Travel exceptions stalled decisions — Concur & CRM unified in Snowflake, CFO?gated

Overview

Executive travel and spend policy at a global enterprise drifted because leaders lacked a clear, shared view of how trips connected to sales outcomes and project delivery. Expense data in Concur, client activity in the CRM, and project results lived in separate tools with different calendars and owners. Intelligex linked SAP Concur, CRM, and project outcomes into a cost–benefit view in Snowflake, encoded exception rules in a simple engine, and introduced a routed approval for CFO sign?off. Leadership made more targeted travel decisions with fewer disputes and clearer justification because every exception and recommendation referenced the same governed evidence.

Client Profile

• Industry: Enterprise technology and services
• Company size (range): Global footprint with field sales, delivery, and corporate functions
• Stage: Mature operator standardizing travel policy and outcome measurement
• Department owner: Strategy, Analytics & Executive Leadership (Corporate Strategy / Finance)
• Other stakeholders: Sales Leadership, Customer Success, Professional Services/Delivery, Finance/FP&A, Procurement/Travel, HR/Policy, Legal & Privacy, IT/Data Platforms

The Challenge

Travel decisions were made with partial context. Concur showed trip costs by traveler and category, the CRM showed meetings, stages, and next steps, and project tools recorded delivery milestones and customer sentiment. None of these systems shared a common identity model or calendar, and “purpose of trip” fields were filled inconsistently. Executives approved exceptions without a consolidated view of impact, which fueled debates about which trips were essential and whether virtual options would have performed just as well.

Policy language became hard to enforce. Regional teams interpreted thresholds differently, after?the?fact approvals were common, and Finance struggled to explain variances because expense reports and outcomes could not be reconciled quickly. Leadership wanted to keep Concur, the existing CRM, and delivery tools while adding a governed way to link trips to business outcomes, surface exceptions with reason codes, and require a lightweight approval before spend went out of bounds.

Why It Was Happening

Identities and calendars were fragmented. Concur keyed on traveler and cost center, the CRM keyed on accounts, opportunities, and activities, and delivery tools keyed on projects or epics. Time zones and cut?offs differed across systems, and unstructured trip descriptions varied by individual. Analysts stitched data together in spreadsheets that drifted over time, leading to defensible but incompatible answers.

Governance arrived late. Exceptions were justified in emails, and CFO approvals happened after tickets were booked. There was no standard attribution rule for linking a trip to an opportunity stage change or a project milestone, and no audit trail binding a policy decision to the exact inputs used. As a result, reviews focused on adjudicating one?offs rather than shaping policy and behavior.

The Solution

We implemented a governed cost–benefit model that connected travel spend to client outcomes and routed exceptions through a simple rules engine with CFO approval. Concur expense headers and line items landed in a conformed model alongside CRM activities and stage changes and project delivery milestones. dbt encoded identity stitching, calendars, and attribution rules that linked trips to opportunities and projects based on timing, participants, and tags. A Power BI view surfaced policy compliance, exception candidates, and outcome context with drill?through to evidence. A lightweight approval workflow in ServiceNow captured CFO sign?off with reason codes before out?of?policy spend. Existing systems remained in place; the orchestration unified data, rules, and governance around them.

• Expense and itinerary data from SAP Concur APIs with traveler, cost center, and category detail
• CRM meetings, activities, and opportunity context via Salesforce REST API or the client’s existing CRM
• Project delivery milestones and customer health from the delivery tool (for example, Jira epics or services milestones)
• Conformed model and snapshots in Snowflake with shared calendars and mastered identities for traveler, account, opportunity, and project
• Transformations and validations in dbt for identity stitching, attribution rules, accepted values, and stale feed checks
• Exception rules engine that flags out?of?policy trips by category, timing, and purpose, and evaluates outcome relevance
• Executive and manager dashboards in Power BI with tiles for policy compliance, exceptions, and outcome drill?through
• Approval workflow in ServiceNow capturing CFO and delegated approvals with reason codes and attachments
• Role?based access with data minimization so Finance sees full detail, while managers see scoped views (identity groups such as Okta Groups)
• Audit log linking each decision to dataset versions, attribution results, rules, and approvers

