Introducing Smart Insights 2.0

Introduction

Modern teams are expected to move faster than ever. Yet despite having access to dozens of productivity tools, many organizations still spend a significant portion of their week on repetitive tasks, manual updates, and administrative work.

From chasing approvals to updating spreadsheets and sending status reports, these seemingly small activities quickly add up.

In many organizations, implementing intelligent automation can save more than 10 hours per employee every week while improving accuracy, visibility, and overall productivity. The good news? AI workflow automation is changing the way teams operate.

The Hidden Cost

Most teams underestimate how much time is spent on repetitive processes.

Consider a typical week:

• Updating project statuses

• Sending reminder emails

• Tracking approvals

• Creating recurring reports

• Syncing information across tools

• Managing task hands off

While each task may only take a few minutes, the cumulative impact is substantial. Across an entire team, these activities can consume dozens of hours that could otherwise be spent on strategic work. AI workflow automation combines traditional automation with artificial intelligence to streamline business processes. This allows teams to automate not only repetitive actions but also decision-making processes that previously required manual intervention.

Automated Status Updated

One of the most common productivity drains is keeping stakeholders informed.

AI workflows can automatically:

• Update project boards

• Notify teams of progress

• Generate summaries

• Share milestones

This eliminates the need for repetitive status meetings and manual reporting. Many workflows become delayed because approvals depend on email chains and manual follow-ups.

AI automation can:

• Route requests automatically

• Notify approvers instantly

• Escalate overdue approvals

• Track completion status

Teams often spend hours documenting discussions and action items. This ensures everyone stays aligned without additional administrative effort.

AI Workflow Automation

Today we’re launching our most powerful update yet for structured extraction: Deep Extract.

Deep Extract is a new agent harness approach to extraction that verifies and corrects its own output until the results are accurate. Much like human-in-the-loop, Deep extract has an agent-in-the-loop, offloading the human reviewer’s burden with an autonomous verification cycle that holds itself accountable for accuracy.

This is particularly powerful when you're dealing with a long list of items to extract — think invoice line items, brokerage statement transactions, equipment manifests, and more. Deep Extract has already extracted over 28 million fields on documents up to 2,500 pages long in our production beta, and we're continuing to expand what's possible.

For the documents that matter most, it gets to 99–100% field accuracy, even out-performing expert human labelers on extraction tasks.

The Challenge with Extraction

Over the past year, we kept hearing the same thing from customers. Their existing extraction pipelines were breaking down on long, complex documents — invoices running dozens of pages, financial statements spanning hundreds. However, totals didn't reconcile, and it flagged to teams that line items were dropped completely.When we asked how they were handling it, the answer was almost always the same: they'd hired people to have a human-in-the-loop (HITL) manually check the output.

The issue isn't that models are bad at reading documents. It's that single-pass extraction has no mechanism to catch its own mistakes, and models get lazy. Models are prone to shortcuts on long, repetitive tasks. Given a thousand line items to extract, they'll often stop short, consolidate, or skip entries rather than working through every last row.

This is amplified even more when citations are needed. For many of our customers, citations are not just a nice to have, but a need in order to prove their outputs.