> ## Content Index > Fetch the complete content index at: https://www.halfofit.net/llms.txt > Use this file to discover other available public pages before exploring further. # Your SAAS renewal no longer make sense - URL: https://www.halfofit.net/your-saas-renewal-no-longer-make-sense/ - Published: 2026-09-26T11:48:56.000Z - Updated: 2026-09-26T11:52:09.000Z - Description: Your renewal came in higher "for AI features" you never asked for or adopted. How to tell whether you're renewing the same product - or repricing onto a new one? - Author: halfofit - Tags: Opinion For a few months I've been sitting with the renewals I worked on this year. The part of this job I actually enjoy is the commercial modelling: pricing a deal, shaping the terms, working with a sales team to land on a structure that best works for both of us. The other part I like is getting to see an absurd number of products up close, and working out what each one actually does once you get past the sales pitch. Sitting on that side of the table gives you a strange vantage point. You start noticing things about how software is sold that a regular user never has any reason to think about. This year, I kept noticing that the cost increases were largely attached to the same two letters in the product pitch: AI. Renewal after renewal came in with 35%, 50%, sometimes 100% increases, and almost every one of them justified it with AI. New AI features. AI across the platform. The increases weren't imaginary, either - supplier "AI uplifts" have been clustering in the [20-37% range](https://www.tropicapp.io/blog/what-is-a-credit-ai-pricing?ref=halfofit.net) across renewals. After about four months of watching, the question stopped being "how much can I get this down?" and turned into a basic question: am I even buying the same product I renewed every year before this? Last week I asked a supplier straight out: what changed? The account manager had it ready. "AI capabilities. New intelligent features. Enhanced automation across the platform." While he talked, I pulled up our own usage dashboard. The AI features were barely touched. The workflows the team actually ran were the same ones they'd been running eighteen months earlier. We were being asked to pay more for a version of the product that nobody had adopted. ## The product BOM There's a move procurement people make that gets misread constantly. We ask to see the price first, before the roadmap, before the demo, before anyone opens a deck. We're trying to answer a simple question: what does this number actually include, and what logic sits underneath it? A few years ago, that was a ten-minute conversation. Now it's one of the harder problems in the whole renewal cycle. I've started treating SAAS contract as having its own version of a bill of materials: seats, modules, credits, transactions, AI actions, support levels, minimum commitments and overage rates. The problem is that suppliers increasingly change that BOM between renewals. The line item still says the same product name. The invoice still arrives once a year. But what's underneath the price may have changed enough to make last year's figure almost useless as a comparison. Sometimes that's a deliberate packaging choice. More often, the product genuinely moved underneath the price, and the pricing followed it. Either way, the question is the same: **what, exactly, am I renewing?** And that's why I've come to think of a SAAS renewal differently. It's easy to treat it as a price negotiation: last year we paid X, you're asking for Y, so let's find a number somewhere in between. But increasingly, that's the wrong starting point. **A SAAS renewal has become a product-and-pricing architecture review.** ## The denominator moved Before Gen AI became a thing, a SAAS renewal was boring in the good way. The unit was a seat. You knew what a seat cost, you knew how many you had, you negotiated a discount for committing, you signed. Both sides knew the script well enough to skip to the end of it. That model still exists, but across a lot of business tools the unit underneath it has changed. Seat pricing made a person the thing you paid for. Credit pricing makes a unit of AI output the thing you pay for. Consumption pricing makes your team's behaviour the thing you pay for - Salesforce's Flex Credits, launched in 2025, meter roughly $0.10 per action. And outcome pricing makes the supplier's definition of a good result the thing you pay for. The rough progression: Seat -> credit -> consumption -> outcome. Intercom is the cleanest live example. Pull up its pricing and you'll see the seat prices have actually come down - over 2024 the entry plan fell from around $39 to $29 a seat, with the top tiers landing at $85 and $132\. If seats were still the whole story, that reads as a price cut. But sitting alongside every one of those seat prices now is a second meter: Fin, the