Off-the-shelf CRM or custom CRM? Eight test questions to decide

Are your processes largely standard?

If leads, contacts, activities, quotes, simple pipelines, email integration and standard reporting cover your everyday work, an off-the-shelf CRM is usually the most economical solution. If your workflows differ from that, the first thing to check is whether those differences are business-critical.

In most companies the following are relatively standardised: capturing leads, managing contacts and companies, documenting activities, following up quotes, running simple sales pipelines, connecting email and reporting through standard reports.

This question comes first deliberately, because it resolves the most common misjudgement: not every peculiarity of a company is a requirement for the software. Some deviations are pure habit — and adapting them to the standard saves an expensive special solution.

Does an off-the-shelf system cover 80–90% without special constructions?

What matters is not whether a vendor can theoretically enable every feature, but whether the system models your daily reality without permanent workarounds. The answer usually shows during the selection phase already.

Warning signs that become visible during selection

• Staff maintain the same data in several systems.
• Spreadsheets remain a central part of the work despite the CRM.
• Processes are simulated through free-text fields.
• Workflows only work through third-party tools.
• Important information sits in notes rather than in structured data.
• Staff have to know which buttons they must never press.

If these points already appear during the selection phase, they do not disappear after go-live. They become the operating state — and every one of them permanently costs working time that appears on no invoice.

Is tool sprawl developing?

The problem grows gradually: after the CRM come a quoting tool, project management, time tracking, a field service app, an ERP, commission accounting, reporting, document management, an automation platform and spreadsheets for the special cases. Every single tool can be good and the overall system still be bad.

The critical question is this: at the end of it, is there still one unambiguous source of data for customer, order, project and revenue?

When the same information is maintained in several places, the result is contradictory data, synchronisation errors, manual work, high integration costs and unclear accountability. The effort does not grow linearly with the number of tools, but with the number of transitions between them.

Do you have complex company structures?

Off-the-shelf CRMs work well when the world looks simple: one company, one customer, one contact, one deal. As soon as several legal entities, businesses, partners or commission models interact, the system has to model relationships it is often not designed for.

Take a trades group as an example: a holding company, several operating businesses, different regions, shared customers, various services, central administration, local sales staff, sales partners, subcontractors and different commission models.

Relationships a system like this has to represent

• A customer belongs to several businesses.
• A contact is relevant for several legal entities.
• An order is sold by business A and carried out by business B.
• A sales partner receives commission.
• The commission depends on product, region or margin.
• Revenue has to be consolidated and analysed separately at the same time.

If an off-the-shelf system is not designed for that, parallel worlds appear: the structure is rebuilt outside the CRM in spreadsheets, and the CRM loses exactly the role it was bought for.

Can roles and rights really be modelled cleanly?

Many systems solve permissions in a technically correct but operationally unhelpful way: users see functions, fields and status changes they will never need. The more unnecessary decisions a system offers, the more errors it produces.

The user interface is an underrated selection criterion. A member of staff typically sees functions they never need, fields that are irrelevant to them, administrative actions, status changes with far-reaching downstream processes, and buttons they can click although they should never use them.

What to check for complex processes

• Can each role be given its own interface?
• Can fields be hidden?
• Can actions be removed entirely?
• Can approvals be enforced?
• Can status changes be tied to conditions?
• Can users be prevented from creating technically invalid states?

If that cannot be modelled, the CRM itself becomes a source of errors — and correcting those errors is work nobody has planned for.

What are the real costs over three to five years?

The list price is almost never the relevant price. What has to be counted is licences, implementation, ongoing external costs, additional software and internal costs — a system at €50 per user can be considerably more expensive over several years than expected.

• Licence costs

  Cost per user, premium features, additional modules, API access, storage.

• Implementation costs

  Implementation partner, data migration, workshops, customising, training.

• Ongoing external costs

  Consultants, CRM agency, integration partner, administrators, support contracts.

• Additional software

  Automation, reporting, CPQ, document generation, commission software, e-signature, project management.

• Internal costs

  Manual double entry, media breaks, error correction, training effort, administration.

The most important question is not: what does the CRM cost per user? It is: what does our complete customer process cost us with this CRM?

When the off-the-shelf solution becomes more expensive than the custom one

With highly configurable platforms such as Salesforce, Dynamics or HubSpot, this calculation shifts with every user and every module: licences per user and edition, an implementation project with an implementation partner, add-ons for everything the chosen edition does not cover, and permanent administration or external consultancy. Beyond a certain number of users and special requirements, that total reaches a level at which a custom-built CRM you rent is level with it or below — a system developed by us is often cheaper than introducing Salesforce or Dynamics. That is not a rule, but it is common enough to be worth calculating both routes before you sign.

Whether the calculation works out that way in your case can be estimated together: we go through the number of users, the modules you need and your special requirements, and compare the costs over three to five years. 30 minutes, free of charge: Book an intro call.

How dependent on the vendor will you become?

Vendor lock-in belongs in the assessment before you sign the contract, not before you switch. A system can offer a formally complete data export and still tie you in strongly — because everything except the data stays behind.

