Every few years, a Salesforce renewal lands on someone's desk with a number attached that makes them pause. Usually the response is to negotiate, absorb the increase, and move on. The alternative is ripping out a platform that your sales, service, and reporting processes have run on for years. That has always sounded like a multi-year project with more risk than the savings could justify, and for a long time it was the right call. I believe that calculation is starting to shift, and for a real reason rather than a hypothetical one.
Our own (much smaller) test case
We made a version of this decision ourselves this year, at a much smaller scale. Harvest, the invoicing tool we've used for years, told us our monthly bill was going from roughly $140 to just over $2,200 as they moved customers onto their enterprise plan. That's a 15x increase for a tool that did one job for us: generate invoices and track payments. We had already built our own time tracking system that fed into it, so Harvest wasn't running anything mission critical for us.
I don't begrudge a software provider for adjusting its pricing to reflect the market. We run dynamic pricing on our own services for the same reason. Still, this move was a head-scratcher. Harvest has been elegant, easy to use, and reliable. But companies are aggressively investing in their own AI-assisted development capabilities right now, and raising the price of a tool that simple by 15x at this moment is a strange bet.
One of our consultants, Dustin Tinney, spent about 35 hours over two weeks, on top of his regular client work, building what we needed to leave Harvest completely. Two weeks. No quarter-long migration, no steering committee.
We don't think that timeline is unusual anymore. The tooling and scaffolding we've been investing in changed the math for a tool that small. Gartner makes a similar argument at a much larger scale: it estimates that as much as $234 billion in SaaS spending through 2030 is at risk as AI changes the economics of building versus buying.
Salesforce isn't Harvest
This is where an executive weighing the same decision for Salesforce should slow down. Harvest was low stakes for us: a single function, no deep integrations, and no years of accumulated business logic. Salesforce is almost never that. It's usually the system of record for your pipeline, the trigger for a dozen downstream integrations, and the tool your sales team has built five years of habits around. Replacing a platform like that isn't a two-week project, and anyone telling you otherwise is selling something. (Yes, I recognize the irony of a consultancy saying that.) The real cost of a Salesforce replacement is rarely the software itself. It's the migration, the retraining, and the risk of breaking a reporting process your board relies on.
The calculation has still moved, though. An honest version of this decision separates what is actually hard to replace from what only feels hard to replace because it's always been there.
What are you actually paying for?
Most platforms like Salesforce bundle two very different things. The first is a set of core capabilities that are genuinely complex: pipeline tracking, forecasting, and permissioning at scale. The second is a much larger layer of configuration, workflows, and reports that exist because your business is specific, not because the underlying problem is hard. That second layer is usually where both the licensing cost and the customization pain live. It's also the part that has gotten dramatically cheaper to build well.
Further, I expect this same split to reshape the SaaS vendors themselves. The products that hold up over the next five years will likely be a smaller core of licensed functionality at a reasonable price, plus a services layer for the customization work that has usually been bolted on badly. ERP vendors have run a version of this model for decades. Plenty of companies have spent millions on licensing, then spent millions more customizing the platform, or customizing their business to fit it. If you're evaluating a build versus buy decision today, the useful question is which of those two layers your pain actually falls into, rather than whether to replace "Salesforce" as a monolith.
When does a rebuild make sense?
A few questions tend to separate a real opportunity from an expensive distraction.
How much of your Salesforce instance is already custom? A heavily customized instance has effectively become bespoke software running on someone else's licensing terms. That's often the strongest case for a rebuild, because you're paying platform prices for something that's already closer to custom software.
How much of the system is actually in use? Many companies deploy large, complex SaaS platforms and end up using 10% or less of the features they're paying for. Paying for 100% of a product while using a fraction of it is pure waste. In my experience, though, the bigger cost lands on the people who use the system every day. Every unused feature adds surface area: more screens to click through, more fields to skip, and more configuration to keep working through every upgrade. That complexity keeps eating your team's time long after the invoice is paid.
What's mission critical, and what's just habit? Pipeline and forecasting data that feeds investor reporting carries a very different risk than a dashboard nobody outside RevOps looks at.
Are the integrations the product? If most of the value lives in how Salesforce connects to your other systems, the replacement conversation is really an integration conversation. That's a far more manageable problem than rebuilding a CRM from scratch, and it's one we've spent years helping clients solve.
None of this argues for replacing Salesforce wholesale on a whim. It argues for treating your next renewal as a real evaluation rather than a formality, which for most companies it hasn't been in years. We're still learning where the line falls for different businesses, one organization at a time. If you're facing a renewal and want a second set of eyes on which parts of your instance are actually hard to replace, I'd welcome the conversation.
(Source: Gartner, via ChannelDive, 2026)
Todd Kaufman is the CEO of Test Double, a 100% employee-owned software consulting company that has been remote since 2011. Test Double helps organizations solve complex technical challenges through people, process, and technology consulting.