AI agent, at [$0.99 per resolution](https://www.intercom.com/pricing?ref=halfofit.net) \- you're billed only when it actually closes a conversation. Watch what happened to the denominator. The unit you could forecast - how many seats you need - got cheaper. The unit you can't forecast - how many conversations the AI resolves in a given month - became the part of the bill that moves. Satya Nadella on a [Q3 FY2026 Microsoft](https://view.officeapps.live.com/op/view.aspx?src=https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/TranscriptQandAFY26Q3&ref=halfofit.net) (page 42) earnings call said "seat-based pricing is just entitlement to some consumption right". ## Your benchmark is measuring a product that's gone Every renewal benchmark you've built assumes the unit stays still. Cost per seat, year over year. Discount off list. Percentage increase against last year's baseline. All of it works by comparing the same unit to itself across time. Move half the bill onto a per-resolution meter and the comparison stops meaning anything. Last year you paid a figure per seat. This year you're paying a lower figure per seat, plus an amount per resolution that rides on your ticket volume, your customers' behaviour, and how often the model decides it "resolved" something. Comparing this year's blended bill to last year's per-seat price tells you almost nothing. Your benchmark is now a well-kept record of a thing that no longer exists. This is the part that costs real time. It isn't ordinary inflation, where a 5% bump on a stable model is a decision you can weigh in an afternoon. Every restructured renewal makes you re-learn the product's commercial architecture from scratch: what the unit is, what it means when your usage drifts outside the expected band, what happens in the month the AI has a busy quarter. The work of evaluating a renewal now costs more than it did the year before, and it does that every single year. \>**Fun fact:** *Across the top 500 SAAS and AI companies, there were more than 1,800 pricing changes in a single year - an average of 3.6 per company, with some suppliers changing their pricing every month \[*[*Kyle Poyar, PricingSaaS analysis, 2026*](https://www.growthunhinged.com/p/2025-state-of-saas-pricing-changes?ref=halfofit.net)*\].* ## Why AI is the accelerant Suppliers have always experimented with pricing. What's different now is that AI changes the economics underneath in a way that forces the unit to move. Traditional software ran at something like 75-80% gross margins because the marginal cost of one more user was close to zero. Once the code was written, serving it to the 10,000th user cost almost the exact same as serving it to the 10,001st: the supplier's cost didn't rise when your usage did. With Generative AI, every single prompt, long-context window retrieval, or multi-step agent execution triggers a **net-new variable cost** via cloud GPUs or third-party APIs (like OpenAI or Anthropic). If a customer heavily utilizes an AI feature under a flat-rate per-seat price, their variable compute cost can easily exceed the subscription fee, actively eroding the supplier's profitability. This margin pressure is the reason companies are aggressively shifting toward the consumption-based packaging. Something has to give, and what gives is the seat. Two years ago nearly every SAAS company launched AI as a separate, optional line item, testing whether buyers would pay for it à la carte. Most have since folded it back in. Microsoft ran the most visible version: in January 2025 it raised Microsoft 365 for consumers for the first time in twelve years, Personal from $69.99 to $99.99 a year - roughly 43% - and bundled Copilot into Word, Excel, PowerPoint and Outlook whether you open it or not \[[CNBC, Jan 2025](https://www.cnbc.com/2025/01/16/microsoft-raises-price-of-consumer-microsoft-365-first-time-since-2013.html?ref=halfofit.net)\]. The add-on you could decline last year is the base price you can't decline this year. ## A different way to read the renewal The reflex when a renewal lands high is to fight the headline number. Push the percentage, get it to something that fits the budget or looks good on a slide. Twelve months later the model shifts again and the same fight restarts from a higher baseline. The first number the supplier sends is doing more than anchoring. It's a question aimed back at you. Have you been paying attention? Do you know what changed? Have you opened the usage data and mapped the new unit onto what your team actually consumes? The question to start the conversation with is: *"Help me understand what I'm paying for now that I wasn't before - and show me the usage data that makes the case."* If they can answer it clearly, with your own consumption sitting in front of both of you, there's a real negotiation to have. If they can't, or won't, you've already learned what you came to learn.