What is strategic sourcing for corporate travel?
Strategic sourcing for corporate travel is using your own booking and spend data to decide which airlines, hotels, and ground suppliers to put under contract next, before an RFP is written, rather than renewing existing deals out of habit. It answers which vendors belong on the list in the first place.
Before the RFP: Using Cogent Agentic AI to Decide Which Vendors to Contract Next
The RFP is the wrong place to start. Most sourcing conversations open with "let's renew these contracts," when the smarter question is "which vendors should be on the list in the first place?" That question is almost never answered with data, because the data lives across too many systems.
Strategic sourcing for corporate travel, done properly, answers it before the RFP opens. Nearly two-thirds of business travel spend sits outside any managed deal, according to Euromonitor International (2025). This guide shows how Cogent by PredictX, the agentic AI platform for travel and expense, identifies high-usage non-preferred suppliers and contract-performance gaps in seconds, so you walk into the RFP already knowing which vendors matter.
In This Article
- What two questions does strategic sourcing have to answer?
- Why is this analysis usually skipped?
- What are enterprise teams querying with Cogent?
- How does the new sourcing workflow turn data into a decision?
- How does the negotiation dynamic shift before the meeting starts?
- How does Cogent get you there?
- Frequently Asked Questions
What two questions does strategic sourcing have to answer?
Strategic sourcing has to answer two questions: which non-preferred suppliers are we already spending heavily with, and are we hitting our market-share targets with the preferred suppliers we already contract? The first finds contract candidates. The second measures contract performance. Most programmes answer neither with data.
The first question is about opportunity. If a non-preferred hotel chain is quietly absorbing a large share of your room nights in a key city, it is a contract candidate, and you are paying rack rates for volume you could put under a deal.
The second is about leverage. If you committed to a carrier in return for a market-share target and you are missing it, the deal is underperforming and the renewal should reflect it. Tracking that, contract by contract and route by route, is what Air Contract Management, powered by Cogent agentic AI, does: it matches booked travel against contracted terms and surfaces your Net Effective Discount and the routes with no term applicable.
Both questions draw on data most programmes already hold but rarely connect: route-level air volume, hotel night concentration, preferred-vendor lists, and contract terms. PredictX sets out how that group-wide view is built in its guide to entity-level travel spend analytics.
Why is this analysis usually skipped?
This analysis is usually skipped because it requires joining spend data, preferred-vendor lists, and contract terms that sit in different systems, which is at least half a day of analyst prep per supplier category. By the time the analysis is done, the RFP cycle has already moved on, so teams default to renewing the incumbents.
The cause is the same data fragmentation that hides off-channel leakage. To answer "which non-preferred hotels in our top cities deserve a contract?", an analyst has to pull spend by property, cross-reference the preferred list, layer in room-night volume, and repeat it for every market. PredictX calls the compounding cost of that context-switching the Toggle Tax.
Faced with it, most teams renew the incumbents and benchmark against a consultant average, because building the real picture takes longer than the RFP timeline allows.
Common ways the RFP starts on the wrong foot:
- Renewing incumbent carriers without testing market-share performance
- Missing non-preferred suppliers that already carry significant spend
- Benchmarking against industry averages rather than your own data
- Opening the RFP before contract candidates are even identified
What are enterprise teams querying with Cogent?
Enterprise teams ask Cogent plain-language sourcing questions that each return in seconds, for example: "Show top non-preferred hotels in NYC by spend and room nights," or "What is our market share with United on top 10 transatlantic routes versus contract target?" Each answer comes back with the underlying data ready to drop into an RFP brief.
There is no analyst queue and no custom report request. As an agent, Cogent does three things at once:
- Cross-references spend against your preferred-vendor lists automatically
- Handles regional and entity-level breakdowns natively
- Surfaces contract-performance gaps proactively, including ones you did not think to ask about
A third common query, "Show Marriott global and regional market share for the trailing 12 months," returns supplier share you can take straight into a renewal.
The agent that runs this is the Air Sourcing Navigator, powered by Cogent agentic AI. Its continuous benchmarking and query process are set out in the complete guide and the continuous air sourcing solution.
How does the new sourcing workflow turn data into a decision?
The new workflow has three steps: identify high-usage non-preferred vendors in seconds, quantify the spend opportunity if they were brought under contract, and walk into the RFP knowing exactly which suppliers matter. It replaces a multi-week analyst exercise with a live capability.
The shift is from a once-a-year project to a standing view. A traditional sourcing exercise goes stale the moment route and supplier patterns change. Agentic AI monitors route share, supplier concentration, and contract performance continuously, so travel procurement analytics is a live picture rather than an annual snapshot.
Gartner projects that 40% of enterprise applications will feature task-specific AI agents by the end of 2026, up from less than 5% in 2025. With global business travel spending forecast to reach $1.69 trillion in 2026, according to Business Travel News Europe, the cost of targeting the wrong suppliers scales with the programme.
Consider an anonymised deployment pattern. At a global professional-services firm with more than 15,000 travellers, mapping route share showed that roughly 68% of transatlantic air volume ran on routes with no negotiated agreement, and one carrier already sat well above its fair market share on the two busiest routes. That produced a ranked target list before the RFP opened, pointing to a 6 to 9% reduction on the affected routes, based on enterprise deployment patterns; individual results vary.
