A 25-person operations team is an interesting size. You are past the spreadsheet phase. You have enough people that shared tooling matters. You probably have at least one person whose job touches systems, reporting, or process design. You are also not yet at the scale where a full enterprise platform runs itself.
At this size, the Salesforce question comes up almost every year. Sometimes it is the right answer. Sometimes it is an expensive detour. The honest answer depends on what your team actually does, not on what category of software you think you need.
Where Salesforce Genuinely Shines
Salesforce is not a weak product. It earned its reputation for real reasons, and pretending otherwise does not help anyone make a good decision.
It is a strong fit when:
- Your primary workflow is a sales pipeline with leads, opportunities, stages, and forecasting
- You need mature reporting on pipeline health and rep performance
- You expect to plug in well-known ecosystem tools like marketing automation, CPQ, or industry clouds
- You already have at least one Salesforce admin or a consultant you trust
- Your compliance or governance posture is built around an enterprise vendor
If most of those are true, Salesforce is doing work that would be expensive to recreate. You are buying a tested CRM data model, a huge partner ecosystem, and a predictable hiring pool of admins and developers. For a sales-led business, that is real value.
The Hidden Costs at 25 People
The sticker price is rarely the full story. Teams in the 20-to-40 person range tend to get surprised by the same line items.
Per-seat licensing that grows with headcount. Enterprise or Unlimited editions are where most of the interesting features live, and the per-user cost compounds quickly. Adding five operations staff next year is not a rounding error.
Add-ons that are separately licensed. CPQ, Service Cloud, Experience Cloud portals, advanced analytics, sandboxes, additional API capacity. Features that feel like they should be included often are not.
Implementation and consultants. Standing up a meaningful Salesforce org usually involves a partner. Ongoing changes often do too. For a 25-person team without a full-time admin, the consultant line item can exceed the license line item.
Certifications and admin labor. If you do hire internally, Salesforce admins and developers are not cheap, and they expect a career path. That means certifications, training budgets, and eventually more than one of them.
Upgrade and release churn. Three releases a year, periodic deprecations, and ecosystem apps that sometimes lag behind. Someone has to watch that.
None of this makes Salesforce a bad product. It does mean the total cost of ownership at 25 people is often two to four times the quoted license figure once you add consultants, admins, add-ons, and internal time.
When a Custom Portal Wins
A custom-built operations portal is the better choice when your workflow does not map cleanly onto a CRM-shaped object model. That is more common than people expect.
Specific situations where custom pulls ahead:
- Niche workflows. You have a process that is core to your business and genuinely unusual. Field service dispatch with specific routing rules, a claims intake pipeline, a custom approval chain, a quoting engine with real pricing logic. Forcing this into Salesforce objects means fighting the platform every release.
- Operations, not sales. Your team does fulfillment, service delivery, compliance, or internal coordination. You do not have a pipeline in the traditional sense. The CRM metaphor adds overhead without adding value.
- Tight integrations with legacy tools. You are connected to an ERP, an accounting system, a warehouse management tool, or an industry-specific system that does not have a polished Salesforce connector. You will be writing custom integration code either way, so the hosting environment matters less than the integration logic.
- Data that is not customer-shaped. You are tracking assets, jobs, inventory, shipments, or cases that do not fit naturally into Account, Contact, Opportunity, or Case.
- A small, stable user base. At 25 users, per-seat economics hurt more than they help. A single-tenant custom app amortizes very differently.
Custom also wins when you want the roadmap to belong to you. No forced migrations. No features deprecated out from under you. No platform update that breaks a critical flow two days before month-end.
The 5-Minute Decision Test
Answer these honestly. Count the answers on each side.
- Is your primary workflow a sales pipeline with leads, opportunities, and forecasting? Yes = Salesforce. No = custom.
- Do you already have a Salesforce admin on staff or a consultant on retainer? Yes = Salesforce. No = custom.
- Do you expect to use three or more Salesforce ecosystem products (Pardot, CPQ, Service Cloud, etc.)? Yes = Salesforce. No = custom.
- Is your core process unusual enough that standard CRM objects feel like a bad fit? Yes = custom. No = Salesforce.
- Are most of your integrations with legacy internal systems rather than SaaS tools? Yes = custom. No = Salesforce.
- Do you care more about predictable fixed costs than about ecosystem breadth? Yes = custom. No = Salesforce.
- Is your team sales-led or operations-led? Sales-led = Salesforce. Operations-led = custom.
Four or more Salesforce answers and you should probably be on Salesforce. Four or more custom answers and you will likely spend less and ship faster with a purpose-built portal.
If it lands three-to-four, the tiebreaker is usually whether sales or operations is the center of gravity. Salesforce is built around selling. Custom is built around whatever you build it around.
If you want a direct, non-salesy read on which side of this line your team sits on, contact Merkra. We will tell you honestly when Salesforce is the right call and when it is not.