Record CRUD
Create and update leads, contacts, opportunities, and campaign members, every change shown before it ships, every one recorded in the audit trail.
Salesforce connector
Salesforce MCP Connector – Query Your CRM Without Writing SOQL
Salesforce holds the revenue truth, guarded by SOQL, object models, and an admin team with a queue. Marketing's questions are simple: did the leads convert, what did the campaign touch, where are those opportunities now. Getting them answered has never been simple.
The Salesforce connector changes the cost of asking. It links the AI you already use, like Claude, ChatGPT, or Copilot, to your Salesforce org through Mad Fish Elements. Your AI writes and runs SOQL and SOSL from plain language, describes objects to ground itself in your org's actual schema, performs record operations with your approval, and reaches the full REST surface, composite and bulk included, through passthrough. The org your company runs on becomes something marketing can question directly.
See it
What Your AI Can See
Connect your org, and your AI works the way a careful admin would. It describes objects first, learning your actual fields, including the custom ones that make every org its own dialect. Then it queries: SOQL for the structured questions, across leads, contacts, accounts, opportunities, campaigns, and any custom object your org defines; SOSL when the question is a search across objects.
Marketing's standing questions become sentences. "Leads created from the webinar campaign, and their statuses." "Opportunities influenced by Q3 campaigns, by stage and amount." "Accounts with open opportunities but no activity in 30 days." The answers arrive as visual reports with the query shown, so your admin can verify the translation any time.
And the loop closes on the same Gateway: ad spend from Google and Meta live on one side, the org's pipeline truth on the other, joined in one conversation.
Do it
What Your AI Can Do
With Write access and your approval, your AI operates on records, not just reads them:
Campaign hygiene
Add the missing campaign members, fix the lead sources that break attribution, and retire the typo-duplicated campaigns, in approved batches.
Scale operations
Composite requests bundle related changes; Bulk 2.0 handles the big jobs; both route through passthrough under the same approval flow.
Schema-grounded safety
Describe-before-write is the habit: your AI checks what fields exist and what they accept before proposing a change, which is how orgs stay unbroken.
A working session
The Lead-Source Cleanup That Finally Happened
In practiceThe session admins recognize: attribution reports have been wrong for months because a third of last quarter's leads carry blank or junk lead sources. You ask, "Leads created last quarter with blank or invalid lead source, grouped by what we can infer from their campaign membership." Your AI describes the objects, runs the queries, and returns the breakdown: 412 leads, most inferable from campaign membership. You say, "Propose the fix." It shows the batch: source values by inference rule, exceptions flagged for human review. You approve the clean 350, review the 62 exceptions yourself. The attribution reports stop lying, and it took a conversation.
The return
What You Get Back
Answers without the admin queue
Plain-language questions against the org, queries shown, results visual.
Attribution that holds
Campaign membership and lead sources kept clean, so the revenue story survives scrutiny.
Spend beside pipeline
Ad platforms live on one side, org truth on the other, one conversation.
An admin team reviewing, not transcribing
Queries and batches arrive ready for verification instead of built from scratch.
Use cases
One Conversation, Real Account Work
"Show opportunities influenced by Q3 campaigns, by stage and amount."Salesforce
The campaign-influence answer as a visual report, SOQL shown.
"Which leads from the webinar have not been touched in two weeks, and who owns them?"Salesforce
The follow-up gap, ranked, with owners.
"Describe the Opportunity object so I know what fields we actually use."Salesforce
Your org's real schema, custom fields included, in plain language.
"Add these 40 webinar attendees as campaign members, matched by email."Salesforce
The batch shown, exceptions flagged, shipped on your approval.
"Put Q3 ad spend next to the pipeline those campaigns influenced."Salesforce · Google Ads · Meta
The budget-defense report, live on both sides.
For in-house teams
Close the expectation gap
Marketing gets its standing questions answered from the same org sales trusts, without becoming SOQL literate, and the admin team's queue drains of everything that never needed a ticket.
For marketing teamsFor agencies
Run the whole book
Client orgs hold the proof of performance. Query them under per-client permission walls, keep campaign hygiene as a deliverable, and report spend-to-pipeline in numbers the client's own RevOps team can verify query by query.
For agenciesReports worth sharing
From Live Account to Clear Story
Ask in chat and share the dashboard without building exports or taking screenshots.
Campaign Influence Board
Opportunities and amounts by campaign and stage. Built for proving marketing's number.
Lead Flow Report
Leads by source and status, follow-up gaps flagged. Built for the marketing-sales handoff meeting.
Org Hygiene Summary
Blank sources, orphaned members, stale records. Built for the monthly cleanup.