How does the negotiation dynamic shift before the meeting starts?
The negotiation dynamic shifts because you arrive already knowing which suppliers carry your spend, where you over-concentrate, and where contract targets are being missed, so the leverage is established before anyone sits down. Sourcing readiness plus negotiation prep equals full procurement readiness.
This is not an argument against negotiation. It is an argument for sequencing: targeting sets the ceiling on savings, negotiation decides how close you get to it. The negotiation itself, the rate cards, clauses, and tactics, is covered in PredictX's analysis of vendor negotiation intelligence for corporate travel. This guide owns the step before it.
How does Cogent get you there?
Cogent gets you there by cross-referencing spend against your preferred-vendor lists automatically, handling entity-level and regional breakdowns natively, surfacing contract-performance gaps proactively, and answering in plain language with no custom report request. It turns a half-day-per-category analysis into a question you ask and answer in seconds.
This is the difference between a query tool and an agent. A dashboard waits for a report request, and a generative AI assistant answers the question you type and stops. Cogent is an intelligent workforce of AI agents: it acts on a plain-language question or an autonomous trigger, runs a multi-step analysis rather than a single lookup, scores and ranks candidate suppliers, and surfaces the contract candidate before anyone asks.
Put another way, reporting tells you what was spent, analytics explains where your share sits, and agentic investigation recommends which suppliers to target and produces the brief.
What Cogent brings to strategic sourcing:
- Acts as an autonomous agent, analysing on a plain-language question or an unprompted trigger
- Runs a multi-step analysis across air, hotel, and ground, then scores and ranks the targets
- Monitors route share and supplier concentration continuously, surfacing candidates proactively
- Cross-references spend against preferred-vendor lists and recommends the next move
No habit-based or average-based method produces a defensible target list, which is why procurement teams adopt agentic targeting. PredictX's travel and expense analytics platform brings corporate travel analytics across air, hotel, and ground together from over 200 sources, and connects to its wider work on sourcing and policy management.
PredictX CEO Keesup Choe frames sourcing as a data-access problem: the inputs that decide which suppliers to target almost always exist already, and the constraint is reaching them before the RFP is scoped. Cogent was named the 2025 BTS Europe Innovation Faceoff Winner, recognition of the shift from periodic, consultant-led sourcing toward continuous, evidence-led targeting.
Frequently Asked Questions
What is strategic sourcing for corporate travel?
Strategic sourcing for corporate travel is deciding which suppliers to contract next from your own route and spend data, before an RFP is written. It answers two questions: which non-preferred suppliers already carry heavy spend, and whether preferred suppliers are hitting their market-share targets. Cogent by PredictX runs this analysis continuously rather than once a year.
What is non-preferred vendor analysis?
Non-preferred vendor analysis identifies suppliers outside your preferred programme that already carry significant spend, marking them as contract candidates. A non-preferred hotel chain absorbing a large share of room nights in a key city is leverage you are not using. Cogent surfaces these automatically by cross-referencing spend against your preferred-vendor list.
How does Cogent help with travel procurement analytics?
Cogent maps spend to suppliers and routes, benchmarks share against contract targets, and scores candidate vendors continuously. Rather than answering a question only when asked, it keeps a live target list ready for the next RFP. In one anonymised deployment, this surfaced that 68% of transatlantic volume ran uncontracted, based on enterprise deployment patterns; individual results vary.
Which vendors should I contract first?
Contract first where the gap between your spend and your negotiated coverage is largest and the supplier already carries strong volume. That combination, high winnable spend and a wide coverage gap, marks the targets where a new deal moves the most money. Cogent ranks these automatically so procurement starts with evidence, not the previous cycle's roster.
How is strategic sourcing different from running an RFP?
Strategic sourcing decides who belongs in the RFP; the RFP itself collects and negotiates bids from that list. Sourcing is the analytical step that sets the ceiling on savings by choosing the right targets. Running an RFP on the wrong supplier list, however well negotiated, leaves the largest savings untouched.
Is agentic AI different from generative AI for sourcing?
Yes. Generative AI answers a sourcing question when you ask it, while agentic AI monitors route share and supplier concentration continuously and surfaces the targets before you ask. For sourcing, that means the target list is current the moment the RFP window opens. This is the basis of agentic AI for travel and expense management.
Key takeaway The RFP is not where your savings are decided. They are decided the week you choose which suppliers to target, on which routes, and that choice should come from your own data, not last cycle's contracts. The question is not how hard you can negotiate. It is whether you are negotiating with the right suppliers in the first place.
See Cogent run a sourcing readiness analysis on your data
Ask one question your current process cannot answer quickly: which non-preferred hotels in your top cities carry the most spend, and what would it be worth to put them under contract? If you cannot answer it before the RFP, your target list is built on habit.
About Cogent by PredictX
Cogent is the agentic AI solution from PredictX, built for travel, finance, and procurement teams as the way they work changes fast. It deploys in seconds, not months: you ask in plain language and the agent returns the answer, with no reporting build and no analyst queue.
- Named the 2025 BTS Europe Innovation Faceoff Winner
- Trusted at scale: 4 of the 6 largest travel programmes rely on PredictX
- Agentic by design: Cogent works as a virtual full-time equivalent, not a query box