Test questions before you sign

• Can all data be exported completely?
• Can the relationships between the data be exported as well?
• Are attachments and documents exportable?
• Can automations be exported?
• Is there an open API — and what limits does it have?
• Who owns custom extensions?
• What happens on termination, and how long does data remain available?
• Can another service provider take the system over?

The most critical case is the one where the customer data can be exported but workflows, automations, reports, permission logic and custom modules cannot. You then own the data but no longer the working system — and rebuilding it costs many times the export.

How flexible are contract term and scaling?

A CRM should fit not only today but the organisational structure planned for three years from now. What has to be checked: minimum term, automatic renewal, notice periods, minimum number of users, price increases, and the cost of additional and temporary staff, several legal entities and growing data volumes.

These points look like side issues during selection and later decide how much room for manoeuvre you have. A contract that is cheap at fifteen users can produce a different calculation at forty users and two legal entities — and that is precisely when switching is most expensive.

When is off-the-shelf software the better decision?

When your processes are standard for your market, at least 80–90% work without special solutions, few integrations are needed, roles and rights remain cleanly configurable and a single data base emerges. Custom software would then be unnecessary complexity.

Three economic conditions come on top of that: the special cases do not represent a strategic competitive advantage, licence and consultancy costs remain calculable in the long run, and the lock-in is acceptable. And one organisational condition: the company is prepared to adapt certain processes to the standard. In practice that means off-the-shelf software is the right entry point with a small budget and standard processes. As soon as you want to scale, staff or per-user licence costs become noticeable, or new business is meant to come out of the data, the calculation tips — and the custom option deserves serious examination.

When should custom software be examined seriously?

When several of these points apply at once — not when one of them is annoying. A single workaround does not justify building your own system; several of them occurring together does.

• The business model cannot be represented cleanly in the standard data model.
• Several legal entities or businesses have to work together.
• The same customers exist in different systems.
• Sales partners and complex commissions have to be modelled.
• Important processes still run in spreadsheets.
• Staff maintain the same data several times over.
• Numerous add-ons are needed and external consultants are permanently necessary.
• Users see functions that are dangerous or irrelevant for their role.
• Processes require many special approvals.
• The CRM increasingly determines the business process rather than the other way round.
• Reporting requires exports and manual consolidation.
• The cost of the off-the-shelf solution rises with every additional module.
• The vendor creates a strong vendor lock-in.

The third option: custom software you rent

Classic custom software has one economic drawback: the customer carries a substantial part of the project risk before the system runs productively for the first time. A rental model shifts the financial burden to the point at which the software is doing its work in everyday operations.

The typical sequence of a classic project runs: concept, quote, high project costs, months of development, go-live — and only then does it become clear whether the system really works in everyday use. At the point of purchase, custom software therefore rarely competes only against off-the-shelf software; it competes against the risk of a high upfront investment.

In the rental model the software is developed specifically for the company's processes, but the financial burden only begins with productive use. That combines the advantages of custom development — fitting processes and data models, role-specific interfaces, fewer unnecessary functions, less tool sprawl, one central data base — with an economic characteristic of SaaS: no classic large project investment before go-live.

We have described what a custom-built CRM looks like when customers hold several roles and contract relationships run in parallel: Custom CRM.

Quick check: twelve questions

For every question you answer with yes, score one point. The scoring below places the result — it does not replace a calculation, but it shows which direction the examination should take.

1. Do staff regularly use spreadsheets alongside the CRM?
2. Is customer data maintained in several systems?
3. Do you have several legal entities, sites or businesses?
4. Are there complex relationships between customers, partners and legal entities?
5. Do you need custom commission models?
6. Are external consultants regularly needed for CRM changes?
7. Do several additional tools have to be connected to the CRM?
8. Can users carry out actions they should not actually be able to carry out?
9. Is the existing system too complex for many users?
10. Does data regularly have to be exported and merged by hand?
11. Can core business processes only be modelled through workarounds?
12. Would changing CRM vendor be very difficult today?

Scoring

• 0–3 points

  Off-the-shelf software is probably sufficient. Start by optimising configuration and processes.

• 4–6 points

  A configurable CRM platform or tighter integration should be examined. Compare the costs over three to five years while you do.

• 7–9 points

  Custom software should be seriously costed as an alternative. The organisational costs of the off-the-shelf solution are probably considerable already.

• 10–12 points

  The company is probably working around its software more than the software is working for the company. A custom CRM or ERP architecture should at least be evaluated economically and technically.

The central decision

In the end the choice comes down to three questions: can off-the-shelf software model our core processes cleanly? Do we have to adapt processes, people and additional tools permanently to the limits of the software? Is the high upfront investment the main reason against custom software?

The best CRM is not the system with the most functions. It is the system that produces the fewest unnecessary decisions, workarounds and media breaks in your company.

Or put more pointedly: if your company has to adapt to the CRM more than the CRM adapts to your company, it is time to reassess the software decision.