The charts do the talking: funnel for stage flow, bar for campaign comparison, table-first for record detail with queries attached. And when your leaders or clients want these anytime, the Data Visualization add-on turns your reports into always-current dashboards they open from a link. Ask in chat. Share the dashboard.
The useful signals
The Numbers That Matter
The org answers best on a few figures. Leads by source and status: the funnel's top, with the handoff visible. Campaign-influenced pipeline: opportunities and amounts the marketing calendar can claim. Stage conversion: where deals advance and where they stall. Time in stage: the momentum reading behind the pipeline snapshot. And the hygiene pair: blank lead sources and orphaned campaign members, the two data faults that quietly break every attribution report downstream. Read pipeline weekly and hygiene monthly. Fix the faults in small approved batches. The org's answers are only as good as these fields, which is why this connector treats cleanup as a first-class workflow and not a favor.
A practical start
Your First Week
Day one: connect the org and get grounded. "Describe the Opportunity object." Learn your own fields, custom ones included; day two: ask a real question. "Opportunities influenced by last quarter's campaigns, by stage." Read the SOQL it wrote. Day three: have your admin verify that query. This is the step that turns RevOps from skeptic to sponsor; day four: run one hygiene pass. "Leads with blank sources last quarter, with inferable fixes." Approve the clean batch; review the exceptions yourself. Day five: schedule the standing answer. "Monthly, run the campaign influence board." By Friday, marketing has answers from the org without a ticket, the admin team has reviewed instead of transcribed, and one attribution-breaking data problem is smaller than it was. The org did not get less guarded; it got more askable.
Works best with
Build the Connected View
Pair Salesforce with Google Ads and Meta for the spend-beside-pipeline report that defends budgets. Add Mautic or Mailchimp where nurture runs outside the org, so the lead journey reads end to end. Revenue truth plus channel truth, one conversation.
Clear limits
What This Connector Will Not Do
It will not bypass your org's security model; it works through the connection your tenant authorizes, with your org's profiles and permissions intact, and Elements' per-person permissions on top. It will not write blind; describe-first and show-before-ship are the habit, and bulk jobs are approved as batches. It will not guess at your schema; when a field's meaning is unclear, it says so. And it will not touch orgs you did not connect.
Governance
Governed Access Your Whole Org Can Live With
Access is set per person, not per account, so Read seats see everything and change nothing while Write seats create and edit with your approval flow in front of every change. Delete access goes only to people you trust to clean up, and many teams start read-only, check answers against the platform, then widen access as the audit trail earns it.
You choose the level of human review and can change it at any time, while every action records what changed, when, and who approved it. Per-user access means Elements never stores your Salesforce password, and your data is never harvested or resold.
Per-user accessRead / write / delete per seatFull audit trailApproval flow on every change
Setup
Setup Takes Minutes
- Start your trialCreate your Elements account and connect it to your AI.
- Link SalesforceSign in through the secure connection screen and choose what to connect. No code and no engineers.
- Set permissionsDecide who can Read, who can Write, and who can Delete. Start conservative and widen as trust builds.
- Ask your first question"How did we do last month?" is a complete instruction.
Plans, account limits, and the multi-client Agency plan are on the pricing page .
FAQ
Salesforce Connector Questions
Do we need to know SOQL?
No; ask in plain language; the AI writes the query and shows it. Your admin can verify every one, which is how trust gets built with RevOps rather than around them.
Does it handle our custom objects?
Yes. Describe works on any object, queries reach custom objects and fields, and the REST passthrough covers the rest of the API surface, composite and Bulk 2.0 included.
What about governor limits and API usage?
Calls run through your org's normal API limits, visible to your admins like any integration. The connector favors bounded, schema-aware queries, and heavy jobs go through Bulk 2.0 the way Salesforce intends.
Will my AI change things without asking me?
No. Your AI acts only with the permissions you set, and changes follow the review level you choose. Read-only seats cannot change anything, because the approval flow is the product working as designed.
Which AI assistants can our team use?
Your team can use Claude, ChatGPT, and Microsoft Copilot, while other AI tools can likely connect when they support MCP, the open standard for linking AI assistants to outside tools.
Is my data used to train anything?
No. Your data stays yours, so nothing is harvested and nothing is resold.
Focus on the Strategy, Delegate the Launch
The teams winning with AI are not handing their accounts to a black box; they keep the strategy and delegate the hands-on-keyboard work, while approvals and an audit trail make it safe to move fast. Connect the account you already have through the Gateway and ask the question you would normally build a report to answer.
Your first 14 days are free. Add your card to start, and cancel before the trial ends to pay nothing.