Skip Navigation Get In Touch
Case study hero — 80% Efficiency Gain in Benefit Extraction
Document Intelligence · Global Mobility · AI Automation

80% Efficiency Gain in Benefit Extraction
Powered by AI.

AIRINC helps companies navigate employee relocation across cities and countries. At the centre of their service is benefit analysis: reading a company's relocation policy, identifying every benefit and sub-benefit it contains, and benchmarking that package against industry standards.

80%Efficiency gain
48+Pages per document
v1Of what's possible
Industry Global Mobility
Solution AI Document Extraction
Efficiency Gain 80%
Output Structured Excel
Approach Human-in-the-Loop
Model Large Language Model
Traceability Source-Linked
Where They Started

Skilled work.
Done manually.

AIRINC’s core service depends on benefit analysis—reading policy documents, identifying every benefit and sub-benefit, and benchmarking against industry standards. The process itself was well-defined. Analysts followed a structured approach, applying domain expertise to interpret complex policy language and extract the relevant details.

The problem was the reading.

Each document often 40+ pages, had to be read end to end. Benefits were identified and catalogued manually, with accuracy dependent on the analyst’s attention and experience.

There was no automated way to track coverage, verify completeness, or standardise output across engagements. The work was precise, but it was time-intensive, and difficult to scale. AIRINC didn’t need to change the process. They needed a system to execute it.

Every engagement. Every time.
48
Pages. Read manually.
Every single engagement.
What Was Breaking

Four problems.
One breaking point.

01
Scale Without Leverage
Every new client engagement meant another full day of manual reading. There was no way to take on more volume without adding more people, and the work didn't scale with headcount the way it needed to.
02
No Audit Trail
Once a document was processed, there was no easy way to verify what had been found, where in the document it appeared, or whether anything had been missed. The output was only as reliable as the analyst's attention.
03
Inconsistent Output
Without a standardised extraction format, the quality and structure of the output varied. Comparing benefit packages across clients or benchmarking against industry standards required additional normalisation work downstream.
04
No Path to Automation
The team knew the process could be improved, but had no internal AI capability to act on that knowledge. The bottleneck wasn't awareness, it was access to the right technical expertise to actually build the solution.
How The Zig Fixed It

A system for extraction,
built for verification.

The Zig built an end-to-end AI extraction system designed around AIRINC's existing workflow.

Documents are uploaded and standardised before processing. Non-PDF formats are automatically converted before processing begins. A large language model then reads the entire document and identifies every benefit and sub-benefit it contains, categorising each one with precision, beyond simple keyword matching.

Critically, the system doesn't just extract. It traces. Each identified benefit is pinpointed to the exact location in the original document, with the relevant text highlighted so an analyst can verify the AI's work at a glance. Nothing is accepted blindly. Every extracted item passes through a human review layer before export.

Engineering Note

The initial architecture used Azure Document Intelligence with vector embeddings and AI Search. After testing, The Zig pivoted to a full LLM approach, delivering meaningfully better accuracy on complex policy language. A custom recall function re-runs extraction if more than 10% of expected items are absent.

System, Step by Step
i.
Upload + Auto-Convert
User selects company profile, uploads policy. Non-PDF formats converted automatically.
ii.
LLM Reads the Document
Full document processed. Every benefit identified, beyond keyword matching.
iii.
Source Tracing
Each item pinpointed to its exact page. Relevant text highlighted for verification.
iv.
Confidence Calibration
Custom normalisation formula calibrates LLM confidence scores reliably.
v.
Recall Safeguard
If >10% of expected items absent, extraction re-runs automatically.
vi.
Human Review + Export
Every item reviewed before sign-off. Clean Excel output ready for benchmarking.
The Result
0%
Efficiency gain in benefit extraction
AIRINC can now process policy documents in a fraction of the time it previously required. The analyst's role has shifted: less reading, more verifying and adding judgment to output that arrives substantially complete.
What It Unlocked

Not just faster.
Verifiable and
Consistent.

The system reduced processing time significantly, but the more important change was traceability.

Every extracted benefit now links directly to its source text, allowing analysts to verify results immediately without re-reading entire documents. Output is standardised across engagements, removing the need for downstream normalisation in the previous workflow and enabling consistent benchmarking.

Beyond the metrics, this project introduced AIRINC to AI as a practical capability.

Policy documents processed in a fraction of the previous time
Speed
Every benefit linked to exact source text, instant evidence for client queries
Traceability
Standardised Excel output ready for benchmarking, normalisation eliminated
Consistency
Analyst role shifts to verification and judgment, higher-value work
Quality
AI proven as a practical capability, generating conversations about what's next
Platform
What This Taught Us
The most valuable AI implementations are the ones that respect existing expertise rather than trying to replace it.

AIRINC's analysts understand nuance in policy language that no system will catch perfectly. The right design puts AI in front of that expertise, not in place of it. Human-in-the-loop isn't a concession to imperfect technology. It's the architecture that makes automation trustworthy enough to actually use.

See what's possible.
Build what matters.

Tell us where manual work is holding your team back. We'll be direct about what AI can actually do for it (no slides, no theatre).