. Okta and Microsoft Entra ID are the verified providers, and Sprig notes that Google doesn’t currently support SCIM. SCIM is an Enterprise plan feature.
SurveyMonkey’s Enterprise plan lists SCIM to link your identity provider to track accounts. If you use it for joiners today, your joiner process changes: new users arrive by invitation, domain discovery or SSO, and SCIM handles their removal.
Divisions and Workgroups
SurveyMonkey Enterprise organizes teams into Divisions, with their own admins and libraries, and Workgroups around shared projects. Map each division to a Sprig workspace, list its admins, and decide who owns each library of saved questions, since a custom question bank has no documented import.
The Identity Cutover Sequence
Run the identity work in this order:
- Export SurveyMonkey’s user list by division, with roles
- Configure SAML single sign-on in Sprig with your identity provider
- Set up domain-based discovery or invitations for the users who need access
- Assign workspace roles to match each SurveyMonkey division’s admins and members
- Configure SCIM deprovisioning in Okta or Entra ID and test it on one account
- Remove SurveyMonkey access only after cutover, so survey owners can still export
Sprig’s SCIM page notes that deactivating the only organization admin returns an error, so make sure at least two admins exist before you test deprovisioning.
The decision rule: if your security policy requires identity-provider-driven account creation, resolve it with Sprig before cutover, and otherwise set up SSO first and SCIM deprovisioning second.
Integrations and Downstream Data
Inventory the consumers of your SurveyMonkey data, not just the integrations. A dashboard fed by a scheduled export breaks as silently as a native connector, and the person who’d notice is usually not on the migration team.
The Integrations SurveyMonkey Enterprise Lists
SurveyMonkey’s Enterprise feature list names integrations with Salesforce, Microsoft Power BI, Tableau and Marketo, plus direct API access. For each one in use, record what data flows, which reports read it, and who owns those reports. Then check Sprig’s current integrations documentation for each destination rather than assuming parity, because a connector that exists may carry different fields.
Getting Data Out of Sprig Continuously
Sprig documents three routes for downstream data. CSV export handles up to 950,000 rows per file. The API’s Retrieve Responses endpoint returns up to 1,000 responses per request. The MCP connector reads responses, themes and summaries up to 1,000 records at a time for teams working in an AI client. For a warehouse pipeline, plan the API pull around those 1,000-response pages.
Email Distribution Has Its Own Setup
If your SurveyMonkey program sends email invitations, Sprig’s email delivery needs setup before the first send. Sprig documents email distribution for Enterprise plans on link surveys only, from a custom sending domain an admin verifies with DNS records.
Contact lists upload as CSV files with Email, First Name, Last Name and User ID columns. Unsubscribes join a team-wide Do Not Contact list, and a 24-hour cooldown caps each person at one Sprig email a day.
There’s no documented automatic reminder. A follow-up to non-respondents is a second Email Run. Sprig also notes that panel and email distribution can’t both be active on the same survey.
Respondent Identity Carries Over Differently
SurveyMonkey custom variables and Contacts data travel with the survey link or the contact list. Sprig identifies respondents through a user ID and stored attributes.
Before you rebuild, decide which identifier joins a respondent across systems, usually your internal user ID, and make sure every Sprig survey carries it. Without a shared ID, the bridge study and every downstream join have to match on email, which breaks when people change addresses.
The decision rule: every downstream consumer gets a named owner and a tested replacement feed before the SurveyMonkey source stops.
The Parallel Run
A parallel run keeps SurveyMonkey collecting while Sprig starts, for the surveys that need it and no others. Done selectively, it’s affordable. Done across the whole account, it doubles the work of the migration’s most expensive months.
What Needs an Overlap
Only three kinds of survey need a parallel run:
- Tracked metrics that feed a bridge study
- Surveys with a downstream consumer that can’t tolerate a gap
- Surveys whose rebuilt version changed a question type or logic path
Everything else cuts over directly.
How to Keep the Overlap Affordable
Split rather than duplicate. Send half the sample to each platform instead of surveying everyone twice, which keeps respondent load flat and gives the bridge study its randomized arms. Keep the overlap to whole reporting cycles, and end it when the bridge study’s decision threshold is met, not on a fixed date.
What to Compare During the Overlap
Compare more than the headline number. For each overlapped survey, track the response rate, the completion rate, the distribution of answers on every tracked item, and the share of open-text answers that are blank or very short. A shift in completion rate is often the first sign that the rebuilt instrument behaves differently, and it shows up before the headline metric has enough responses to compare.
Keeping Respondent Load Flat
Respondents shouldn’t receive both versions. Randomize each person to one platform, suppress the other invitation for them, and keep your contact-frequency rules in force across both systems. Sprig’s 24-hour email cooldown applies only inside Sprig, so a person can still get a SurveyMonkey invitation and a Sprig invitation on the same day unless you manage the lists together.
