SAP AI Core & Launchpad: Negotiation Approach

Why SAP AI Licensing Negotiations Differ From Traditional SAP Licensing

If you've negotiated SAP ERP licenses before, stop assuming the same playbook applies to SAP AI Core and Launchpad. SAP's AI pricing model operates under entirely different economic pressures and contract mechanics.

Traditional SAP licensing (ERP, analytics, CRM) follows a predictable model: you agree on named users or ABAP Workbench pricing, lock in maintenance costs, and operate in relatively stable territory for 3-4 years. SAP's incentive is to retain predictable ARR (Annual Recurring Revenue) and maintain customer relationships across the portfolio.

SAP AI Core and Launchpad flip this on its head. These are consumption-based services tied to SAP's quarterly revenue targets and competitive pressure from hyperscalers. SAP launched AI offerings aggressively to compete with AWS SageMaker, Azure AI Services, and Google Cloud Vertex AI. This creates three critical negotiation dynamics:

• Consumption unpredictability: You don't know your actual costs until usage happens. SAP reserves the right to charge based on actual API calls, inference compute, model training hours, or data storage—often with vague definitions in the contract.
• Aggressive upsells: SAP bundles AI with RISE, S/4HANA, and Datasphere deals, embedding AI costs into larger agreements where they're harder to scrutinize.
• Hyperscaler competition: SAP knows enterprises are evaluating Azure OpenAI, Bedrock, and SageMaker alongside its AI portfolio. This dynamic creates negotiating room if you credibly demonstrate alternatives.

The Leverage Points: SAP's Quarterly Revenue Targets and Market Reality

SAP publishes quarterly guidance on AI revenue growth. In recent cycles, SAP Cloud Business unit targets have included significant AI adoption metrics. Your negotiation leverage sits at the intersection of SAP's desperation to close deals before quarter-end and your credible willingness to build AI capabilities on competing platforms.

SAP's public statements acknowledge the competitive threat. In earnings calls and investor presentations, SAP management notes that hyperscaler AI services are gaining adoption faster than anticipated. This admission signals that SAP is willing to negotiate more aggressively on AI contracts than on legacy ERP.

Leverage Point #1: Fiscal Year and Quarter-End Timing

SAP's fiscal year ends December 31. The company's fiscal quarters close on March 31, June 30, September 30, and December 31. In the 30-60 days before these dates, SAP sales teams face maximum pressure to close deals. November-December and late February-March are peak negotiation windows.

During these windows:

• SAP's sales teams have higher discounting authority (sometimes 40-50% off list pricing)
• Service credits and favorable term structures become available without escalation to SAP legal
• Multi-quarter commitments offer better per-unit pricing than month-to-month usage

The worst time to negotiate is mid-quarter (February, May, August, November), when SAP's pipeline is full and sales teams have less urgency.

Leverage Point #2: Hyperscaler Competitive Threat

This is your most powerful card. SAP knows:

• AWS SageMaker offers lower per-hour inference costs and better integration with existing AWS infrastructure
• Azure OpenAI Services provide GPT-4 access with enterprise governance that rivals SAP's claims
• Google Cloud Vertex AI includes foundation models, data governance, and MLOps at lower TCO than SAP's bundled approach

In negotiations, reference your technical evaluation of these platforms. Say: "We've assessed Azure OpenAI for our AI roadmap. SAP Core can integrate, but pricing needs to reflect that we're not locked into your ecosystem." This statement alone typically triggers SAP legal to become more flexible on contract terms.

Leverage Point #3: Your Company's Growth Stage and Adoption Trajectory

If you're a large, established customer with 5+ SAP modules and 10,000+ named users, SAP fears you're a churn risk if you begin evaluating Azure AI or Bedrock. Your negotiating position is stronger because you represent potential expansion (more AI modules, more data consumption) if the contract is favorable.

Conversely, if you're a mid-market customer with 2-3 SAP modules, frame the negotiation as a "proof of concept" or "pilot platform." SAP will often offer discounted pricing on pilots to establish a beachhead for future expansion.

Timing Your AI Negotiation: Fiscal Windows, Contract Renewal, and the Pilot Trap

The Fiscal Year-End Window (November-January)

This is the peak negotiation period. SAP's sales organization faces year-end quota pressure. Deals signed in December and January close SAP's fiscal year revenue targets. You have maximum leverage.

