How sprawl gets there in the first place
No SME sets out to run 80 software subscriptions. The number arrives slowly and almost always for reasonable-sounding reasons. A sales lead needs a prospecting tool for a quarter and the trial converts. The marketing team picks up a second analytics product because the first one cannot answer one specific question. Engineering signs up to a logging service to debug an incident and never decommissions it. HR adds an onboarding app because the previous head of people preferred it. None of these decisions is wrong on its own. Together, they are SaaS sprawl.
The same dynamic that makes SaaS easy to buy makes it almost impossible to stop. A €29 per seat per month line item never triggers a procurement review. It clears any expense policy. Three of them across a 30-person company is €31k a year, and nobody in the building is responsible for noticing that the total grew.
By the time a finance director starts pulling vendor lists, the typical mid-sized company is running between 70 and 130 distinct SaaS products. Industry benchmarks from Productiv, BetterCloud and Zylo all land in roughly the same range, and the SMEs we work with rarely beat it. The split between tools that are actively used, tools that are licensed but mostly idle, and tools that nobody on payroll has touched in a quarter is typically something like 40, 35 and 25 percent. The last bucket is pure tax.
The four invisible costs, not just the licence fee
When SMEs do try to size the problem, they usually look at the line items in the bank feed. That captures roughly half of what sprawl actually costs. The other half is harder to invoice and easier to ignore.
- Direct licence waste. Seats provisioned for people who left, plans on the wrong tier, annual contracts renewed without negotiation, and overlapping products that solve the same job in two different teams. For an SME between 25 and 150 people, this typically runs €15k to €30k a year. It is the only bucket you can recover by reading invoices.
- Integration and glue work. Each new tool needs to talk to the others. Zaps, Make scenarios, custom scripts, manual CSV exports between systems that should have shared a database. The cost is rarely a vendor invoice — it is hours of someone in operations rebuilding the same pipeline every time a tool changes its API or pricing.
- Cognitive load on operators. Every product has its own login, permissions model, notification queue and quirks. A salesperson spending 20 minutes a day navigating five overlapping tools instead of one is not a line item, but at company scale it dwarfs the licence cost. This is the bucket that quietly justifies headcount that should not need to exist.
- Risk and compliance debt. Each SaaS product is a place customer data can live, a vendor that can change terms or be acquired, a login that can leak. SMEs typically have no inventory of which tools store personal data and which do not. Under GDPR that is not theoretical risk — it is a finding waiting to happen during the first audit that takes it seriously.
Added together, the four buckets routinely reach €30k to €50k a year of avoidable spend for a 50-person company, and double that by the time you cross 100. The licence-fee waste is the smallest of the four. It is also the only one most people measure.
How to audit your stack in a week, without buying another tool
The instinct when sprawl gets visible is to buy a SaaS management product to fix it. That is the joke. A workable audit takes five working days and a spreadsheet, and it has to happen before you decide whether tooling can help.
Pull the bank feed and corporate card statements for the last 12 months and flag every recurring software charge, including the ones in personal expense reports. Cross-check with the SSO provider if you have one — the gap between "what we pay for" and "what people log into" is usually the most expensive line in the audit. For each product, capture four things: who owns it, what job it does, how many active users it has in the last 30 days, and what would happen if it disappeared next Monday. The last question is the one that matters.
The output is a list sorted into four piles. Tools that are clearly load-bearing and used daily by the team that pays for them. Tools that overlap with another tool already in the stack. Tools whose owner has left or moved roles and that nobody currently champions. And tools that exist because of a single specific need that may or may not still be active. Three of those four piles are candidates for cancellation, downgrade or consolidation. The first pile is the only one the company should be paying full price for.
An honest audit usually reclaims 20 to 35 percent of the SaaS spend within a quarter, without anybody losing a tool they actually use.
What to do once you have the list
Cancellation is the easy part. The harder decision is what replaces what. Two patterns work for SMEs and one does not.
The pattern that works is consolidation onto a smaller number of products that each cover more ground, even if no single one is best-in-class for a specific job. A company running HubSpot and Pipedrive and a third lead-scoring tool is not three times better at sales than a company running one of them well. The pattern that also works is targeted custom internal tools for the workflows that two or three SaaS products are stitching together badly — usually around how orders, customers or operational records flow between systems. The pattern that does not work is replacing a SaaS product with a slightly cheaper SaaS product without addressing why the first one was there. That is sprawl pretending to be discipline.
Pick the consolidation first because it is reversible. Custom internal tooling second, and only where the audit shows you are paying multiple vendors to do something the business clearly considers core.
When a custom internal tool quietly beats the subscription
Not every SaaS subscription should be replaced by something built. Most should not. The ones worth building tend to share three traits and they are easy to spot once you have the audit in front of you.
First, the workflow is specific to how the company actually operates and not generic enough for an off-the-shelf product to fit without bending. Second, the company is currently paying two or three vendors to approximate it, with manual glue between them. Third, the data that flows through the workflow is the kind that the company already considers proprietary — customer records, internal pricing logic, the specifics of how a service is delivered. When all three are true, a focused internal tool built once and maintained lightly tends to outlast any combination of subscriptions, and the total cost over three years is usually below what the subscriptions alone would have charged.
When even one of the three is missing, stay on the SaaS. Building software you do not need is a more expensive mistake than paying for software you barely use.
The AEKIOS take
SaaS sprawl is not a procurement failure. It is the predictable consequence of a stack that nobody is responsible for as a whole. The fix is not another tool. It is one named person inside the company who owns the software list end-to-end, a yearly audit that is treated with the same seriousness as a finance close, and a willingness to consolidate or replace the two or three workflows where the company is clearly paying three vendors to do one job badly. Most SMEs that take this seriously recover the cost of doing it in the first quarter.
Frequently asked questions
How much does SaaS sprawl actually cost a 50-person SME
Direct licence waste alone typically lands between €15k and €30k a year for a 50-person company. Once integration glue work, cognitive load on operators and compliance exposure are added, the total avoidable cost is closer to €30k to €50k a year. The pure licence figure is the smallest of the four buckets and the only one most SMEs measure, which is why the problem looks smaller than it is.
Do we need a SaaS management platform to fix sprawl
No, and buying one as the first move usually adds a line item without solving the underlying problem. A five-day audit using bank statements, SSO logs and a spreadsheet captures most of the value. Dedicated SaaS management platforms only start to pay for themselves once the company crosses roughly 150 employees or runs more than 100 active subscriptions with multiple owners.
When does building an internal tool beat staying on SaaS
When three conditions hold at once: the workflow is specific to how the company operates, the company is already paying two or three vendors to approximate it with manual glue between them, and the data flowing through it is something the company considers proprietary. If even one of the three is missing, the SaaS is almost always the cheaper option over three years.
Who should own the SaaS list inside an SME
One person, not a committee. In SMEs without an IT function this usually sits with operations or finance, and the role is less about gatekeeping new purchases and more about running an annual audit, holding owners accountable for active usage and consolidating overlapping products. Sprawl returns within twelve months in any company where this responsibility is shared between three departments because none of them owns the total.