Last month I sat with a business owner who was convinced his company barely spent anything on software. We pulled the credit card statements and added it up. His SaaS subscription costs came to Rp 14.2 million per month. That is more than he pays his most senior admin staff, and he could not name half the tools on the list.
Nobody decides to spend that much. That is exactly the problem. SaaS subscription costs do not arrive as one big invoice you can question. They arrive as Rp 150 thousand here, USD 12 there, a "per seat" fee that quietly doubles when the team grows. Each purchase felt cheap. The total is not.
I like SaaS. I build software for a living and I use plenty of it. But the industry has trained us to treat every monthly fee as a rounding error, and that training is costing Indonesian SMEs real money.
How subscription creep actually happens
Subscription creep is not a discipline problem. It is a structural one. The pattern looks the same in almost every company I audit:
- Someone hits a small friction. Scheduling meetings is annoying, so they sign up for a scheduling tool. Reasonable.
- The free trial converts silently. The card gets charged. Nobody notices because the amount is small.
- The person who signed up changes roles or leaves. Now the tool has no owner. It keeps billing.
- A second team buys a competing tool. Marketing uses one project tracker, operations uses another. Both bill per seat.
- Seats accumulate. People get invited "just in case." Each seat is Rp 100 to 300 thousand per month, forever.
None of these steps involves a bad decision. That is why willpower and "being more careful" never fix it. I have watched companies run a big cleanup, feel good about it, and be right back at the same monthly spend within a year.
The math nobody does
Here is a composite of what I typically find in a 20 to 40 person Indonesian company:
| Category | Tools found | Monthly cost |
|---|---|---|
| Communication and meetings | 3 to 4 overlapping | Rp 2 to 4 million |
| Design and marketing | 4 to 6 | Rp 2 to 3 million |
| Project and task tracking | 2 to 3 competing | Rp 1.5 to 3 million |
| Storage and file sharing | 2 to 3 | Rp 1 to 2 million |
| "Someone tried it once" | 5 to 10 | Rp 1 to 3 million |
The bottom row is the killer. In one audit I did for a retail chain in Tangerang, 9 of 31 active subscriptions had zero logins in the previous 60 days. That was Rp 2.8 million per month, Rp 33.6 million per year, paying for nothing. That is a decent laptop for every new hire, or a real chunk of a marketing budget.
And this is only the direct cost. Overlapping tools also cost you in fragmented data, in onboarding time, and in the "where is that file" tax every employee pays daily.
Willpower is not the fix. A rule is.
Here is the rule I install in every company I advise, and it is deliberately boring:
Every subscription must have a named owner and a cancellation review date. No owner, no renewal.
That is it. In practice:
- Named owner means one specific person, not a department. That person can explain what the tool does, who uses it, and what would break if it disappeared tomorrow. When they leave the company, the subscription is orphaned and automatically goes on the review list.
- Cancellation review date flips the default. Today, the default for every subscription is "renew forever." The rule changes the default to "cancel unless someone actively defends it." Set the review at 6 or 12 months. When the date arrives, the owner gets one question: do we keep this? If the answer takes more than a sentence, cancel it.
Put the whole list in a single spreadsheet: tool name, owner, monthly cost, seat count, review date, payment method. One sheet, reviewed quarterly, thirty minutes. Companies that do this typically cut 20 to 35 percent of software spend in the first pass without losing anything anyone misses. This is the same thinking behind data-driven pricing: stop guessing, look at the actual numbers, then decide.
A slightly contrarian note on "SaaS everything"
The current advice climate says every problem deserves a tool. I disagree more often than my industry peers would like.
Before subscribing to anything new, ask two questions. First, can an existing tool you already pay for do this at 80 percent quality? A shocking number of "we need a new tool" requests are solved by a feature already sitting inside Google Workspace or the project tracker you already own. Second, is this a recurring problem or a one-time one? A one-time problem does not deserve a forever fee.
And sometimes the honest answer is that a simple internal tool beats a subscription entirely. A shared sheet with a form in front of it costs nothing per month and does exactly what your workflow needs, nothing more. I wrote about that pattern in building internal tools without hiring a developer, and it applies here: owning a simple thing often beats renting a complicated one.
None of this means avoid SaaS. It means treat a subscription like a hire. You would never hire a person without knowing who they report to and when their performance gets reviewed. Apply the same bar to software.
Do this before the end of the month
A practical sequence you can finish in one afternoon:
- Pull three months of card and bank statements. Highlight every recurring software charge, including annual ones divided by twelve.
- Build the one-sheet inventory: tool, cost, seats, owner, last real usage.
- Anything with no owner or no logins in 60 days: cancel it today. You can always re-subscribe. You almost never will.
- Assign owners and review dates to everything that survives.
- Add the rule to your purchasing process: no new subscription without an owner and a review date written down first.
Most owners who run this exercise recover Rp 2 to 5 million per month within a week. Not by being frugal, and not by willpower. By replacing a silent default with a written rule. The leak was never the individual tools. It was the absence of anyone whose job it was to notice.