Implementation

• Discovery: Cataloged Concur fields, CRM objects, and delivery milestones; inventoried current travel policy and exception patterns; mapped traveler, account, and project identifiers; reviewed representative disputes to understand attribution gaps; and aligned on CFO approval criteria and delegation.
• Design: Defined the shared identity model and calendar, attribution logic for linking trips to opportunities and projects, and a simple rules engine for exceptions. Authored required fields for exception requests and reason codes. Designed Power BI views for Finance and business leaders, and the approval flow with comments and attachments.
• Build: Landed Concur, CRM, and delivery feeds into Snowflake; implemented dbt models for identity stitching, attribution, and validations; built the rules engine to flag exceptions and compute outcome context; developed Power BI dashboards with drill?through to expense lines and activities; and configured ServiceNow approvals with change logging.
• Testing and QA: Replayed recent quarters to reconcile travel costs and outcomes; tuned attribution rules with Sales Ops and Delivery; validated exception flags against policy language; verified role?based access and data masking; and dry?ran approvals with CFO delegates using real scenarios.
• Rollout: Launched dashboards in read?only mode alongside the legacy process. After validation, enabled the approval workflow for exceptions and made the governed view the source for executive travel reviews. Maintained a manual override path for urgent travel with post?review documentation.
• Training and hand-off: Delivered quick guides for managers on reading exception tiles and submitting rationale, for Finance on verifying attribution and approvals, and for Sales Ops/Delivery on stewardship of tags and milestones. Established ownership for rules, definitions, and mappings with a change?control cadence and human?in?the?loop review.

Results

Executives and managers reviewed travel through a single lens that connected spend to client meetings, opportunity movement, and project delivery. Exceptions were flagged with clear rules and linked evidence, and CFO approvals captured rationale before spend occurred. Debates about whether a trip was necessary shifted to a discussion of trade?offs grounded in consistent attribution and policy language.

Operational friction declined. Finance reconciled faster because identities and calendars were harmonized and validations caught stale feeds and missing tags. Sales and Delivery teams saw which trips correlated with key outcomes and could plan travel against a shared standard. With approvals, rules, and lineage visible, policy drift slowed and disputes tapered.

What Changed for the Team

• Before: Expense, CRM, and project data were stitched manually. After: A conformed model linked trips to opportunities and milestones with drill?through to evidence.
• Before: Exceptions were justified in email after booking. After: A rules?based flag and approval captured CFO rationale before out?of?policy spend.
• Before: “Purpose of trip” fields were inconsistent. After: Attribution rules and required tags aligned travel to outcomes under a shared calendar.
• Before: Reviews focused on anecdotes. After: Dashboards presented governed tiles for compliance, exceptions, and outcomes.
• Before: Accountability for policy changes was unclear. After: An audit log tied decisions to datasets, rules, and approvers.

Key Takeaways

• Unify travel, CRM, and delivery signals under a shared identity and calendar; cost–benefit decisions need comparable inputs.
• Encode attribution and exception logic in transformations and a simple rules engine; definitions in code reduce drift and disputes.
• Require CFO approvals with reason codes before exceptions proceed; governance in the flow beats after?the?fact adjudication.
• Deliver executive views with drill?through to evidence; transparency builds trust in policy and enforcement.
• Keep Concur, CRM, and delivery tools; add a governing layer for integration, approvals, and auditability rather than replatforming.

FAQ

What tools did this integrate with?
We ingested travel and expense data from SAP Concur APIs, meetings and opportunity context from the CRM (for example, Salesforce REST API), and project milestones from the delivery platform. Data landed in Snowflake, transformations and rules ran in dbt, decision views were delivered in Power BI, and approvals executed in ServiceNow with access governed by identity groups.

How did you handle quality control and governance?
A mastered identity model aligned travelers, accounts, opportunities, and projects under a shared calendar. dbt encoded attribution rules and validations for accepted values, stale feeds, and missing tags. A rules engine flagged exceptions with clear criteria, and CFO approvals captured reason codes and attachments before spend occurred. An audit log recorded dataset versions, rules, and approvers.

How did you roll this out without disruption?
We ran dashboards in read?only mode while teams compared results to legacy reports. After validating attribution and exception behavior, we enabled the approval flow for exceptions and kept a manual override for urgent travel with post?review documentation. Core systems remained the same; the new layer orchestrated integration, rules, and approvals on top.

How did you attribute business impact to travel?
Attribution relied on timing, participants, and tags. Trips were linked to opportunities when traveler and account matched and meetings bracketed a stage movement within a defined window. For delivery, trips linked to projects through traveler, account, and milestone proximity. Rules favored conservative associations, and ambiguous cases routed to a human reviewer with notes stored for future reference.

How did you address privacy and sensitive traveler data?
Role?based access and data minimization limited person?level details to Finance and designated managers. Dashboards showed aggregated views by default, with drill?through gated by permissions. Sensitive fields from Concur were masked in analytics tables, and approvals and exports were logged for audit and policy compliance.