You Already Have the Engine. Here is What Activates It.

Written by Steny Sebastian | 9 July 2026 11:30:03 PM

How IBM Planning Analytics (PA) + Bob (IBM PA Agent) + IBM watsonx Orchestrate (WXO) unlock the two-track model - and give your team back 150 minutes a day


If your team runs IBM Planning Analytics, read this. The planning engine is already there. The governed model is already there. The data your CFO needs to make faster decisions is already there. What is missing is the layer that runs it autonomously - so your analysts stop losing 150 minutes a day to work the system should already be doing.

In June 2026, IBM shipped the IBM Planning Analytics Agent - known internally as Bob. The diagram shared widely this month shows what it actually is: seven specialist finance agents, six MCP tool layers, one orchestration hub, all sitting on top of your existing Planning Analytics environment.

This is not a new platform. It is not a migration. It is the intelligence layer that activates what you already have.

IBM PA Agent MCP tool layers - the six categories of direct access into your PA model:

Planning Analytics Data Access

Scenario Simulation

Forecast Calculations

Reporting Outputs

Workflow Orchestration

External Business Context

Wire this into IBM watsonx Orchestrate (WXO) with OAuth-secured connectivity and you have a governed, auditable, AI-driven finance operation running at enterprise scale - on your existing IBM Planning Analytics investment.

The two-track model: what PA + Bob + WXO runs, and what your people do best

The reason PA + Bob + WXO produces real savings - not pilot-scale improvements - is that it operates on a two-track model. AI and humans are not competing for the same work. They are running parallel tracks simultaneously.

What separates the two tracks is not intelligence. It is where each type of work actually belongs.

HUMAN COGNITIVE TRACK	AI
1. Critical Thinking	Plans, schedules and initiates the finance workflow
2. Active Learning	Ingests and Validates - collects, checks quality, validates completeness against the PA model
3. Quality Control	Processes and Analyses - applies rules, AI models and logic across planning data
4. Judgement and Decision Making	Reconciles and Aligns - matches figures across systems and entities within PA
5. Problem Solving	Verifies and Controls - applies controls, audit checks and ensures compliance

The outcome of running both tracks simultaneously: faster close cycles, higher accuracy, lower effort on the AI side - and better decisions, smarter strategy, and stronger business outcomes on the human side.

This is what 1.6 FTEs of untapped capacity - currently consumed by process overhead inside your PA environment - looks like when it is released back to the team.

What the two-track model delivers: the numbers

These figures come from live discovery workshops and Octane's implementation experience running the two-track model on IBM Planning Analytics environments. Not theoretical. Not vendor estimates.

150

minutes saved
Per analyst, per day

$38,250

per year
Value per analyst (AUS)

$191,250

per year
Value across a team of 5

Based on an average fully loaded cost of AU$68 per hour and 225 working days per year.

Across all knowledge workers in your finance function:

Analysts, FP&A professionals, controllers - we benchmark the value at approximately $7,800 per person per year as a conservative cross-function starting point.

For accounting-specific roles running IBM Planning Analytics, the documented figure is significantly higher.

A day in the life of a finance analyst using AI in Finance

Here is what the two-track model looks like, hour by hour, for an analyst whose team runs IBM Planning Analytics with Bob and WXO deployed:

Time	Activity	What Bob + WXO does	Time saved	$ Value/day	Human focus
8:30 am	Review JE agent overnight report	Anomaly detection and flagging inside PA	~20 min	~$23	Policy judgement
10:30 am	Clear anomalies and unprocessed items	Root-cause and exception triage	~30 min	~$34	Accounting treatment
12:30 pm	Validate journals and calculations	Journal recommendation and approval routing	~30 min	~$34	High-value JE review
2:30 pm	Coordinate close and consolidation	WXO workflow orchestration, data aggregation	~25 min	~$28	Escalation decisions
4:30 pm	Review trial balance commentary	Variance review and MDA commentary drafting	~25 min	~$28	Narrative refinement
5:30 pm	Advise stakeholders, improve controls	Follow-up action, continuous improvement	~20 min	~$23	Strategic guidance

Total: approximately 150 minutes saved per analyst per day. Total daily value: approximately $170 per analyst.

