There is a quiet shift happening in how Salesforce work gets done.

Not the headline-grabbing kind — not Agentforce announcements or Einstein rebrands. Something more fundamental. The tools consultants use to do the work are changing. And if you are not paying attention, you are already behind.

At JourneyForce, we have started using Claude — Anthropic's AI coding assistant — as a core part of how we deliver Salesforce projects. Not as a novelty. Not as a chatbot bolted onto a demo. As an integral part of our consulting workflow.

This is the first in a short series explaining why, and how.

The problem with traditional Salesforce consulting

Salesforce consulting has a productivity ceiling that nobody talks about.

A senior consultant can hold a lot in their head: object models, automation logic, data architecture, governor limits, best practices accumulated over years. But the mechanics of the work — writing SOQL queries, analysing metadata, reviewing field usage, preparing reports, documenting findings — are time-consuming and manual.

Health checks, for example, typically involve logging into an org, navigating Setup, exporting reports, cross-referencing field usage, and manually compiling findings into a document. A thorough Marketing Cloud audit might take seven to eight days. A Sales Cloud review, five to seven.

The expertise is in knowing what to look for and what it means. The bottleneck is the mechanics of getting there.

Where Claude changes the equation

Claude is not a Salesforce tool. It is a general-purpose AI that can read code, query APIs, analyse data, and write structured output. What makes it powerful for Salesforce work is that it can be taught how to interact with Salesforce programmatically — through skills.

A Claude skill is a reusable instruction set that tells the AI how to perform a specific type of work. You define the steps, the tools it can use, the format of its output — and then it executes reliably, every time.

We have been building skills that connect to Salesforce orgs via the Salesforce CLI, run SOQL queries, analyse the results, and produce structured reports. The consultant still drives the analysis. The AI handles the mechanics.

The result is not that we replace expertise with automation. It is that we free up expertise to focus on what actually matters: understanding the business, interpreting findings, and advising on what to do about them.

What this means for our clients

Three things, practically:

Faster time to insight. What used to take a day of manual data gathering can now happen in minutes. We spend less time in Setup screens and more time in workshops with your team.

Deeper analysis. When the mechanics are handled, we can afford to look at things we might not have time for in a traditional engagement — field completion rates across every object, automation dependency chains, permission set overlaps.

More consistent output. A skill produces the same structured report every time. The quality of our deliverables does not depend on which day of the week it is.

What comes next in this series

In the next post, we walk through how we built a specific skill: a Salesforce reporting tool that connects to a live org, runs complex queries, and produces formatted analysis. We will cover the technical approach, the challenges, and what we learned.

If you are a Salesforce professional curious about how AI tools can augment your own delivery, or an organisation wondering how your consultancy could be working smarter — this series is for you. You can also explore our Salesforce consulting services and AI & Data consulting to see how we apply this approach.