Strategy: If you're initiating an SAP AI Core evaluation or renewal, schedule your RFP release for October-November. This gives SAP 6-8 weeks to prepare a proposal before December closes, and creates deadline pressure on their side.

Contract Renewal Windows

Most SAP AI pilots launched in 2024 are now entering 3-6 month evaluation periods. If you agreed to a pilot in early 2024, your production contract decision point is late 2025 or early 2026.

At this juncture, do not let SAP convert your pilot terms (often heavily discounted) to standard production pricing without renegotiation. This is the pilot-to-production trap.

Mid-Quarter vs. End-Quarter Negotiations

Avoid negotiating in February, May, August, or November unless you're responding to a contract renewal notice. In these months, SAP's pipeline is full, and sales teams have less discounting authority.

Optimal timing:

Critical Contract Terms to Negotiate for SAP AI Core & Launchpad

1. Consumption Caps and Overages

This is non-negotiable. SAP's default contract often includes vague language on "actual consumption" with open-ended overage charges. You must establish:

• Monthly/quarterly consumption cap: A fixed maximum (e.g., "$50,000 per month" or "1M inference calls/month"). Anything over that requires SAP approval and is charged at a defined rate (typically 10-20% overage premium).
• Overage pricing: Define the per-unit overage cost upfront. Do NOT accept "overage pricing to be determined" or "at SAP's standard rates." Specify: "Overages billed at 1.15x the negotiated per-unit rate."
• Overage notification: Require SAP to notify you when you're at 80% of your consumption cap with 14 days lead time. This prevents surprise bills.

2. Credit Rollover and Non-Expiring Credits

SAP often includes service credits or usage credits in AI contracts. Negotiate:

• Credit non-expiration: Credits should not expire at calendar year-end or contract anniversary. If you don't consume $200K in credits in Year 1, you carry them forward to Year 2.
• Credit transfer rights: You can apply credits to any SAP cloud service (not just AI Core, but also Datasphere, Analytics Cloud, S/4HANA Cloud).
• Credit true-up clause: At contract end, unused credits convert to refunds (not forfeited).

3. Burst Pricing and Auto-Scaling Controls

SAP AI Core and Launchpad pricing models include dynamic scaling. When your workload spikes (e.g., month-end financial close processes trigger heavy model inference), SAP charges premium rates for burst capacity.

Negotiate:

• Burst allowance: You can exceed your consumption cap by 20% in any month without overage charges, provided you don't exceed it in more than 2 months/year.
• Burst pricing cap: If you exceed burst allowance, overage charges are capped at 1.25x your negotiated rate (not 2-3x as SAP standard allows).
• Workload predictability statement: Provide SAP with your expected peak usage months (e.g., month-end closes, quarterly reviews) so they don't charge burst rates for predictable workloads.

4. Exit Clauses and Termination for Convenience

This is critical. SAP AI Core and Launchpad contracts often include 3-year terms with automatic renewal and steep exit penalties. Negotiate:

• Termination for convenience: You can exit the contract with 90 days notice after Year 1, 60 days notice after Year 2, and 30 days notice after Year 3. No early termination fees beyond the notice period.
• Performance exit clause: If SAP AI Core uptime drops below 99.5% in any quarter, or if SAP discontinues a core feature, you can terminate with 30 days notice.
• Pricing increase caps: SAP cannot increase pricing by more than 5% annually. If SAP proposes increases above this, you can exit without penalty.

5. Service Level Agreements (SLAs) for AI Services

SAP Cloud SLA guarantees 99.5% uptime for compute and storage. For AI workloads, also negotiate:

• Inference latency SLA: Model inference responses within 5 seconds (or specify your requirement). If SAP misses this in any month, service credits apply (e.g., 5% monthly fee credit).
• Model availability SLA: Pre-trained models (e.g., generative AI, predictive models) available 99.9% of the time.
• Support response time: For production incidents, 1-hour response time from SAP support (vs. SAP's standard 4-hour SLA for cloud services).

The "Pilot to Production" Trap: How Free Trials Become Expensive Commitments

This deserves its own deep dive because it's where 70% of enterprises get blindsided on AI costs.

