The CFO's Guide to SaaS Cost Elimination (Not Just Optimization)

You've tried SaaS optimization. You've renegotiated contracts, consolidated vendors, killed unused licenses. Maybe you even hired a SaaS management platform to do it for you.

And the bill still went up.

That's not a failure of execution. It's a failure of strategy. You're optimizing a cost structure that's designed to grow. SaaS vendors build annual price escalators into every contract. They add "AI features" nobody requested and charge 30-45% premiums. They get acquired by private equity firms whose entire thesis is extracting more revenue from the same customer base.

Optimization is playing defense on a field that tilts 11.4% steeper every year. This guide is about playing offense. It's about eliminating SaaS costs — permanently — by owning the software your business depends on instead of renting it forever.

Key Takeaways

The SaaS Cost Spiral: Why Optimization Alone Doesn't Work

Here's a thought experiment. Your company spends $1.2M on SaaS today. You run a great optimization cycle — renegotiate contracts, cut unused licenses, consolidate redundant tools. You save 20%. Excellent. Your spend drops to $960K.

Now apply the 11.4% annual SaaS price increase. In 12 months, that $960K becomes $1.07M. In 24 months, $1.19M. By month 26, you're back where you started. All that work bought you two years of treading water.

This is the SaaS cost spiral. It works like this:

Year 1: Optimize. Save 15-25%. Feel good about it.

Year 2: Vendors raise prices 8-12%. Half your savings evaporate. New tools get added because teams still need things. Net savings: maybe 5%.

Year 3: Another round of increases. The AI surcharge hits. Your "optimized" stack now costs more than your un-optimized stack did two years ago.

The math is merciless. A 10% annual price increase compounds to 61% over five years. A 12% increase compounds to 76%. No amount of license pruning keeps up with that trajectory.

And this assumes your headcount stays flat. If you're growing — even modestly — per-seat pricing turns every new hire into a SaaS cost event. At $4,830 in SaaS spend per employee (2025 figures, up 21.9% YoY), adding 50 people means $241,500 in new annual software costs. Before they've done a single day of productive work.

The Real Numbers: What Your SaaS Stack Actually Costs

Let's get specific. Most CFOs underestimate their SaaS spend by 2-3x because shadow IT and decentralized purchasing obscure the real number. Here's what the data shows.

Per-Employee SaaS Spend by Industry (2025)

Those numbers are total spend. Now apply the waste layer.

The Waste Multiplier

According to BetterCloud's State of SaaSOps report, 30-53% of SaaS licenses go unused in any given month. Not underutilized. Unused. Nobody logs in. The seats sit empty.

For a 300-person company at the cross-industry average:

Vertice's analysis pins it more precisely: organizations with 100-500 employees waste approximately $915,000 per year. That's not a rounding error. That's headcount. That's R&D budget. That's margin.

For large enterprises, the number is $18 million annually in SaaS waste. Eighteen million dollars. Every year. On software that goes unused.

The PE Acquisition Tax

There's another cost that doesn't show up in benchmarks: what happens when your vendor gets acquired by private equity.

PE-owned SaaS companies have implemented price increases as high as 900% post-acquisition, as documented by Gartner's research on vendor risk management. The playbook is simple: acquire, cut costs (R&D, support, engineering), raise prices. Customers are locked in by switching costs and data gravity. Most pay the increase rather than migrate.

If your critical tools are owned by Vista Equity, Thoma Bravo, or similar firms, budget for 15-30% annual increases. That's not pessimism. That's their investor presentation.

Why SaaS Management Platforms Are a Band-Aid

You've probably been pitched Zylo, Vendr, Productiv, or one of the dozen SaaS management platforms that promise to "optimize your SaaS spend." They do three things well:

All useful. None of them solve the structural problem.

Here's why: these platforms optimize your rental agreement. They help you pay a fairer price for the privilege of renting software. But you're still renting. You still face annual price increases. You still pay per seat as you grow. You still have zero control over the product roadmap, the pricing model, or whether the vendor gets acquired next quarter.

Vendr negotiates a 15% discount on your Salesforce renewal. Great. Salesforce raises list price 10% the following year. Your "discount" is now 5%. The year after that, it's gone.

SaaS management is a necessary tactic. It is not a strategy. The strategy is ownership.

The Elimination Alternative: Own Instead of Rent

Here's what elimination actually looks like.

You take a SaaS tool your company pays $78,000/year for. You use maybe 20% of its features. You need it to integrate with three internal systems and follow your specific workflow — not the vendor's idea of a generic workflow.

Instead of renewing that contract, you build a custom replacement. Purpose-built for your exact needs. No features you don't use. No seats you don't need. No annual price increases.

The build costs $15,000-$45,000 depending on complexity. Maintenance runs $500-$3,000/month. Infrastructure costs $50-$500/month. After the initial build, your total annual cost is $6,600-$42,000.

You just replaced a $78,000/year recurring expense with a one-time investment and a fraction of the ongoing cost. And the gap widens every single year because your costs stay flat while the SaaS price keeps climbing.

This isn't theoretical. According to Retool's 2026 survey, 35% of teams have already replaced at least one SaaS tool with a custom build. And 78% plan to build more. eXp Realty eliminated roughly $1M/year in per-seat licensing. ClickUp saved $200K/year replacing their automation tools. A 12-person startup cut SaaS spend from $3,200/month to $480/month — an 85% reduction.

The economics have shifted. AI-assisted development has compressed build timelines by 30-55%. What used to take six months and $150K now takes 4-10 weeks and $15K-$45K. The payback period has collapsed.

CapEx vs. OpEx: The Financial Advantage of Owning Software

For a CFO, the SaaS-to-custom shift isn't just about total cost. It changes the financial character of your software spend in ways that matter for planning, tax treatment, and board reporting.