Sequencing Against the Contract
Work backward from the contract end date. The parallel run has to finish before the final export, and the final export has to finish with enough margin before the end date to repeat a failed file. If the arithmetic doesn’t fit, negotiate a short extension or a read-only period with SurveyMonkey before you start, not after the overlap runs long.
The decision rule: overlap only the surveys whose numbers someone defends, and end each overlap on its bridge result.
Cutover and Decommissioning
Cutover is the date each survey stops collecting in SurveyMonkey and starts in Sprig. Decommissioning is the date the SurveyMonkey account stops existing. Keep the two apart by weeks, not days.
The Cutover Sequence
For each survey, in order:
- Confirm the Sprig version passed its validation check (see Validating the Migration)
- Update every link, embed and email template to point at Sprig
- Close the SurveyMonkey collector
- Run a final response export and check its row count against SurveyMonkey’s total
- Confirm the first Sprig responses are arriving and reaching every downstream consumer
What to Keep Before Decommissioning
Before the account closes, keep the final exports, design PDFs and API details files, collector settings, user lists by division, and the integration inventory. Keep any benchmark reports SurveyMonkey’s Industry Benchmarks produced, because they won’t be retrievable later.
The Deletion Decision
Decide in writing whether you’ll ask SurveyMonkey to delete your data at contract end or let its retention schedule run, and record who made the call. That decision belongs to your privacy owner, not the migration lead.
Communicating the Cutover
Tell three groups before each cutover:
- Survey owners, who need to know where their survey now lives
- Report consumers, who need the date the data source changes and any series break
- Support teams, who’ll get questions if a respondent clicks an old link
A closed SurveyMonkey collector shows its own closed message, so update any public page or email template that still points at it.
The Rollback Plan
Keep a rollback path for each survey until its first Sprig reporting cycle closes cleanly. That means the SurveyMonkey survey stays intact with its collector closed rather than deleted, and the old links stay recorded. If the Sprig version fails validation after launch, reopen the collector and fix the rebuild, rather than patching a live tracker.
The decision rule: the SurveyMonkey account closes only after every survey’s final export is verified and every consumer reads from Sprig.
Validating the Migration
Validation checks four things: the instrument, the data, the measurement and the access. A migration that passes the first and skips the rest has moved surveys, not a program.
The Validation Checklist
| Area | Check | Owner |
|:---:|:---:|:---:|
| Instrument | Wording, options, order, logic paths, required flags, languages | Research lead |
| Data | Archive row counts, SPSS labels, custom variable notes | Analytics owner |
| Measurement | Bridge window complete, gap compared with threshold, decision recorded | Metric owner |
| Access | Roles correct, leavers removed, SCIM deprovisioning tested | IT admin |
| Consumers | Every downstream report reading Sprig data | Report owners |
Each line of that checklist is explained below, and the prose carries the detail the table compresses.
Instrument Fidelity
For each rebuilt survey, compare against the SurveyMonkey source:
- Every question present, with identical wording
- Every answer option present, in the same order unless randomized
- Every skip, display and piping condition producing the same path
- Required flags matching
- Every language version present and reviewed by a fluent speaker
Test the paths by taking the survey yourself through each branch, not by reading the logic panel.
Data Integrity
Check that each archive file’s row count matches SurveyMonkey’s response count, that the SPSS labels open correctly, and that custom variable columns are documented. Then check that the first Sprig export carries the fields your downstream reports expect.
The Measurement Check
For each tracked metric, confirm the bridge study ran its full window and the result was compared against the written threshold. Record the decision and the measured gap.
The Access Check
Confirm everyone who needs access has it at the right role, everyone who left during the migration doesn’t, and SCIM deprovisioning works by testing it on one account.
A Soft Launch Before Full Cutover
Before switching a high-volume survey fully to Sprig, send it to a small share of its usual audience for one cycle. Check completion rates, open-text quality and every logic path in the real responses, then compare the response distribution with the same period in SurveyMonkey.
A soft launch catches the problems a test run misses, like a branch nobody took in testing or an attribute that’s missing for a segment of users.
Final Approval
Get one named approval per area: instruments from the research lead, data from the analytics owner, measurement from each metric owner, and access from IT.
The decision rule: a survey isn’t migrated until its validation row is signed, whatever the cutover date says.
Common Mistakes
These are the mistakes that turn a planned migration into an emergency, roughly in the order teams make them.
Treating It as an Import
There’s no documented import on either side, so every hour budgeted for "importing" surveys has to come from somewhere else once the team finds out. Budget a rebuild from day one.
Uploading the Design PDF Unannotated
SurveyMonkey’s design PDF leaves out logic and dropdown choices. Feed it to AI File Upload as exported and the draft looks complete while missing its branches. Annotate first.
