Every small business I work with carries a stack of SaaS subscriptions that almost-but-don't-quite fit the way they actually work. The CRM is too generic for the niche. The booking tool wants the workflow inverted. The invoicing app is missing the one feature the accountant insists on. So the team works around the gap manually, every week, and pays €30, €60, €120 a month for the privilege.
This is the case for the alternative: a small custom internal tool, built once, owned outright, that fits your workflow exactly. Not for everything. For the specific parts of the business that are specifically yours.
Where SaaS genuinely makes sense
Some categories really are universal. Email. Calendar. Document signing. Payroll. Accounting. The off-the-shelf product is built by a team of fifty engineers, audited by lawyers, and used by ten thousand other businesses, which means it gets compliance, edge cases, and security baked in for free. Building your own version of these would be expensive, slow, and worse.
Buy the SaaS for those. The rest of this article is about everything else.
Where SaaS quietly fails small businesses
A small business has workflows that are not universal. The way you assign athletes to lanes. The way you reconcile yesterday's POS terminal totals with the booking calendar. The way you batch-print labels in three different courier formats depending on country. None of those are universal. None of them have a SaaS that fits exactly.
So a SaaS that almost fits gets adopted. A few obvious problems show up over time.
The 10% problem. Your team uses 10% of the platform's features, but pays for 100% of them. The other 90% is built for an enterprise you are not, and the platform's roadmap is driven by the enterprise's needs, not yours.
The workaround tax. Wherever the SaaS doesn't fit, somebody on the team works around it manually. Export to spreadsheet, sort by hand, re-upload. Every week. Forever. The workaround is invisible in the SaaS bill but real in the team's time.
The subscription stack. Five years in, the business has eight subscriptions, three of them overlapping, two of them only used by one person who left, and nobody can remember what the original problem was that each one solved. €40 + €60 + €25 + €80 + €30 + €15 + €50 + €90 = €390/month, before VAT, indefinitely.
Lock-in. Your data is in their database, in their schema, accessible through their export-to-CSV button on a good day. The day you want to switch tools, the export is incomplete, the new tool's import is incompatible, and you have already spent two years training the team on the current vendor's UI.
Pricing changes. The vendor was €15/month when you adopted them. Three years later they got acquired, "value pricing" rolled out, and the same plan is €60/month. Your dependency on the tool is exactly why they can do that.
The build-vs-rent math
Here is the actual calculation for a typical small-business workflow.
Say the SaaS that almost-fits is €60/month. A custom equivalent, built once and tailored exactly to your team, costs €3,000. The break-even is 50 months, just over four years. That sounds long.
But that comparison undersells the custom version on three axes.
- The custom tool is built around your real workflow, so the manual workaround disappears. If the workaround was costing you two hours a week at €25/hour of staff time, that is €2,600/year, recovered.
- The custom tool has zero ongoing rent. Hosting a small internal tool is roughly €0 to €15/month depending on whether it needs a database. The €60/month bill stops the day you cut over.
- You own it. If your business changes and the tool has to change with it, that is a small revision, not a vendor roadmap negotiation.
Add those up and the real break-even is closer to 12 to 18 months. After that, every month is pure savings.
This math does not apply to the universal categories above, where the SaaS team really is doing more for you than €60 of your developer's time would. It applies to the specifically-yours parts of the business, where what you are paying for is the gap between the SaaS and your real workflow.
What "small custom tool" actually looks like
The phrase "custom software" does a lot of damage. It sounds like a six-month enterprise project, so people don't ask. The reality, for the kind of internal tool a small business needs, is closer to:
- Two to four weeks of build time.
- One developer.
- A handover that fits in one phone call.
- A README the owner can read in ten minutes.
- Hosting on a free tier or a single small VPS.
This is the shape of every internal tool I have shipped. They are not impressive. They are useful. They get out of the way. The business runs on them, and nobody notices, which is the whole point. See Internal Tools for Small Businesses, A Practical Guide for the four shapes these tend to take.
When the rented version is still the right call
I am not going to tell you to build everything. There are real cases where SaaS still wins:
- The need genuinely is universal (see above).
- The volume is so low that even €120/year is too much to spend on a custom build.
- The workflow is going to change so fast that locking in a specific custom shape is premature.
- Compliance or auditability is the actual deliverable, and the SaaS already has the certifications.
For everything else, the question worth asking is not "should we buy a tool for this?" but "is this part of the business specifically ours?". If the answer is yes, the rent version will keep almost-fitting forever, and a small custom build will fit exactly.
Work with me
If you have a SaaS bill that has been quietly growing, or a workaround that the team has been doing every week for a year, the cheapest way to find out whether a custom tool would pay back is a free 30-minute call. Email info@tonibarisic.com or use the contact form. For the cost side of the build-vs-rent question, see How Much Does a Custom Internal Tool Cost. For the broader category overview, Internal Tools for Small Businesses, A Practical Guide is the umbrella post.