What Should You Ask a Salesforce Consultant's References Before Hiring One?
Salesforce AI

What Should You Ask a Salesforce Consultant's References Before Hiring One?

By Troy Amyett•November 1, 2026•5 min read
Back to Insights
Speed
AI-narrated

A reference call helps you check how a Salesforce consultant delivers, handles problems and hands work over. Asking only “Were you happy with them?” rarely gives you enough detail to compare partners. The useful questions are about what actually happened: what changed in scope, what broke, and what the client’s team could do on their own afterwards.

This guide covers the reference call itself. If you’re still building your shortlist, start with the questions to ask the consultant directly.

How do you set up a reference call that tells you something?

Ask each finalist for two references from work similar to yours, finished in the last year or so. Similar matters more than famous: a comparable org, a comparable integration, or the same kind of Agentforce use case.

Then ask to speak with the right people. The executive sponsor will give you the summary. The person who owned the project day to day, and ideally the admin who inherited the work, will tell you how it went.

Keep the call to 30 minutes and use the same questions for every reference, so you can compare answers side by side.

What should you ask about scope and delivery?

  • What was the original goal, and what did you end up with?
  • What changed in scope, and who raised it first?
  • When something slipped, how did you find out?
  • Who from the consultant’s side did the actual building? Were they the people you met in the sales process?
  • If you ran the project again, what would you change in the statement of work?

Listen for how surprises were handled. Every project has them. A good consultant raised them early and explained the trade-offs. A weak one let you find out at the deadline.

What should you ask about data, testing and AI work?

If the work involved an agent, these questions matter most:

  • How was the agent tested before real users touched it?
  • Did they test things going wrong, such as bad inputs, misuse attempts and missing data, or only the happy path?
  • Who reviewed conversations after launch, and what changed as a result?
  • What happened the first time the agent got something wrong?
  • Did they ask about your data quality and permissions before they talked about prompts?

You’re checking whether guardrails and handoffs to a person were part of the design or bolted on at the end. A reference who can describe a specific test, a specific failure and a specific fix is describing a team that does this properly.

What should you ask about handoff and support?

  • What could your team change on its own after go-live?
  • What documentation and training did you get, and do you still use it?
  • How responsive were they after the main project ended?
  • Did you feel dependent on them, and was that by design or by accident?
  • Would you hire them again for the same scope?

That last question is the one to listen to carefully. A quick, specific yes is a good sign. A pause, or “for some things”, deserves a follow-up question.

How do you read what references tell you?

References are chosen by the consultant, so expect them to be positive. You’re not looking for criticism. You’re looking for detail.

Strong signals:

  • They name the people who did the work.
  • They describe a problem and how it was handled.
  • Their story matches what the consultant told you.
  • Their team kept improving the work after handoff.

Weak signals:

  • Only general praise, with no examples.
  • The reference wasn’t close to the work.
  • The project they describe is much smaller or simpler than yours.
  • Their timeline doesn’t match the one the consultant described.

What if a consultant can’t give you a reference?

Some reasons are legitimate. A newer team may not have comparable clients yet, and some clients don’t take reference calls. In that case, ask for something else you can check: a live walkthrough of work they built, with a clear answer on whether it came from a client project or an internal build. A small, paid first phase with a defined deliverable is another way to see how they work before you commit to the whole project.

What you shouldn’t accept is a claim you can’t check in any way.

FAQ

How many references should you call?

Two per finalist is usually enough, as long as at least one comes from work similar to yours.

Who should you speak to at the reference company?

The person who owned the project day to day, and if possible the admin who maintains the work now.

Should you ask references about AI work specifically?

Yes, if your project involves agents. Ask how the agent was tested, who reviews conversations and what happened the first time it got something wrong.

Is it a red flag if a consultant has no references?

Not automatically. Ask for a walkthrough of real work or a small paid first phase instead, and be wary of claims you can’t check.

When should you call references?

After you’ve narrowed the list to two or three finalists, and before you negotiate the contract.


Book an intro call

Funnelists Services

Agentforce & AI Agents

Autonomous agents that predict, automate, and execute revenue-driving actions across sales, service, and marketing.

Troy Amyett

Troy Amyett

Founder & Chief Solutions Architect

9x Salesforce certified. Agentforce Specialist and Agentblazer Legend, 2025–2026. Anthropic-certified in Claude Code and MCP.

Get Insights in your inbox

AI-powered perspectives on Salesforce and Agentforce, delivered weekly.

No spam. Unsubscribe anytime.

Your competitors aren't waiting. Why are you?

Let's talk about what AI agents could do for your business. 30 minutes. No pitch deck. Just answers.

Launch before they do