Exporting Too Late
Teams often leave the archive until the final weeks, then hit the 500 calls a day starting API limit, the CSV-only ceiling on large surveys, and the 14-day window on finished exports at once. Start the export well before you think you need to.
Changing the Channel and the Platform Together
Moving a tracker from SurveyMonkey email to a Sprig in-product survey changes the platform and the population at the same moment. The bridge study then measures one gap with two causes. Change one thing at a time.
Reporting the Step Change as a Finding
The first Sprig wave of a tracker will probably land at a different level. Reporting that as customers changing their minds, without a bridge study, turns a measurement artifact into a business conclusion.
Forgetting the Downstream Consumers
The Salesforce field, the Tableau dashboard and the weekly export to finance all stop when SurveyMonkey does. Find them in the audit, not after cutover.
Cutting Over Before Identity Is Ready
A cutover that happens before single sign-on and roles are set up leaves survey owners locked out of their own rebuilt surveys on launch day. Finish the identity sequence first, and keep SurveyMonkey access for survey owners until their final exports are done.
Deleting SurveyMonkey Surveys at Cutover
Closing a collector and deleting a survey aren’t the same step. Deleting removes the rollback path and any chance to re-export a file that turns out to be incomplete. Close collectors at cutover and leave deletion to the decommissioning decision.
Rebuilding Dead Surveys
An account with hundreds of surveys usually has a few dozen in use. Rebuilding the rest spends the migration budget on instruments nobody will field.
Trusting the Draft Without a Line Check
An AI-drafted rebuild is fast, which makes it easy to trust too much. A dropped answer option produces a broken trend line that typically surfaces a quarter later.
Letting AI Follow-Ups Run on a Tracker
AI-generated follow-up questions change the instrument from one respondent to the next. That’s useful in exploratory research and a problem on a tracked survey, where the instrument has to stay fixed for the bridge study to mean anything. Turn follow-ups off on tracked surveys until the bridge is complete.
Skipping the Translation Review
SurveyMonkey’s design PDF covers only the default language, and machine translation produces wording that differs from your approved versions. A tracked survey fielded in a new translation has changed its question in that market. Reuse approved translations and have a fluent reviewer check each one.
Losing the Custom Variables
SurveyMonkey custom variables appear as columns in the response export, but their meaning isn’t written anywhere. A year later, nobody remembers what src=3 meant. Document every custom variable with the archive before the account closes.
How Sprig Supports the Migration
Sprig supports each of the four migration projects through documented features, with limits worth knowing before you plan around them.
For the rebuild, AI File Upload in the Design Agent turns an annotated questionnaire into a draft study, accepting PDF or Word files up to 25 MB on Enterprise plans. It doesn’t handle MaxDiff or prototype links, and researchers remain responsible for checking every drafted question against the source. The Study API and the MCP connector suit teams that want to script the rebuild.
For fielding, Sprig supports link surveys, in-product surveys on websites and web apps, native mobile app surveys, email on Enterprise link surveys, and research panels for reaching people outside your customer base.
The Field Agent asks contextual follow-up questions as people respond, and researchers can disable AI follow-ups for any study where a tracked instrument has to stay fixed.
For analysis, the Synthesize Agent groups responses into themes, shows how many responses support each theme, and keeps every theme traceable to the responses behind it. Sprig documents no significance testing, so the bridge study’s statistics still happen in a stats tool or an AI client working on the export.
For teams that work in AI clients, the MCP connector reads responses, themes and summaries up to 1,000 records at a time and creates draft studies that stay in draft until a person launches them. Admins turn it on and can disable it for the whole organization at any time. Researchers remain responsible for validating anything an AI client drafts or concludes from the data.
Where to Start
Start with one live, low-risk survey that uses only Yes rows from the mapping table. Rebuild it through AI File Upload, check it line by line, field it to a small audience, and compare the results with its SurveyMonkey history. That first survey tests the team’s annotation format, the attribute loading and the export join before any tracked metric depends on them.
Frequently Asked Questions
Can I Import My SurveyMonkey Responses into Sprig?
No, Sprig documents no way to import historical survey responses from SurveyMonkey or any other source as of November 2026. Sprig’s CSV import covers user attributes and Manual Groups, and its API’s response endpoint is read-only. Export your SurveyMonkey responses in CSV and SPSS formats, with each survey’s design file, and keep that archive in a data warehouse or governed storage outside both platforms.
Can I Export a SurveyMonkey Survey Design?
Yes, SurveyMonkey exports a survey design as a PDF, but its help center says the PDF won’t include logic, dropdown and matrix-of-dropdown menu choices, or languages other than the default. For a complete structure, SurveyMonkey’s API returns each survey’s design with question and answer option IDs from its survey details endpoint. Use the API output to fill in what the PDF drops.
