We recently began working with one of the largest nonprofit membership organizations in the country with the goal of modernizing a legacy software system that has been running for nearly thirty years, with a focus on evaluating the feasibility of a custom ERP software development initiative. Several hundred internal and external users rely on the system for their daily work, and the aging technology has become a liability from a technical debt perspective; not to mention the bugs, security vulnerabilities and performance issues that come along with using a system built 30 years ago.
Their modernization project highlights a key aspect of the ERP and workflow software selection process: selecting the right software for your organization is not just about evaluating competing SaaS products, it’s an opportunity to rethink your software strategy as a whole. It’s a comparison between building something you own and renting something you don’t control.
The System That Outlasted Everything Else
Thirty years is an unusual run for any piece of software. It predates SaaS as a business model entirely. It also predates most of the cloud ERP platforms now pitching this organization on a life-long subscription. Whoever built their original legacy software system is very likely retired. And yet it has kept hundreds of people working every day for nearly three decades, without a subscription bill that climbs at every renewal and without a vendor deciding what stays in next year’s release.
Durability is the part almost nobody accounts for when comparing a custom build to an off-the-shelf ERP or any workflow software, for that matter. A system you own outright doesn’t answer to someone else’s product roadmap. There’s no per-seat rate hike waiting at renewal, and no risk that the company behind your ERP gets sold to a private equity firm that decides your module doesn’t fit the next five-year plan and quietly sunsets it. This organization’s 30-year run isn’t luck. It’s what happens when you invest the time and energy to build and own the thing you need, instead of leasing it.
The Risk Baked Into Every SaaS Subscription
Run the same 30 years forward instead of backward, and the math gets uncomfortable fast. Take NetSuite, one of the most common cloud ERPs, as a real example of current pricing: a base platform fee starting around $1,000 a month, plus $99 to $199 per full user per month. For an organization with 200 users, that’s somewhere between $250,000 and $540,000 a year, before a single price increase.
Hold that rate flat for 30 years, a generous assumption since per-seat pricing tends to rise at renewal, and the bill lands somewhere between $7.5 million and $16 million. Factor in even a modest 5% average annual increase, which is on the low end of what this market has actually done in recent years, and it’s closer to $17 million to $36 million.
There’s a risk in all of this that almost no one factors in: software companies get bought and sold. Private equity firms have built real strategies around acquiring SaaS platforms, raising prices at renewal, commandeering payment processors to pigeon-hole users to their bank and payment gateways, and eventually deciding a product line doesn’t fit the new owner’s roadmap. When that happens, you don’t get a vote. You get a migration deadline.
We’ve walked through a version of this reckoning in construction, where the SaaS apocalypse is forcing the same rent-vs-own math. ERP is no different, just with more zeros.
What Custom ERP Development Costs, By Company Size
There’s no single number that fits every organization, but there are real ranges, based on what these builds take to deliver.
| Company size | One-time build cost | Typical timeline | What’s usually included |
| Small (10–50 users) | $25,000–$60,000 | 2–4 months | One core workflow (e.g., accounting plus inventory, or membership plus billing), a single integration, straightforward data migration. |
| Mid-market (50–250 users) | $70,000–$150,000 | 6–10 months | Several connected modules (finance, inventory, purchasing, order management), two integrations, reporting dashboard, moderate data migration. |
| Enterprise (250+ users) | $115,000–$285,000+ | 10–18 months, often phased | Full departmental coverage, 3+ integrations, dedicated security review, and migration off a genuinely old system. |
The single biggest swing factor at every size is typically data migration. Moving clean, well-organized data into a new system is inexpensive. However, untangling three decades of inconsistent entries, retired fields, and one-off exceptions built up over that same span is very expensive, and it’s usually the reason a build costs more than the module count alone would suggest.
Owning vs. Renting Forever
For the organization in this story, the enterprise tier above is the right comparison. Let’s call it a $225,000 cost for the build-out including discovery, design, testing and change management. That’s in comparison to a cloud ERP subscription running at $225,000–$500,000 per year, every year, for as long as they keep using it.
There’s a simpler way to see this without doing 30-year math. A mid-size NetSuite deployment commonly runs $100,000 to $200,000 in total first-year cost, licensing and implementation combined. Building custom, once, lands in roughly that same range, and then the ongoing cost is hosting and maintenance, a small fraction of a recurring per-seat license. Renting has no ceiling, and you don’t get a vote on next year’s price.
We’ve made this TCO argument before, in more general terms. The math holds regardless of which type of system you’re replacing (ERP, CRM, Project Management, Procurement, Timekeeping, etc.).
Zooming out across company sizes, and counting maintenance on both sides of the comparison, since a cloud ERP still needs someone internally to configure, troubleshoot, and manage it, the gap only widens over a system’s real lifespan:
| Company size | Custom, built and maintained | Cloud ERP, subscribed and managed |
| Small (10-50 users) | Build: $25K–$60K Maintenance (30 yr): $113K–$360K Total: $138K-$420K | Subscription (30 yr): $1.4M–$3.9M Maintenance (30 yr): $150K–$300K Total: $1.6M–$4.2M |
| Mid-market (50-250 users) | Build: $70K–$150K Maintenance (30 yr): $315K–$900K Total: $385K–$1.1M | Subscription (30 yr): $5.7M–$12.5M Maintenance (30 yr): $600K–$1.4M Total: $6.3M–$13.9M |
| Enterprise (250+ users) | Build: $115K–$285K Maintenance (30 yr): $518K–$1.7M Total: $632K–$2M | Subscription (30 yr): $9.3M–$19.7M Maintenance (30 yr): $1.9M–$3M Total: $11.2M–$22.7M |
Of course, these are only high-level estimates based on our team’s experience and open-source, public pricing information. However, it’s a safe bet that your organization’s actual number will land somewhere inside these ranges. Furthermore, as you can see above, the discrepancy between the lifetime cost of Custom vs Subscription based software is dramatic enough to account for even major fluctuations in the estimation techniques used above.
The Levers That Move Your Number Up or Down
- How much data has to migrate, and how clean it is coming out of the old system
- How many outside systems it needs to talk to, like accounting platforms, payment processors, or industry-specific tools
- How many departments and workflows the system needs to cover on day one versus later
- Whether the data involved, financial, health, or payment, requires a dedicated security and compliance review
- How much reporting the organization needs, from a handful of dashboards up to a full analytics layer
The Commitment This Requires From You
Building your own ERP is genuinely realistic. Committing to it is the harder part. It works best when someone on your side owns the decision internally, not just approves invoices, when the scope gets broken into phases instead of one all-at-once cutover, and when you pick a partner who tells you what a feature costs before you commit to it, not after.
Do that, and the payoff isn’t only a lower number over time. It’s that you stop managing your business around somebody else’s release schedule.
If you’re weighing this decision for your own organization, we’d rather talk through the real scope than hand you a generic range. Pell Software is a 100% US-based team in Denver, and every project runs direct between you and the people actually building it. Reach out at contact@pellsoftware.com or (720) 504-1112.
1. We respond in 1 business day to schedule a 30-min call.
2. You speak directly with a US-based software expert, not a salesperson.
3. We’ll determine how we can help, or point you in the direction of someone who can!
