Last updated: October 2026
Changelog
We summarize notable product improvements and fixes so administrators and builders know what shipped. Severity and rollout timing vary by tenant; private instances receive updates according to their maintenance agreement.
October 2026 — Online diaries, home use tests and research communities (Pro & Corporate)
Three new ways to study people over days and weeks, included with the Pro and Corporate plans and with Enterprise and private instances. Online diaries: qualitative and quantitative diaries with photo, video and voice-note answers, schedules by study day, fixed dates or always open in each participant’s own time zone, repeats with late grace, event logging, email reminders, and follow-up questions on any entry. Home use testing: blind-coded products, one product per period, balanced product order with a Williams Latin square (or random or fixed order), gated tasks such as a “product received” check, and per-product results with a full rotation export. Research communities: private, invite-only spaces with nine discussion formats, five visibility modes (open, blind until first post, moderator-only, private and group-restricted), moderation, groups, and diaries and home use tests inside the community. Researchers get one Research inbox, a Member 360° profile, combined completion and engagement metrics, search, AI summaries with links to the source entries, and CSV/Excel exports. Participants take part in a mobile portal in English, Arabic and Bahasa Indonesia, with no app to install.
October 2026 — Survey scripting toolkit and exact Word imports
Variables in the Logic tab: counts of selected options, sums, averages, minimum/maximum, today's date, age and days between dates, and if/then/otherwise values for hidden "dummy" questions—pipe them with {name}, use them in rules and save them with each response (dates as dd.mm.yyyy). Randomization now keeps chosen options, grid rows and grid columns in place, grid columns can be shuffled, and the order each respondent saw is exported. New rank-all requirement, choices copied from several questions, typed "Other" text piped like any answer, and bold, italic and underline formatting in question and answer text. Word, PDF and Excel questionnaires now import exactly as written: original question numbers as names, original answer codes, grid rows named Q20_1, Q20_2…, scripting notes kept, and nothing added.
October 2026 — Numeric allocation and show-N-of-M panels
Numeric allocation questions (e.g. split 100 points) with per-item and total limits, a live running total and clear errors, analysed as numbers in results, AI reports and PowerPoint. Panels can show only N of their questions per respondent, at random or balanced across respondents, with the selection exported. Copied choices can exclude options and add the question's own; grids can include an "Other (please specify)" row.
September 2026 — Sentink Field mobile app
Sentink Field brings survey collection to Android for field teams: download surveys, collect offline with native screens for every question type (grids, signature, file upload, image picker and more), optional voice recording and location, and sync when back online. Downloaded surveys update automatically, the app supports Bahasa, and Continue with Google now works in the app and on the web.
September 2026 — Qualitative research upgrades
Transcript-only studies: upload video or audio and get an AI transcript with speaker turns, de-duplication and translated downloads. AI qualitative reports export to PDF and PowerPoint, discussion guides can be generated from a brief or uploaded document in several languages, and deleted studies go to Trash for restore or permanent deletion.
September 2026 — Open-ended coding, screen-out logic and panel redirects
Open-Ended Codes studio builds a codebook and codes free-text answers for surveys and qualitative studies. The Logic tab adds a Screen out action for disqualifying respondents, panel redirect settings send completes and screen-outs back to sample providers. A refreshed app-style respondent design is used across the live survey, preview and theme designer.
August 2026 — Live qualitative research
Run live online interviews and focus groups in Sentink: lobby, video room, chat, live transcription and subtitles, discussion guide panel and in-session activities, with recordings, AI transcript enhancement and AI reports. Student plans were also added.
August 2026 — Image hotspot questions and heatmaps
Added Image hotspot as a first-class question type: respondents click a point on an uploaded image (pack shot, ad, UI screenshot, or store plan). Each response shows the photo with the selected marker; Visual Insights overlays every click as a heatmap so packaging, creative, and UX tests stay visual. Available in English and Arabic, including RTL layouts.
May 2026 — AI Survey Agent, Conjoint, and MaxDiff
Clarified and shipped AI Survey Agent as one built-in experience: draft surveys from a natural-language prompt or uploaded brief, then continue with the same AI in the builder for structure, advanced blocks, and alignment with AI Quality Check and iterative editing. Added Conjoint-style choice flows with in-product analysis for bundles and preferences. Added MaxDiff (most vs least important) tasks with aggregate scores and results diagnostics—ideal for ordered priority lists alongside pricing and packaging research.
What this page covers
Highlights for the Sentink web application: builder, respondent experience, dashboards, integrations, Arabic and RTL behavior, governance, exports, and security-related improvements. Routine fixes may be grouped.
Where you see releases first
Many changes appear in staging or trial environments before general availability on Sentink Cloud. Enterprise customers sometimes receive previews when agreed.
Older history
As the product grows, older entries may be archived for clarity—contact support if you need details about a legacy change.