Spradley vs Microsoft Viva Glint: what each tells you

Microsoft Viva Glint is an engagement survey platform run inside Microsoft 365, and through its Copilot Impact survey it shows how people rate Copilot. Spradley tells you why, and helps you act on it. It runs one-on-one AI interviews by text or voice, asks a follow-up question on every answer, and reports themes with quotes and reasons within days. A Personal Agent for each leader then answers questions with cited evidence, launches the follow-up interview and reminds you of what you committed to, including from inside Microsoft 365 Copilot. It never sees one person’s answer. If you need to understand a result and change it, Spradley is the stronger choice.

Quick verdict

What is Microsoft Viva Glint?

Viva Glint is Microsoft’s employee engagement survey product, run from inside your Microsoft 365 tenant. It supports ad hoc, recurring, lifecycle and always-on survey programmes, and it can send survey notifications through Teams.

Glint also has Copilot built in. According to Microsoft’s guide for managers, Copilot Highlights summarises a team’s scores and changes, and comment summarisation groups up to 8,000 comments into at most 10 themes. It gives no result when the comment count is below the confidentiality threshold.

What is Spradley?

Spradley runs one-on-one AI-moderated interviews, by text or voice. People get an invitation by email or Slack and answer on their own phone or computer, in their own time. The interviewer asks a follow-up question on every answer, so you hear the reasons, not just the rating.

Within days you get a report: themes with the quotes behind them and the reasons behind each theme, compared across departments, and followed across interview rounds over time. Export it to PowerPoint, PDF, Word or Google Docs. Numbers and ratings are never shown for fewer than five people, and a privacy judge checks each written finding. The database is in Frankfurt and the AI runs in Google Cloud’s EU regions.

On top of the reports sits a Personal Agent for each leader, working on what their people said. A leader asks a question in plain language and gets an answer that cites the themes, quotes and reports it rests on. The agent runs the numbers, draws the chart and drafts the next step, from a message to the team to a deck for the board.

What should you ask before you choose?

Does it ask why?

Glint asks everyone the same items on the same scale, so you can compare a sales team in Munich with one in Leeds, and this year with last. Comment boxes add detail, and Copilot sorts the comments into themes.

Spradley adapts to each person. If a nurse says handovers have got worse, the interviewer asks what changed, when, and what it costs her on a shift. You get the cause behind the score, which is what you need to fix it.

What happens after the results come in?

Glint’s Copilot summarises scores and comments for a manager, and Focus Areas come with suggested actions and resources. Spradley’s Personal Agent goes on from the evidence itself. Broad questions fan out to research agents working in parallel, and their findings merge into one answer. When a company-wide report lands, a readout is written for leaders straight away.

The agent drafts what comes next: a message to the team or the company, a team conversation guide, a deck, a PDF or a Word file, or a Google Doc. Leaders can add their own files (PDF, CSV, text) or Google Drive documents as context, and it all works on a phone.

Does it launch the follow-up for you?

Glint is built for a rhythm: a regular engagement survey, lifecycle surveys at onboarding and exit, and recurring pulses. If an admin enables it, a manager can send a Viva Pulse follow-up built from the Glint results. Spradley runs a round of interviews whenever a question comes up, such as a drop in a Glint score, a reorganisation or a slow Copilot rollout. Themes are followed across rounds, and when the evidence on a question is thin, the Personal Agent drafts a follow-up interview on one to three themes. The leader sends it after one confirmation, and the new report joins the others.

Does it check that the action happened?

In Glint a manager sets a Focus Area status (on track, off track or complete) and a due date by hand. In Spradley a leader saves what they committed to, and reminders come at four and eight weeks until it is done. A follow-up interview then shows whether the change reached people, not just whether the status says complete.

Can anyone be identified, even by the AI?

Glint’s confidentiality settings default to five responses for scores, adjustable down to three, and ten for comments. Thresholds can’t be changed once a programme has collected responses, so decide early.

In Spradley privacy is built into the agent itself. Numbers and ratings are never shown for fewer than five people, and a privacy judge checks each written finding. The Personal Agent never sees one person’s answer, sentences naming an employee are cut, and it refuses to identify anyone, pick a team to cut or monitor intent to leave. Open answers can be easier to trace to a person than a rating, so keep interview groups larger where you can.

Can leaders use it from Microsoft 365?

Glint lives inside Microsoft 365. Spradley integrates with HiBob, Slack, Notion and Google Drive and Docs, and its MCP server brings each leader’s Personal Agent into Microsoft 365 Copilot, Claude and ChatGPT, never below five people. A leader can ask Copilot what people said about workload and get an answer from the interviews, with the evidence cited.

Does it find what stops people using Copilot?

This is where Glint comes up most in AI adoption work. Glint offers a Copilot Impact Survey template with four core items on productivity, speed, mental effort and quality of work. They must stay word for word. If you also have Viva Insights, the results of those four items appear as sentiment on the Copilot Dashboard, next to usage data. The template also has optional items and open questions on barriers and concerns.

That shows how many people use Copilot and how they rate it. It is harder to see what stops the people who don’t. A low score on “Using Copilot allows me to complete tasks faster” in a claims team could mean the prompts don’t fit their work, the data they need isn’t in Microsoft 365, or a manager told them not to use it on customer files. An interview asks which, and then asks why. A Personal Agent can then draft the guidance for that team and a follow-up interview to check it worked. Our post on measuring Copilot adoption beyond licence data covers the full picture, and our AI adoption page shows how Spradley fits into a rollout.

When should you choose Spradley?

• A Glint score dropped and the comments don’t explain it.
• Copilot usage is flat in some teams and you need to know what stands in the way.
• The group is too small for scores to be reliable, such as one site or one shift.
• You want reasons and quotes you can put in front of leaders within weeks.
• You want leaders to ask their own questions of the evidence, from Copilot if they like, and get a draft of the next step.
• You want the follow-up launched for you and commitments checked at four and eight weeks.

If all you need is a score once a year, a survey platform does that. Our comparison of employee listening platforms covers them, and Spradley vs Qualtrics EmployeeXM covers another enterprise option.

Can you use Viva Glint and Spradley together?

Yes. Run Glint for the engagement survey and the Copilot Impact items. When a result stands out, such as low Copilot scores in finance or a fall in manager scores at one depot, interview those teams with Spradley on that topic. The survey tells you where to look, the interviews tell you what to fix, and each leader’s Personal Agent helps plan it and checks it happened.

With the MCP server switched on, leaders who live in Microsoft 365 can ask about the interviews from Copilot without opening another tool.

Frequently asked questions

Key terms

Copilot Dashboard

A Viva Insights dashboard that shows Microsoft 365 Copilot readiness, adoption and impact, with survey sentiment where it is available.

Copilot Impact Survey

A Viva Glint survey template on how employees experience Copilot. Its four core items must be used word for word.

Follow-up question

A question the interviewer asks based on what someone just said, to find out what they mean and why.

Minimum group size

The smallest number of people whose answers can be shown as a group. Below it, results are hidden to protect individuals.

MCP server

A connector using the Model Context Protocol, which lets AI assistants such as Microsoft 365 Copilot read data from another product.

Personal Agent

Spradley's AI agent for each leader. It answers with cited evidence and drafts the next step, never seeing one person's answer.

Follow-up interview

A short interview on one to three themes, drafted when the evidence is missing and sent after the leader confirms. Its report joins the others.