In 2024, SAP heavily promoted "free AI Core pilots" (often 6-12 months at no cost). Enterprises spun up pilots, loaded data, built integrations, trained teams, and achieved measurable ROI in AI use cases. By month 9, the organization was dependent on the pilot and reluctant to terminate.

Then SAP proposed production contracts. What happened?

• The pilot's free tier ended, and SAP quoted production pricing at 10-15x the pilot's cost (because the pilot cost was $0).
• Enterprises had no exit ramp. Terminating the pilot meant reversing AI initiatives mid-flight, a politically costly move.
• SAP knew this and offered a "compromise" production contract at 50-60% off list pricing—still 5-8x higher than enterprises expected.

How to Avoid the Trap

Step 1: Establish a pilot-to-production economics agreement upfront. Before signing the pilot SLA, negotiate what production pricing will be. For example:

Step 2: Define pilot success metrics and graduation criteria. The pilot should include explicit criteria for "go/no-go" at 6-month mark. Example: "If actual ROI exceeds $500K annually and model accuracy exceeds 85%, the pilot graduates to production on [specified terms]. If either metric is missed, the pilot is terminated at no cost to the enterprise."

Step 3: Negotiate a pilot exit clause with 90-day wind-down.. You can terminate the pilot with 90-day notice for any reason (or no reason). During the 90-day wind-down, you can reduce consumption to zero without penalty. This prevents lock-in.

Step 4: Cap pilot resource investments. Establish a pilot budget (labor, infrastructure, data prep) of $200-400K max. If the pilot requires more investment than this, it's not a pilot—it's a production deployment, and it should be on production terms from the start.

Bundled vs. Standalone AI Negotiation Tactics

Scenario 1: SAP AI Core Bundled into RISE with SAP

If you're evaluating RISE (which includes S/4HANA Cloud, integrated analytics, and business process intelligence), SAP often bundles SAP AI Core at no incremental cost for the first 2 years.

Red flag: "No incremental cost" means SAP has embedded AI pricing into your RISE contract at an inflated base rate. After Year 2, AI costs suddenly appear as line items. By then, you're dependent and have no leverage.

Negotiation tactic: Require SAP to separately itemize AI costs even if bundled. Example: "RISE contract is $5M for 3 years. Itemize: S/4HANA ($3.2M), Analytics Cloud ($1.2M), AI Core ($600K). This clarity shows you the true cost of each component and prevents surprise AI billings post-Year 2."

Scenario 2: Standalone SAP AI Core Contract

If you're negotiating SAP AI Core independently (not bundled with RISE), you have more flexibility. SAP has less leverage to force unfavorable terms because you're not locked into a larger deal. Use this.

Negotiation tactic: Propose a shorter initial term (1 year instead of 3 years) with proven performance metrics. Tell SAP: "We'll commit to 12 months on AI Core at [price]. If the service meets our 99.5% uptime and 5-second inference SLA targets, we'll extend to 24 months at 5% discount. Tie our commitment to your performance, not our lock-in."

SAP often accepts this because it gives them a foot in the door, and they expect your AI consumption (and switching costs) to increase over Year 1, making you reluctant to churn.

Red Flag Contract Clauses: What to Push Back On

Hyperscaler Alternatives as Negotiation Leverage

Your strongest leverage point is a credible, documented evaluation of competing AI platforms. SAP fears customer churn to hyperscalers more than any other competitive threat.

AWS SageMaker as a Negotiation Lever

SageMaker includes:

• Pre-trained foundation models (similar to SAP's generative AI capabilities)
• AutoML for automated model building (comparable to SAP BTP AI services)
• Lower per-hour inference costs than SAP AI Core (typically 40-60% cheaper for similar workloads)

In negotiations: "We evaluated SageMaker and can build our AI roadmap entirely on AWS. SAP Core's advantage is ERP integration, but that's not a cost driver for us. Price SAP AI within 15% of SageMaker economics, and we'll stay in your ecosystem."

Azure OpenAI Services as a Negotiation Lever

Azure OpenAI provides access to GPT-4, GPT-4 Turbo, and Azure's own models with enterprise governance. Key advantages:

• Token-based pricing is predictable and transparent
• No lock-in; you can move models across cloud regions or providers
• Integration with Azure Synapse, Cognitive Services, and Power BI (if you're a Microsoft shop)

In negotiations: "Our finance team evaluated Azure OpenAI for regulatory and governance reasons. It's viable for our use case. We prefer SAP's ERP integration, but only if pricing reflects the hyperscaler alternative."

