Improve your existing Salesforce setup—or start over?
When your team is frustrated with Salesforce, starting over can sound appealing. But before choosing a rebuild, separate two questions: what is making work difficult, and how much of the current setup actually needs to change?
A confusing handoff, an unreliable report, and an outdated process may feel like one system-wide problem. They do not necessarily require the same solution. Start with evidence from the work people are trying to do.
Follow one process before judging the whole system
Choose a workflow that matters: qualifying an inquiry, preparing a proposal, handing a sale to delivery, or coordinating a customer request. Walk through it with the people who perform it.
Ask where information first enters the process, who owns the next step, and what causes someone to leave Salesforce for a spreadsheet or email. Identify what works as well as what fails. A useful assessment should protect the parts your team already depends on.
For example, imagine a sales-to-delivery handoff that repeatedly stalls. Is necessary information missing? Is responsibility unclear? Is someone waiting for a notification? Or does the business now need a different approval process? These are investigation questions, not reasons to assume a particular technical fix.
Compare three possible approaches
| Approach | When it is worth investigating | What to establish first |
|---|---|---|
| Targeted improvement | The difficulty is concentrated in a specific workflow, report, or handoff. | Whether the proposed change can be isolated without disrupting dependent work. |
| Phased redesign | Several connected processes need attention, but useful parts of the current setup can remain. | A practical sequence, clear ownership, and how the old and new processes would coexist. |
| Broader rebuild | The existing design appears fundamentally misaligned with the business's current needs. | Evidence that narrower changes are insufficient, plus the implications for data, integrations, users, and transition. |
Treat this comparison as a starting point, not a diagnostic score. An assessment may reveal a different scope from the one you initially expected.
Start with the workflowWhat prevents the current operating model from working reliably?
How much of the current foundation is still sound?
01 · IMPROVE
Keep the foundation.
Repair workflow, governance, adoption, or configuration when the platform still fits.
- Core data model remains useful
- Main gaps are process or configuration
- Change risk exceeds replacement value
02 · SELECTIVELY REPLACE
Change the weak layer.
Preserve what works while replacing a constrained tool, handoff, or integration.
- Failure is concentrated in one layer
- Boundaries can be defined safely
- Phased change reduces disruption
03 · REBUILD
Redesign the operating system.
Rebuild when the existing foundation cannot support the required workflow or controls.
- Architecture blocks essential work
- Risk is structural, not cosmetic
- A governed transition can be owned
Decision principle: Validate assumptions, dependencies, data, controls, and ownership before committing to a path.
Use six questions to explore your options
These questions help you see whether a focused fix, a phased change, or a broader rebuild deserves a closer look. This is a starting point, not a diagnostic score or recommendation. The comparison table and full guide remain available without this tool.
What to establish first:
An assessment may reveal a different scope from the one you initially expected.
Ask what must be preserved
Before deciding what to replace, identify the business history, working processes, reports, and connections that matter. Include the less visible dependencies: a report used at month-end, an integration maintained by another team, or an exception that only one employee knows how to handle.
A fresh design is not a complete plan unless it explains what happens to those dependencies. Ask how the proposed approach will be tested and how people will recognize an acceptable result before rollout.
Evaluate the recommendation—not just the proposed features
A useful recommendation should answer:
- What specific business problem does this address?
- What evidence supports the proposed scope?
- Which alternatives were considered, and why were they set aside?
- What assumptions still need to be checked?
- What will remain unchanged?
- Who will maintain the result, and what will they need to understand?
- How will the transition be reviewed before it affects everyday work?
Be cautious about choosing a rebuild solely because the current setup is old, or a small fix solely because its initial scope looks simpler. The decision should account for the whole workflow and its dependencies.
Agree on what “better” means
Define the intended result in terms your team can observe: a handoff with a clear owner, a report that answers an agreed question, or fewer places to enter the same information. If you want to measure an improvement, first establish a baseline and a consistent way to compare results.
Do not turn an aspiration into a guaranteed outcome. Record uncertainties and decide what needs to be learned before making a larger commitment.
Start with a focused conversation
You do not need to decide between improvement and rebuilding before speaking with a consultant. Bring one workflow, the main difficulty, the people involved, and the result you want. Avoid sending credentials or raw customer records in an initial inquiry.
Use the one-workflow conversation worksheet.
Halo Business Systems helps Cincinnati-area organizations with Salesforce business analysis, implementation, workflow automation, integration, and connected business systems. Explore Halo's Salesforce consulting services to discuss your starting point and the questions a scoped assessment should answer.
