Stage 7: Measure & Iterate — Campaign Ops Optimisation

7.1 The Key Question

Are Campaigns Moving Faster and More Predictably?

That's the whole point of this project. Before, campaigns took anywhere from 45 to 120 days with no way to predict which end of that range you'd land on. If the PoC works, I should see cycle times come down and become more consistent.

I'm not setting hard targets before the pilot — I want to baseline first with real data, then set goals based on what I actually observe. But I do know what questions to ask.

7.2 What I'll Track

Phase 1 Metrics

These are the things I can measure once the pilot is running:

What I'm Measuring	Why It Matters
Campaign cycle time	The big one. Are campaigns getting from brief to live faster than before?
Time ops spends on admin	Are they spending less time copying data and chasing updates?
SLA accuracy	Are the auto-calculated SLAs realistic? Are we getting false breaches?
Auto-assignment accuracy	Is the routing Flow putting campaigns with the right builders?
Workload balance	Are builders getting a fair share of work, weighted by complexity?
System adoption	Are people actually using Salesforce, or are they still going to email and Excel?
QA checklist usage	Are teams using the Salesforce QA checklist instead of the old Excel one?

Honest note: I'll only know real targets after the pilot runs for a few weeks. Right now, there's no baseline data because everything was tracked informally before. Setting up the system is what creates the ability to measure in the first place.

7.3 How I'll Learn

Feedback Loops

Metrics on their own don't tell me much — I need to talk to the people using the system to understand what the numbers mean.

Weekly Check-in with Ops

What worked this week? What didn't?
Did any auto-assignment feel wrong? Why?
Are the SLAs realistic for the types of campaigns coming through?
Any steps that still feel manual and shouldn't be?

Monthly Review

Look at the data: cycle time trends, SLA hit rates, workload distribution
Adjust the complexity scoring weights if the formula isn't reflecting real effort
Tune the SLA day mappings based on how long things are actually taking
Fix any Flow logic issues that came up

After Each Iteration

What did I change and why?
What did I expect to happen?
What actually happened?
Keep, revert, or adjust?

This sounds basic, but most process improvements fail because nobody checks whether they actually worked. I want to build the habit of measuring before and after every change.

7.4 What Comes Next

Iteration Roadmap

Short-Term (During Pilot)

Refine complexity scoring weights based on real campaign data
Adjust SLA targets based on actual turnaround times
Improve validation rules based on common submission errors
Tune dashboards and list views based on what ops actually uses

After Pilot Success

Roll out to remaining business units
Build better reporting with more data
Identify which parts of the process would benefit most from Agentforce
Start scoping Phase 2 based on what we've learned

Phase 2 (If Phase 1 Proves Itself)

Explore Agentforce for smarter routing — not just lowest workload, but matching builder skills to campaign type
Automated intake validation — catching incomplete briefs before they enter the pipeline
Better notifications — proactive updates to BUs instead of them having to check

Key principle: Phase 2 only makes sense if Phase 1 data proves the process works. Agentforce needs clean, structured data to be useful — that's exactly what Phase 1 creates.

7.5 Keeping the System Healthy

Monitoring the Flows

The three Flows are the backbone of the system. If they stop working properly, everything breaks. Here's what I'll watch:

Flow	What Could Go Wrong	How I'd Catch It
Briefing – Readiness & SLA	Accepting incomplete briefs, or calculating wrong SLA dates	Spot-check new campaigns weekly. Compare calculated SLA vs what feels right.
Assignment Routing	Overloading one builder, or assigning complex campaigns to the wrong person	Review workload distribution weekly. Ask builders if assignments feel fair.
SLA Pause/Resume	Not pausing when it should (making ops look slow for BU-caused delays)	Check paused campaigns. Verify the pause reason matches what actually happened.

If something's off, I'll adjust the Flow logic, test the fix, and deploy in the following week. The whole point of using Flows (not code) is that these adjustments are quick and don't need a developer.

7.6 Summary

Where This Stage Stands

The measurement approach is defined. The metrics are clear. The feedback loops are planned. What's missing is real data — and that comes from running the pilot.

What's Ready

Key metrics identified (cycle time, admin time, SLA accuracy, adoption)
Feedback loops planned (weekly check-ins, monthly reviews)
Flow monitoring approach defined
Iteration process: change, measure, keep or revert
Phase 2 criteria: only proceed if Phase 1 data supports it

Status: Framework defined. Real measurement starts when the pilot goes live.