SaaS: Perpetual OpEx

SaaS subscriptions are operating expenses. They recur indefinitely, escalate unpredictably, and create ongoing P&L drag. You can't depreciate them. You can't capitalize them. They just... keep going.

Worse, they're variable costs disguised as fixed costs. Your contract says $78K/year — until the vendor raises prices, adds mandatory AI surcharges, or changes the packaging to push you into a higher tier. The "predictability" of SaaS is a myth. It's predictable the way inflation is predictable: always up, never down.

Custom Build: One-Time CapEx + Minimal OpEx

A custom software build is a capital expenditure. Under ASC 350-40 (for US companies) and IAS 38, internally developed software can be capitalized and amortized over its useful life — typically 3-5 years.

What this means in practice:

The Controller's View

For financial planning purposes, the shift looks like this:

5-Year TCO Model: SaaS vs. Custom

Let's build a real model. We'll use a mid-complexity business tool — think workflow automation, internal operations dashboard, or specialized CRM — currently costing $78,000/year in SaaS subscriptions with a 10% annual price escalator.

Scenario A: Keep the SaaS

5-year total: $476,198

And this assumes a modest 10% annual increase. If you're on a PE-owned platform or one adding AI surcharges, 15-20% is realistic. At 15% annual escalation, the 5-year total hits $527,000.

Scenario B: Build a Custom Replacement

5-year total: $153,000

The maintenance budget of $1,500/month ($18K/year) is intentionally generous — it covers bug fixes, feature additions, security updates, and dependency maintenance. Infrastructure at $300/month covers hosting, database, CDN, and monitoring. We've even added a $10K refresh in Year 5 for any modernization work.

The Comparison

The custom build pays for itself in 9 months. After that, every dollar is savings. By Year 5, you're spending $114,200/year on SaaS versus $21,600/year on owned software. The gap is $92,600 per year — and it widens every year after.

Scale this across 3-5 tools and you're looking at $1M-$1.6M in savings over five years. For larger enterprises replacing 8-10 tools, the number crosses $5M.

This is the 73% average savings figure you see in replacement case studies. It's not marketing. It's math.

Which Tools to Eliminate First

Not every SaaS tool is a replacement candidate. Your email provider, your cloud infrastructure, your core ERP — those stay. You're looking for tools that meet three criteria:

Here's the priority stack, ranked by typical ROI:

Tier 1: Replace Immediately (6-Month Payback)

Tier 2: Replace Within 12 Months (12-Month Payback)

Tier 3: Evaluate Case-by-Case (18-Month Payback)

A useful filter: if you would describe the tool as "good enough but frustrating," it's a candidate. That frustration means there's a gap between what it does and what you need — a gap you're paying full price to work around.

Building the Business Case for Your Board

The numbers make the case. But boards want structure. Here's the framework.

Executive Summary (1 Page)

The Four Arguments

1. Financial: Present the 5-year TCO model above. Show the compounding cost curve of SaaS vs. the flat cost line of owned software. Boards understand compounding — make sure they see it working against them.

2. Risk reduction: SaaS creates vendor concentration risk. Price increases, acquisitions, product sunset, data sovereignty issues. Every tool you own is a risk you've retired. Emphasize the PE acquisition trend — if your vendor is PE-backed, a 50-100% price increase is a matter of when, not if.

3. Competitive advantage: Custom tools match your workflow exactly. No compromises, no workarounds, no forcing your process into someone else's software. Teams move faster with tools built for how they actually work.

4. Asset creation: SaaS subscriptions are pure expense. Custom builds create a depreciable asset on your balance sheet. You're converting ongoing OpEx into a one-time investment that your company owns permanently.

Key Metrics to Track

The Objection Playbook

"What about maintenance?" Budget 15-20% of build cost annually for maintenance. Even at the high end, it's a fraction of the SaaS renewal. A $500-$3,000/month maintenance retainer covers everything.

"What if the builder disappears?" You own the code. Any qualified developer can maintain it. You can't say the same about a SaaS vendor — if they get acquired or shut down, you're migrating on their timeline, not yours.

"SaaS is faster to deploy." True for generic needs. False for workflow-specific tools where you spend months configuring the SaaS to do what a custom build does out of the box. Build timelines have compressed to 4-10 weeks for most replacement projects.

"We don't have internal engineering capacity." You don't need it. Purpose-built replacements are scoped projects with fixed timelines and fixed budgets — $5K-$15K for quick builds, $15K-$45K for core replacements, $40K-$80K for platform builds. Compare that to what you'll pay your current vendors over the next three years.

How to Get Started Without Risk

Here's the honest truth: you don't need to commit to anything to find out if elimination makes sense for your company.

Step 1: Get a SaaS audit. A proper audit surfaces your real spend (not what finance thinks it is), identifies unused licenses, maps duplicate tools, and flags the highest-ROI elimination candidates. We do this for free. No commitment. No sales pitch. Just the numbers.

Step 2: Pick one tool. Start with a Tier 1 candidate — a workflow automation tool, an internal admin tool, or a form builder. Something with a clear 6-month payback and limited integration complexity. Prove the model on a low-risk target.

Step 3: Measure and expand. Track the actual savings against the projection. When the board sees real dollars recovered — not theoretical savings from a vendor negotiation, but actual line items removed from the budget — the conversation shifts from "should we do more?" to "how fast can we move?"

The companies that start eliminating SaaS costs now will have a structural cost advantage over the next five years. The ones that keep optimizing will keep running on the treadmill — negotiating harder every year just to stay in the same place.

You already know the SaaS model is broken. The question is whether you're going to keep paying for it.

Get your free SaaS audit. We'll show you exactly which tools are bleeding your budget — and what it would cost to own them instead. No commitment. Just math.