Case study · demo org

What one scan found in a Salesforce demo org

15 findings: 10 problems that would cost $14,241 over three years if left alone, and 5 things the org hasn’t built yet. The two urgent ones take about 4.5 hours to fix.

By VanguardOps, written by a Salesforce architect with 13+ years in enterprise orgs. Updated November 18, 2026.

Health score
52/100
Health score
Grade F: needs cleanup
3-year cost if nothing changes
$14,241
3-year cost if nothing changes
From the 10 problems
Agentforce readiness
37/100
Agentforce readiness
Not ready yet: 1 blocker
To fix the 2 urgent issues
$530
To fix the 2 urgent issues
About 4.5 hours of work
VanguardOps dashboard for the demo org: $14,241 three-year cost, 15 findings, health score 52, five fixes waiting for approval, and the next step to take
The app's dashboard right after the scan. Top: the 3-year cost, the number of findings and the health score. Below: fixes waiting for approval, and the issue to start with. Only the org's name is blurred.

The org we scanned

A free Salesforce test org (Developer Edition) that VanguardOps owns, with Service Cloud features turned on and Salesforce’s Agentforce permission sets installed. It holds no customer data. The scan on November 18, 2026 read how the org is built, never the records in it.

Fix first: two urgent issues, $530

  • Report exports aren’t watched. A security policy that would log or block big report exports was never switched on. Left alone: $5,712 over three years. To fix: $262, about 1.5 hours.
  • 23 fields don’t say what they’re for. People, and AI agents, have to guess. Left alone: $1,896. To fix: $268, about 3 hours.
A suggested fix in VanguardOps: what changes, why, the risk, $5,712 saved over 3 years, $262 to fix, about 1.5 hours for an architect
The suggested fix for the export policy: what changes, why, the risk, what it saves and what it costs. Nothing changes until someone approves it.

All 10 problems, priced

Risk is how much harm a problem can do. Cost is what it adds up to over three years if nobody fixes it. The two don’t always line up. Together these ten come to $14,241.

Problems the scan priced, costliest first
Report exports aren't watched: the export security policy was never switched onRisk: High3-year cost: $5,712
An outside app can sign in from any network (connected app with no IP limits)Risk: Medium3-year cost: $2,754
23 fields don't say what they're for (no description)Risk: Critical3-year cost: $1,896
22 permission sets that nobody is assigned toRisk: Medium3-year cost: $1,485
Omni-Channel is on, but nothing tells it where to send casesRisk: High3-year cost: $807
An outside app with no note on what it's for (connected app, no description)Risk: Medium3-year cost: $552
Two case assignment rules with the same name, “Standard”Risk: Medium3-year cost: $402
24 permission sets that don't say what they're forRisk: Medium3-year cost: $309
One flow named outside the org's naming patternRisk: Medium3-year cost: $162
One case record type used for every kind of caseRisk: Low3-year cost: $162

The other 5 findings are things the org hasn’t built yet, like case queues and Knowledge. Those are projects, not problems, so they add nothing to the cost.

The fix plan: who, how long, how much

All 15 items in three phases, urgent first. Each row shows who does it, the hours, the cost to fix and the cost of leaving it.

VanguardOps fix plan: 15 fixes in three phases, with who does each one, the hours, the fix cost and the 3-year cost if left
Rows with $0 in the last column are new things to build, not problems. The plan downloads as a spreadsheet.

Agentforce readiness: 37 out of 100

Agentforce is Salesforce’s AI agent. This score shows how ready the org is to run one, and the answer is not yet. One blocker stands in the way: the 23 fields with no description, which an agent needs to know what each field is for.

Agentforce readiness in VanguardOps: 37 out of 100, not ready, 1 blocker, 4 readiness findings, and a score for each category
Each area gets its own score. N/A means the scan doesn't check that area yet, and it never counts as a pass.

How a fix gets made

  1. Draft. VanguardOps writes the exact change, and you can read it first.
  2. Check. Salesforce test-runs it without saving anything.
  3. Approve. An admin approves that exact change.
  4. Apply. It goes into a sandbox, a test copy of the org, never straight to production.
  5. Undo. One click takes it back out.

What this case study doesn’t show

  • A demo org is small and young, so its findings and dollar figures are smaller than a production org’s.
  • Dollar figures are estimates from the published cost model, not quotes.
  • No fix from this report has been applied to this org yet.
What does the health score mean?
It sums up the shape of the org's setup from 0 to 100. 90 or more is an A, 80 a B, 70 a C, 55 a D, and under 55 an F. It grades this org on its own, not against other orgs.
Why does the dashboard say to fix the fields first?
The dashboard points to the highest-risk item, and the 23 fields are rated critical. The export policy costs the most if left alone. Both sit in the first phase of the fix plan.
Is this a customer's org?
No. It is a free Salesforce Developer Edition org that VanguardOps owns and uses for its own work. No customer data is in it, and the scan read only metadata.
Are the dollar figures real?
They are the scan's own estimates for this org, copied from its report without changes. They come from the published cost model and are scaled to the org's size, so a small demo org prices low. They are estimates, not quotes.
Why are some findings priced at zero?
Something the org has never built, like Knowledge or case queues, is a project to take on, not debt it carries. The report lists those as new builds and leaves them out of the total.
Will my org look like this?
Probably not. This org is small and lightly configured. The same checks run on any org, and an org with more fields, access and automation gives them more to find.