Your organization may already have plenty of information about the people and organizations it works with. Applications explain where someone started. Surveys capture what they say. Staff notes describe what happened. Documents provide supporting evidence.

The difficulty comes when you need to use those sources together. A new question sends someone back through the files: find the right person, match the dates, check what the question meant, read the comments, and assemble an answer. When the next round of updates arrives, much of that work starts again.

Sopact Sense is a workspace for teams that collect information regularly and need to understand it together. It connects forms, files, feedback, and relevant history; supports AI analysis as information arrives; and gives authorized people control over collection, access, and review.

This course explains that approach through one example. You will learn what to connect, what AI needs to know, and what your team should control before relying on an answer.

What “Connected Data Intelligence” means

Connected Data Intelligence brings recurring collection, relevant history, AI analysis, and governance into one continuing evidence workflow.

The words describe two connected responsibilities:

• Connected data: keep information associated with the person, organization, program, or site it describes, together with its source, date, and meaning.
• Connected intelligence: use that information and context to understand changes, identify questions worth investigating, and support decisions.

Governance applies to both. It covers what your team collects, how measures are defined, who can access the information, what AI may use, and who checks the findings.

Sopact uses this category name to describe the complete workflow. Its value should be visible in everyday work: fewer repeated handoffs, evidence that remains usable, and a team that can maintain the process. The name alone does not establish a capability.

How Sopact works: five capabilities, one connected workflow

Sopact’s workflow has three parts working together: shared definitions give information meaning; collection, history, and analysis turn incoming information into usable evidence; and governance determines who can manage and use that evidence. The diagram shows how the five capabilities fit together.

Shared definitions guide the work

Set shared definitions. Before interpreting a response, the workflow needs to know what the question measures, which period it covers, and what counts as a result. In the diagram, “positive” means a score of 4 or 5 on the same five-point scale. A different scale or definition needs review before results are combined.

This is part of context: the information needed to interpret a record correctly. Context also includes who or what the record describes, its relevant history, the rules for using it, and the decision it should support. Chapter three explains how to organize that context.

Collection adds to history, and analysis uses both

Collect flexibly. Forms, files, and feedback bring in different kinds of evidence. A registration form, a partner’s document, and a follow-up response can contribute to the same workflow.

Connect over time. Relevant updates stay associated with the person, organization, program, or site they describe and the period they belong to. Earlier evidence remains available when a new response arrives. Your team must define which records belong together; anonymous feedback should retain useful program or cohort context without being connected back to a person.

Analyze on arrival. Configured AI analysis uses incoming information and the relevant context made available to it. Staff can review findings alongside supporting evidence while there is still time to act. In the diagram, the increase in positive responses starts an investigation: what changed, who responded, and what does the written feedback explain?

Your organization governs the whole workflow

Own and govern. The bottom band applies to collection, history, and analysis. Organizational policies set the boundaries; authorized data owners manage routine questions, collection cycles, and review within them. Different people can have different permissions. A written instruction about confidentiality does not replace an access control.

This is what self-managed means here: your team can maintain the continuing workflow within agreed controls. Initial setup still matters—someone must define the records, context, permissions, and analysis instructions, then check that they work.

The architectural difference becomes clear at the next collection cycle. A separate survey export and AI chat leave your team to rebuild the context and connections. Sopact brings that recurring work into one workspace. Other platforms can support connected data and AI too; chapter two explains what to compare for your team’s workflow.

Start with a question your team recognizes

Imagine a small team running a training program. It needs to know: “Who is using what they learned, and who needs more support?”

The team has registration details, attendance, mentor notes, and a follow-up survey. Each source is useful, but none answers the whole question.

One learner, Maria, says she is not using the new skill. That answer needs context. Did she finish the program? Did she demonstrate the skill during training? Has her employer given her an opportunity to use it?

The following example is fictional.

The team can now investigate a specific barrier. It has more than a low score or an isolated comment.

The same need appears elsewhere: a funder following a grantee’s commitments, a membership team connecting activities with renewal feedback, or a partner manager checking a site report against an agreement.

Why the work outgrows a survey

A survey is a good way to collect a set of answers. It may be all you need for feedback on a single event.

The requirement changes when your question crosses forms, documents, teams, or periods. You need to know which updates belong together, preserve earlier information, apply consistent definitions, and decide who may see the result. You also need to notice who has not responded.

If the tools leave those jobs to separate exports and spreadsheets, your team becomes responsible for maintaining all the connections by hand. An AI summary of the latest survey can help read that survey, but it does not by itself supply the missing history.

This is the point at which a connected workflow becomes useful: the next collection should add to information you can already use.

What self-managed means for your team

Self-managed means the people responsible for the work can maintain routine forms, organize follow-up, apply access controls, and review findings within agreed boundaries.

Your organization still needs to make decisions. Someone must own the workflow, define the measures, approve changes, and check ambiguous evidence. Setup, security, and integrations may require specialist support.

The practical test is what happens after setup. Can your team add the next collection period, understand a change to a question, and keep the process running when a colleague leaves?

That operating responsibility is central to Sopact’s approach. It matters especially when data collection is growing faster than the team available to manage it.

Who this is for

This course is for organizations collecting information repeatedly, often from people or organizations outside their own team.

You do not need to connect every source at once. Some teams should begin by bringing reports together under shared definitions. Others need a continuing participant or partner record. Start with the work that has an immediate use.

Where your existing tools fit

A CRM, survey platform, data warehouse, or AI assistant may still have an important role. Before changing tools, identify the part of the evidence workflow that your team currently has to rebuild.

The next lesson examines those approaches, including enterprise experience platforms such as Qualtrics and Medallia. It shows what happens when a new response arrives and what to ask a provider to demonstrate.

For now, keep one question in mind: “Can our team collect the next update, understand it with the earlier evidence, and check the answer without rebuilding the process?”

Take one example into the next lesson

Choose a recurring question from your own work. Note who needs the answer, which sources contain the evidence, and where someone currently has to connect or interpret information manually.

You can record those notes in the working evidence plan. A simple example is enough to begin.

Next: AI-native vs AI-bolted: what changes in practice? Follow Maria’s next response from collection to a reviewed finding.