That time does not disappear. It gets reinvested into the human cognitive track - control enhancement, proactive issue resolution, business partnering.

PA + Bob + WXO: the three components your team needs

The two-track model runs on a specific three-part architecture. Each component has a distinct role. Together, they replace the manual hand-offs, data exports and reconciliation steps that consume the 150 minutes.

01 Planning Engine

PA - Rigour

02 AI Co-builder

Bob - Velocity

03 Orchestrator

WXO - Governance

IBM Planning Analytics (TM1)

IBM Planning Analytics Agent (via Bob)

IBM watsonx Orchestrate

The governed financial model your team already runs. All planning data, scenarios, forecasts and actuals live here. TM1 is the source of truth - the structure that ensures every number means the same thing across the organisation.

Seven specialist agents. Thirty-plus MCP tools connected directly into your PA model. Analysts describe what they need in plain English. Bob builds the structure inside the governed model, runs the scenario, and surfaces the result - in an afternoon, not three weeks.

The binding layer. WXO sequences the agents, enforces access controls, maintains audit trails and ensures every AI-driven action is traceable and accountable. PA + Bob produces the speed. WXO makes it enterprise-grade.

What this means at team level

Individual savings compound fast when you apply the two-track model across a finance team.

For a team of 5 finance analysts on IBM Planning Analytics:

~750 minutes freed per day (12.5 hours)

~$850 in value created per day

~$191,250 in annual value (at AU$68/hour, 225 working days)

These are recoverable savings inside your existing PA investment - no new platform, no data migration, no heavy lift.

The 45-minute offer - for IBM Planning Analytics teams

If your team runs IBM Planning Analytics, give me 45 minutes. I will show you exactly where the 150 minutes is hiding in your current process - and run a live demonstration activating the two-track model on a real PA environment.
Not a slide deck about what it could do. A working system doing it.

In 45 minutes you will see:

Bob (IBM PA Agent) running natural language queries against a live TM1 environment - real model, real results WXO (IBM watsonx Orchestrate) governing the workflow - audit trails, access controls, all visible
The two-track model in operation: month-end close, variance analysis, scenario modelling, board commentary
The $7,800 per knowledge worker benchmark mapped to your team size and existing PA environment
Clear answers on data residency, sovereignty and compliance - no vague AI governance promises
A straight breakdown of managed services, SLAs and what ongoing looks like - no surprises

No heavy lift. No replacing your existing PA investment. No new data model.

Before we speak - two questions

These will shape the 45 minutes. Rough answers are fine. We will sharpen them together.

Question 1:
How much time does your finance team spend each month on manual reporting, reconciliations and month-end close inside your IBM Planning Analytics environment?

Question 2:
Have you quantified what activating Bob and WXO on your existing PA investment could save your organisation in both time and dollars?

Join the roundtable - IBM Planning Analytics in the room

I run invite-only AI Roundtable Lunches for CFOs and senior finance leaders across Australia and Asia. Each session includes a live demonstration running real finance workflows on a live IBM Planning Analytics environment. No slides, no pitches. Thirty senior finance leaders per city.

If you are on IBM Planning Analytics and want to see the two-track model running before you commit to anything, this is the room.

City	Date	Format	Register
Perth	28 July	30 senior finance leaders - live PA + Bob + WXO demo	octanesolutions.com.au/perth-roundtable-ai-lunch-summit-2
Melbourne	30 July	30 senior finance leaders - live PA + Bob + WXO demo	octanesolutions.com.au/melbourne-roundtable-ai-lunch-summit
Sydney	11 August	30 senior finance leaders - live PA + Bob + WXO demo	octanesolutions.com.au/sydney-roundtable-ai-lunch-summit-1
Mumbai	August (TBC)	Invite-only CFO Lunch - live demo	DM Steny Sebastian on LinkedIn
Auckland	October (TBC)	Invite-only CFO Lunch - live demo	DM Steny Sebastian on LinkedIn

View full post