Most growing companies do not have a software spending problem. They have a software visibility problem. Finance sees the line items, department heads see their own tools, and nobody has a complete picture of what the business actually uses. A recent Vendr analysis found that the average mid-market company now runs 130 SaaS applications and wastes roughly 33 percent of its software budget on tools that are underused, duplicated, or abandoned entirely. That is real money leaking out of the business every month, usually in $50 and $200 increments that never trigger an internal review.
The good news is that cutting software costs does not require a painful consolidation project or an enterprise buying committee. A disciplined audit, a short list of consolidation rules, and a willingness to replace a few overpriced tools will typically recover 20 to 40 percent of your annual software spend in under a quarter. This guide walks through a repeatable framework for auditing your stack, identifying the right cuts, and keeping every workflow your team depends on.
Why Software Costs Quietly Spiral
Software procurement usually starts clean. A founder picks a CRM, an accounting system, and a project tool, and the stack looks rational on a whiteboard. Then the company hires. Every new department lead brings opinions about tooling. Sales wants a dedicated prospecting platform. Marketing wants a separate analytics layer. Customer success wants a ticketing tool that integrates with Slack. Within two years, the whiteboard diagram has 40 boxes and nobody can explain how they connect.
Photo by www.kaboompics.com on Pexels
The second driver is price creep. SaaS vendors routinely raise prices 8 to 15 percent a year, often quietly at renewal. Gartner research has shown that enterprise software is one of the fastest-growing IT categories, and a large share of that growth comes from existing customers paying more for the same features. Unless finance tracks year-over-year cost per seat, these increases slip through unnoticed.
The third driver is shadow IT. Individual contributors sign up for tools on personal credit cards, get reimbursed as expenses, and never register them as vendors. Productiv's 2024 State of SaaS report found that more than half of SaaS applications in a typical company are unknown to IT or finance. Those tools are not governed, not audited, and not part of any renewal review.
The Real Cost Is Not the Subscription
When teams think about software cost, they usually think about the monthly bill. That is only half the picture. The hidden costs include onboarding time, integration engineering, training, security review, and the productivity tax of context-switching between tools that do similar things. A $200-per-month tool with a painful onboarding experience and no API can easily cost more than a $500-per-month tool that plugs into your existing workflows cleanly. Any serious audit has to weigh total cost of ownership, not just the invoice.
The Four-Step Audit Framework
A useful software audit does not try to be comprehensive on day one. It focuses on the categories where waste is most likely and builds a ranked list of cuts from there. The following framework has worked consistently for businesses between 20 and 300 employees.
Step 1: Export Every Software Line Item From the Last 12 Months
Start with finance. Pull every recurring software charge from your accounting system, your corporate card provider, and your expense reporting tool. Match them against vendor contracts if you have them. The output is a single spreadsheet with vendor name, monthly cost, annual cost, owning department, and renewal date. Most companies are genuinely surprised by this list. Expect to find 15 to 30 percent more tools than anyone on the leadership team remembers paying for.
Step 2: Map Each Tool to a Business Capability
For each tool on the list, write down the specific business capability it delivers. Not the product category, the capability. "CRM" is too vague. "Track sales pipeline and log customer interactions" is the kind of capability statement that lets you spot overlap. When two tools have the same capability statement, you have a consolidation candidate. A typical mid-market stack will surface four to seven obvious duplicates in this pass alone.
Step 3: Pull Usage Data Before Making Any Decisions
Photo by Negative Space on Pexels
Cost data alone is not enough. A $300-per-month tool that three people use daily is worth more than a $100-per-month tool that 40 people logged into once last quarter. Every SaaS vendor worth paying for exposes usage metrics through an admin panel or an API. For the tools that matter most, pull actual login counts, active users, and feature adoption over the last 90 days. The Okta Businesses at Work report is a useful benchmark for what "healthy" usage looks like in different tool categories.
Step 4: Score Each Tool on Value Density
Create a simple scorecard for every tool: monthly cost divided by monthly active users, weighted by how critical the tool is to revenue or operations. Tools in the top quartile for cost per active user and the bottom quartile for criticality are your first cuts. Tools that score badly but sit in a critical workflow are candidates for replacement or renegotiation rather than elimination.
Where the Savings Actually Come From
Once the audit is complete, the cuts almost always fall into the same five categories. Knowing what to look for saves weeks of analysis.
Duplicate category spend. Most companies end up with two analytics tools, two project management tools, two customer support tools, and two password managers. One is a holdover from a previous team lead, one is the current standard, and both are being paid for. Cutting the holdover is usually the single largest line-item savings from any audit.