Google Cloud Vertex AI as a Negotiation Lever

Vertex AI bundles foundation models, AutoML, and MLOps with strong data governance and explainability. Key advantages:

• Lower entry-level costs than SAP Core (especially for startups and pilots)
• Vertex AI Workbench provides a Jupyter-like notebook environment (more familiar to data scientists than SAP's UI)
• Integration with BigQuery for analytics (if you're already on GCP)

In negotiations: "We're a Google Cloud house. Vertex AI is our baseline for AI. SAP Core only makes sense if it's cheaper or provides irreplaceable ERP integration. Right now, it's more expensive with fewer features."

Building Your Negotiation Team: Who Needs to Be at the Table

SAP negotiates differently depending on who shows up from your side. A procurement-only team gets a sales-friendly deal. A procurement + legal + IT + finance team gets a balanced agreement.

Procurement Lead

Owner of vendor relationships, pricing strategy, and contract terms. Procurement negotiates consumption caps, discounts, and payment terms. SAP's standard sales team reports to procurement.

IT/Architecture Representative

Validates technical requirements, SLAs, integration points, and exit strategy. IT should ensure:

• Consumption metrics are technically measurable and not ambiguous (e.g., "inference calls" not "compute units")
• SLAs for uptime, latency, and support response are realistic and enforceable
• Exit strategy (data export, decommissioning timeline) is operationally feasible

Legal/Contracts

Non-negotiable for any SAP contract >$500K annually. Legal should:

• Flag auto-renewal, termination, and liability clauses
• Ensure pricing escalation language is capped (3-5% annually max)
• Verify data retention, security, and compliance clauses align with your policies

Finance/Budget Owner

CFO or finance director who owns P&L for AI initiatives. Finance should:

• Define minimum ROI thresholds for AI investments (e.g., $500K annual benefit required to justify $800K annual spend)
• Verify consumption forecasts are realistic (IT often overestimates AI usage in pilots)
• Require true-up and reconciliation clauses in the contract so monthly invoices match actual consumption

SAP CIO or ERP Owner

Your internal SAP stakeholder (CIO or VP of SAP Operations). This person has political sway with SAP's account team and can escalate if SAP is inflexible. Include them late in negotiations (not upfront, to avoid SAP leveraging personal relationships with the CIO to push unfavorable terms).

Negotiation sequence: Procurement + IT + Legal first. Once you have a position, loop in Finance for cost validation. Only at the end, if SAP pushes back, escalate through your SAP CIO to SAP's account executive.

Pre-Negotiation Checklist for SAP AI Contracts

Post-Signature Monitoring and Renegotiation Triggers

Your negotiation doesn't end when you sign the contract. SAP often "scope creep" after signature, adding service fees, increasing consumption charges, or proposing add-ons that inflate costs.

Monthly Monitoring Activities

• Reconcile invoices to actual consumption: SAP's billing systems sometimes double-count consumption or bill at incorrect rates. Audit 100% of invoices in Month 1-3, then sample 20% thereafter.
• Track actual vs. forecasted consumption: If you forecasted 500M API calls/month and you're at 800M/month, you'll hit overages. Accelerate cost optimization or renegotiate the consumption cap.
• Monitor service health: If SAP misses the 99.5% uptime SLA in any month, document it and request service credits immediately (don't wait until contract renewal to mention it).
• Review SAP updates and feature changes: If SAP discontinues a model you rely on, or deprecates an API, document the impact and notify SAP that this triggers your right to renegotiate.

Renegotiation Triggers

The contract should specify conditions that allow either party to renegotiate without penalty:

• SAP exceeds pricing increase cap: If SAP proposes a price increase >5% annually, you can renegotiate or exit.
• Uptime SLA miss: If SAP misses 99.5% uptime in 2 consecutive months, you can demand service level credits or contract modification.
• Material feature deprecation: If SAP discontinues a core model or feature your contract depends on, you can renegotiate.
• Consumption variance: If your actual consumption is >20% different from forecast (in either direction), you can renegotiate the consumption cap to reflect reality.

Renegotiation Approach

When renegotiation is triggered, don't wait passively for SAP to propose changes. Be proactive:

FAQ: Common SAP AI Negotiation Questions