Improve the skill that is limiting your ability to deliver a useful result. If you can build quickly but struggle to define the problem, focus on business skills. If you understand the problem but cannot judge whether a solution is reliable, strengthen your technical skills.
Both matter, but your next step depends on your role
The World Economic Forum’s Future of Jobs Report 2025 identifies analytical thinking as a leading core skill, while AI and big data, cybersecurity and technological literacy are among the fastest-growing skills. That supports learning across both areas rather than assuming one replaces the other.
For Salesforce work, our recommendation is to keep a strong base in your role and learn enough of the adjacent discipline to make better decisions. A developer should understand the reason behind a request. A business owner should be able to question data access, limitations and the evidence behind an AI result.
Business skills help you choose the right work
Useful business skills include asking clear questions, mapping a process, deciding priorities and explaining trade-offs. They help turn a request such as “we need AI” into a specific problem that can be tested.
Consider a sales team asking for automatic lead scoring. Before selecting a tool, find out what makes a lead useful, whether that information is recorded consistently, and what a salesperson should do with the score. If nobody agrees on those answers, a technically sound build can still fail to help.
Start with the decision someone needs to make. Then work out what information and software would help them make it.
Technical skills help you test the answer
Technical understanding lets you check how a solution handles data, permissions, errors and changes. For Salesforce admins, this might mean data modelling, sharing rules, Flow design and testing. For developers, it includes integration design, code review, automated tests and monitoring.
When using AI-generated code or configuration suggestions, be able to explain the result before you deploy it. Test normal and failure cases in an appropriate non-production environment. A convincing explanation from a tool is not a substitute for checking what the implementation actually does.
You do not need to become a specialist in every area. You do need to recognise when a task requires someone with deeper experience.
Choose a skill based on the problem you keep facing
| Your current challenge | Skill to strengthen | A practical exercise |
|---|---|---|
| Requests change after you build them. | Discovery and acceptance criteria. | Write three real examples and agree the expected result before building. |
| You struggle to assess an AI suggestion. | Technical review and testing. | Explain the proposed change, then test an edge case and a failure case. |
| Your automation works but people avoid it. | User research and change support. | Watch a colleague complete the task and record where they hesitate. |
| Reports disagree across systems. | Data definitions and integration fundamentals. | Trace one record and document which system owns each field. |
| You cannot explain the value of a project. | Measurement and commercial judgement. | Compare the current effort and errors with the proposed result and full cost. |
A four-week learning plan
This is a suggested learning plan, not a promise of proficiency in a month. Choose a small exercise that fits your responsibilities and access permissions.
- Week one: observe. Pick one repetitive task and ask the people doing it what causes difficulty. Record a baseline.
- Week two: design. Map the process, define a useful result and list the data and permissions needed.
- Week three: test. Build a limited example with approved test data. Review it with someone whose skills complement yours.
- Week four: explain. Present what worked, what failed and what you would change. Include the impact on the user and the maintenance effort.
The deliverable should be something another person can review: a process map, test cases, a small prototype or a documented recommendation. That is more useful than measuring progress only by how many courses you completed.
How teams can support this learning
Pair business and technical colleagues on a real, limited problem. Give them time to understand each other’s decisions. Let the business colleague review the process and acceptance criteria, while the technical colleague checks access, implementation and failure handling.
For a Salesforce project, Adaptal’s consulting services can help turn business requirements into an agreed scope. Our managed services support ongoing improvements. Start with the work your team needs to do better, then decide what skills and support are missing.
Discuss your Salesforce project
Tell us what you need help with. We’ll get back to you.