Unused seats on per-seat plans. Sales tool seats that belonged to employees who left six months ago. Design tool seats that were provisioned for a freelance contract that ended last year. Per-seat SaaS vendors rarely proactively audit this because they benefit from the inertia. A manual seat audit on the top five most expensive tools in your stack typically recovers 10 to 20 percent of those specific line items immediately.
Features that duplicate your existing platform. Your project management tool probably includes time tracking, your CRM probably includes email marketing, and your cloud provider probably includes a managed database that your team bought separately from a specialty vendor. When a core platform has already added the feature you are paying extra for elsewhere, the specialty tool is a candidate for removal.
Overpriced tools with free or open source equivalents. Some categories have matured to the point where commercial pricing no longer reflects the value delivered. Analytics dashboards, internal form builders, and simple reporting tools are the most common examples. Open source alternatives like Metabase and PostHog have reached production quality for many teams. Swapping a paid tool for an open source alternative makes sense when the delta in engineering effort is smaller than 12 months of subscription fees.
Tools that should be custom-built. The final category is the one most companies ignore. When a SaaS tool costs more than $25,000 per year and only does three things your team actually uses, a custom internal tool built on top of your existing database is often cheaper over a three-year horizon and more aligned with how your team actually works. This is where custom software development and data automation become strategic rather than discretionary.
Replacing Software Without Breaking Workflows
Cutting a tool is easy. Cutting a tool without disrupting the people who depended on it is the hard part. Three rules keep migrations smooth.
The first rule is to run the replacement in parallel for at least one full work cycle. If the tool is used for monthly reporting, run both tools for a full month. If it is used in the sales process, run both through a complete deal cycle. Parallel running surfaces edge cases that pure feature comparisons miss. The cost of the overlap is always smaller than the cost of a botched cutover.
The second rule is to preserve the data, not just the workflow. Before you cancel a subscription, export everything. Historical records, attachments, comment threads, custom fields, audit logs. Some SaaS vendors make this easier than others. The OWASP guide to data portability is a reasonable sanity check for what a clean export should include. Store the export in your own storage, not in another vendor's system.
"The single biggest mistake I see companies make during a software audit is treating it as a finance project instead of an operations project. If the people who use the tool every day are not in the room when you decide what to cut, you will cut the wrong thing, and you will hear about it for months. Bring them in early, show them the numbers, and let them defend what matters." - Dennis Traina, 137Foundry
The third rule is to document the workflow before you migrate. Not the product, the workflow. What happens when a new customer signs up? What happens when a deal closes? What happens when a support ticket is escalated? Write those flows down in plain language, then verify that the replacement supports all of them before switching off the old tool.
Photo by Walls.io on Pexels
Common Mistakes That Sabotage Software Audits
Three mistakes come up often enough to call out explicitly.
The first is cutting based on cost alone. A cheap tool that everyone uses daily is more valuable than an expensive tool that looks good on paper. If the scorecard does not account for usage and criticality, the cuts will feel arbitrary to the teams affected.
The second is negotiating renewals without data. Vendors routinely offer 15 to 30 percent discounts when the customer comes prepared with usage numbers and a specific alternative. Walking into a renewal call without that data is how you end up paying last year's price plus an increase.
The third is ignoring the integration layer. Many tools are sticky not because of the core feature set but because they are wired into the rest of the stack. Before cutting a tool, map every integration it touches. An unexpected Zapier connection or a webhook into an internal dashboard can turn a routine cancellation into a week of firefighting. Our guide to building resilient webhook integrations covers how to audit and replace these connections without breaking downstream systems.
More Resources
For teams running a software audit for the first time, a few additional resources are worth bookmarking.
- The Zylo SaaS Management Index publishes annual benchmarks on average tool count, spend per employee, and waste percentages by company size. Useful for setting internal targets.
- G2 Track offers a free tier for tracking SaaS spend and renewals across vendors, which lowers the friction of running these audits quarterly.
- The 137Foundry business technology guide explains how custom software and SaaS replacement decisions fit together for growing teams.
- For companies considering a move toward custom internal tools, our overview of web development services describes how a typical internal tool project is scoped.
A software stack audit is not a one-time event. The companies that consistently control their software costs repeat this exercise every six months, which keeps waste from compounding and gives finance a reliable lever to pull when budgets tighten.
The most important shift is treating software spending as a design problem rather than a procurement problem. Every tool added to the stack creates a permanent maintenance cost in attention, integration complexity, and vendor management. A smaller, more intentional stack almost always outperforms a larger one, even when the headline subscription numbers look similar. If you want a second set of eyes on your stack or help replacing an overpriced tool with something built around how your team actually works, 137Foundry is worth a conversation.