What Is the Fastest Way to Rebuild a SurveyMonkey Survey in Sprig?
The fastest documented route is Sprig’s AI File Upload, which drafts a study from a PDF or Word questionnaire up to 25 MB on Enterprise plans. Annotate the SurveyMonkey design PDF first with SKIP LOGIC, DISPLAY LOGIC, RANDOMIZE and REQUIRED notes and the missing dropdown choices, then add MaxDiff questions by hand, and check the draft line by line against the source.
What Doesn’t Transfer from SurveyMonkey to Sprig?
Survey designs and historical responses don’t transfer, because neither has a documented import. Slider, Click map, File upload and Date / time questions have no documented Sprig equivalent. Same-page logic, dynamic redirects, respondent authentication, offline kiosk mode, native SMS sending and EU or Canadian data center hosting aren’t documented in Sprig either. Each has a remedy in this guide’s does-not-port table.
Which SurveyMonkey Question Types Map Directly to Sprig?
Multiple choice, checkboxes, comment boxes, single textboxes, matrix and rating scale grids, and Net Promoter Score questions map directly to documented Sprig question types with no plan restriction. Ranking, Best worst scale, image choice, dropdowns and multiple textboxes map with restrictions, several of them Enterprise-only. Slider, Click map, File upload and Date / time questions have no documented Sprig equivalent and need redesigning.
What Happens to My SurveyMonkey Data When the Contract Ends?
SurveyMonkey documents that a non-renewed Teams or Enterprise account is paused for 30 days, then users are removed and accounts drop to the free Basic plan. Surveys stay only in each creator’s individual account, responses over the Basic limit are deleted under SurveyMonkey’s retention period, and the free plan can’t export. Export everything well before the end date, with time to repeat any failed file.
Will My NPS or CSAT Score Change When I Migrate?
It probably will, and the change won’t necessarily reflect customers. A platform switch changes presentation and often the channel, and Pew Research Center measured a mean 5.5-point mode difference across 60 questions in a randomized 2015 study, with satisfaction items moving most. Run a bridge study with a written decision threshold before reporting any post-migration level.
How Long Should SurveyMonkey and Sprig Run in Parallel?
No published rule sets the length. The US Office of Management and Budget recommends bridge studies to preserve trends without naming a duration, and precedents range from 18 months for the Current Population Survey to over seven years for the University of Michigan’s consumer surveys. Size the overlap from a power calculation and whole reporting cycles, and end it on the bridge result.
How Long Does a SurveyMonkey Migration Take?
No published figure applies, because the length depends on how many surveys are live, how complex their logic is, your API limit and the reporting cycles your bridge studies need. The rebuild is usually the shortest part. The bridge studies set the floor for any program with tracked metrics, since they run in whole reporting cycles. Start from the contract end date and work backward.
Can We Run SurveyMonkey and Sprig at the Same Time?
Yes, running both is how a bridge study works, and it’s the recommended approach for any tracked metric during the migration. Randomize each respondent to one platform, suppress the other invitation, and keep contact-frequency rules in force across both systems. Run only the surveys that need an overlap in both places, and cut everything else over directly to keep the cost down.
Does Sprig Support SSO and SCIM?
Sprig documents SAML single sign-on on Enterprise plans, with Auth0, Google Workspace, KeyCloak, Microsoft Entra ID, Okta and OneLogin. Sprig’s SCIM is an Enterprise feature for deprovisioning only: it removes users your identity provider marks inactive but doesn’t create users or sync groups. Okta and Microsoft Entra ID are the verified SCIM providers.
Can I Keep My SurveyMonkey Data in the EU After Migrating?
No, not inside Sprig on current documentation, because Sprig hosts its software on AWS facilities in the USA and documents no regional hosting option. SurveyMonkey’s Enterprise plan documents EU and Canadian data centers. If residency is a hard requirement, resolve it before migrating, and note that SurveyMonkey says EU data center data can’t be migrated to a US account after downgrade.
Do SurveyMonkey Quotas Work the Same Way in Sprig?
Not exactly. SurveyMonkey allows one quota per survey, in Simple or Combination form. Sprig’s response-based quotas are Enterprise-only, have no documented limit on their number, and screen on single-select multiple choice, rating scale or NPS questions. Sprig quotas can’t be deleted from an active survey or edited after launch, so define them fully before going live.
Should We Migrate at All?
Teams whose surveys map mostly to Yes and Partial rows in this guide’s tables can migrate with a rebuild, an archive and a bridge study. Teams that need EU or Canadian data residency, offline kiosk collection, native SMS sending, built-in significance testing or full SCIM provisioning should resolve those requirements with Sprig first, and shouldn’t migrate if they can’t.