The B2B SaaS renewal trap is a pricing model, not an accident.

Pattern 1

Auto-escalators

The clause: the contract includes a "price adjustment" of X% per year, compounded, regardless of usage change or market conditions. The typical range is 5-10% annually. The math on a 7% compound escalator: a $100,000 year-one commit becomes $107,000 in year two, $114,490 in year three -- a 22.5% increase over the contract period with no change in scope or value.

Named examples from public-record sources (SEC filings, GSA schedules, and well-sourced secondary reporting as of April 2026): Workday has historically included 5% annual escalators in multi-year ELA agreements; Oracle Cloud Infrastructure (OCI) contracts frequently include 3-8% floor escalators; ServiceNow enterprise agreements commonly include 3-7% annual adjustments; HubSpot hub-specific contracts have included 5-10% escalators depending on package.

Tactic: escalator removal or capping at CPI is always negotiable before signing. Once signed, it is contractually binding and extremely difficult to remove. Your leverage is at signature -- after that, escalator removal typically requires a material scope change (large expansion, new product line purchase) as a trigger for renegotiation. See the contract red flags page for the exact clause language.

Pattern 2

Multi-year lock-ins

The pitch: "Sign a three-year deal and we will give you 15% off year one." The trap: early-termination fees are typically 50-100% of the remaining contract value. A buyer who signs a three-year $300,000 deal (discounted from $360,000) and needs to exit at month 18 owes 75% of the remaining 18 months -- approximately $101,250 -- in addition to having already paid $150,000. The net position versus not signing is often worse than the discount.

Vendors that lean heavily on multi-year structures (as of April 2026): Snowflake and Databricks (cloud data platforms, typically 3-year consumption commits with annual true-ups), Datadog (infrastructure monitoring, 3-year platform commits common at mid-market), Workday (multi-year ELA is standard above 500 employees), Oracle (multi-year cloud and on-premise agreements are the primary model).

Tactic: negotiate the ETF clause separately from the discount. A standard counter is to cap the ETF at 25-50% of remaining contract value (not 100%), or to negotiate a specific exit ramp: "Buyer may terminate for convenience after Year 1 with 60 days notice, with no ETF after Year 2." This protects the buyer if business conditions change while still giving the vendor the multi-year revenue signal they want.

Pattern 3

Net-new-license (additive) pricing

The mechanism: your negotiated contract sets a discounted price for the seats or units in the deal. Any growth -- new hires, new API calls, new projects -- is billed at "then-current list price," not at your negotiated rate. For fast-growing companies, net-new-license pricing turns a 20% discount into a blended rate barely below list within 12 months.

Classic examples (April 2026): Notion team plans (per-member pricing, growth seats at full rate); Slack (per-active-user pricing, new active users at list); Datadog (per-host pricing, infrastructure growth at list); HubSpot per-contact tier (contacts purchased in tier upgrades at list price). These are not obscure -- they are by design.

Tactic: negotiate "discount protection for growth" -- the discount rate applies to any net-new seats or units added during the term. Alternatively, pre-buy a buffer of seats (say, 20% above current headcount) at the negotiated rate. Or negotiate a usage-based model instead of seat-based, with a committed minimum and overage at a negotiated rate.

Pattern 4

Shelfware

Shelfware is purchased SaaS licenses that go unused. Gartner's 2024 SaaS Management Survey found 25-40% average shelfware across mid-market organizations. Shelfware becomes a renewal trap because most B2B SaaS contracts renew on current seat count -- not active-user count. The vendor has no incentive to flag unused licenses; the renewal desk is measured on ARR retention, not usage optimization.

Common shelfware vectors: Salesforce (sales rep seats held for months after rep departure; admin licenses for departed ops staff); Zoom (Large Meeting licenses purchased for an event that does not recur); Microsoft 365 (E5 licenses assigned to users who only use E3-tier features -- common after a "we should all be on E5" executive decision that never translated to actual E5-feature adoption); Slack (full workspaces with multiple legacy channels where 30-40% of accounts are inactive).

Tactic: run a monthly-active-user audit 60 days before renewal. Pull usage logs from the vendor portal (most enterprise SaaS vendors expose these in the admin console). Map active users against licensed seats. Drop unused seats -- vendors typically allow a "true-down" of up to 10-15% at renewal with no fee for some products. For deeper cuts: negotiate the true-down explicitly in the renewal negotiation.

Most B2B SaaS renewals trip at least two of these four patterns. The compound math: a 3-year multi-year deal signed at 15% off list, with a 7% annual escalator, 30% net-new-license growth at list price, and 25% shelfware throughout.

The multi-year discount is real, but the compounding of the four patterns typically exceeds it within 18-24 months. The procurement playbook shows how to model this math before signing.