Forecasting Illuminate agent cost means turning agent skills and action volume into a monthly Flex Credit burn. This guide gives a step by step model a buyer can run during the complimentary window and take into a renewal.
Key takeaways
- Illuminate agents meter through the same Flex Credit rate card as every other agent.
- Cost is set by the skill mix, not the agent count.
- Forecast burn as actions per skill times credits per skill.
- Retrieval skills draw near 1 credit, autonomous completions near 5.
- Use the complimentary window to capture the real action volume.
- Size the paid tier against observed burn, not a vendor estimate.
How do you build an Illuminate cost forecast?
You build the forecast by mapping each Illuminate agent to its skills, estimating monthly actions per skill, and multiplying by the credit value from the Flex Credits rate card. Workday expanded Illuminate across HR, Finance, and industry in 2025, so the agent list can be long.
Step one. Inventory agents and skills
List every active Illuminate agent and the metered skills it runs. Use the Agent System of Record as the source so nothing is missed.
Group agents by function so the forecast maps to budget owners. HR, Finance, and industry agents each carry their own skill profile, and a shared inventory stops duplicate agents from being double counted in the burn.
Step two. Estimate monthly actions
Estimate actions per skill from real usage, separating retrieval from autonomous completion. The Workday AI agents catalog shows which skills each agent exposes.
Pull volume from a full month, not a busy week, so seasonality does not distort the base. Where usage is still thin, hold the number conservative and let the heavy case carry the upside rather than inflating the base.
Step three. Multiply and total
Multiply actions by the per skill credit value and sum across agents. That total is the monthly credit burn.
Keep the blended credits per action visible as a single figure. It is the fastest sanity check at the table, because a blended rate drifting above 2 signals an autonomy heavy estate that needs more headroom.
Illuminate monthly burn forecast, worked example
| Illuminate agent and skill | Monthly actions | Credits each | Monthly credits |
|---|---|---|---|
| HR Self Service, retrieval | 35,000 | 1 | 35,000 |
| HR Self Service, guided draft | 6,000 | 2 | 12,000 |
| Finance Agent, autonomous | 5,000 | 5 | 25,000 |
| Total | 46,000 | Blended 1.57 | 72,000 |
Is Illuminate a separate SKU or part of Flex Credits?
Illuminate is not a separate SKU. Its agents draw on the same universal Flex Credit rate card as every other Workday agent, so there is no standalone Illuminate price line to negotiate. You negotiate the credit pool and the rate card behind it.
When does the meter actually run?
Workday meters an agent when a task is complete, not per prompt or token. A retrieval skill draws about 1 credit per action and an autonomous completion draws about 5, per the Flex Credits rate card. Mid range skills sit between.
How do complimentary credits fit in?
Every subscription includes an annual allotment of complimentary Flex Credits sized to your company and renewed each contract year. You only pay once production usage exceeds that allotment. Experimentation outside production draws no credits, which is why the complimentary window is safe to load test.
Flex Credit rate card by skill type
| Skill type | Example Illuminate agent | Credits per action | Meter trigger |
|---|---|---|---|
| Retrieval, self service | Case Agent, Business Process Copilot | 1 | Answer returned |
| Guided draft | Employee Sentiment, Job Architecture | 2 to 3, illustrative | Draft produced |
| Autonomous completion | Financial Close, Cost and Profitability | 5 | Task completed end to end |
Workday publishes 1 credit for retrieval and 5 for autonomous completion. Mid range values are illustrative, so confirm your own rate card.
What drives the size of the forecast?
The size of the forecast is driven by the share of autonomous actions, because they draw five times a retrieval action. A small shift toward autonomy moves the total sharply.
The autonomy share is the swing factor
Hold actions flat and raise the autonomous share, and the burn climbs fast. This is why the mix matters more than the count.
The arithmetic is plain. Move 5,000 actions from retrieval to autonomous and burn rises by 20,000 credits a month, because each of those actions now costs 5 instead of 1. Volume held flat, the total still climbs.
Adoption changes the mix over time
As users trust the agents, they hand over more complete tasks. Model that drift, or the forecast ages badly.
Adoption rarely stays where it launches. Users start with questions, then let the agent draft, then let it complete. Model that shift as a rising autonomous share across the year, or the forecast reads low by the second quarter.
Where the common advice on Illuminate cost is wrong
The common advice is to forecast AI cost from the number of agents or the number of users. We disagree. In the forecasts we built, agent count barely predicted burn, while the skill mix predicted it closely, and the autonomous share moved the total more than any other input. The buyer side move is to forecast from actions per skill, run a heavy autonomy case as a sensitivity, and size the paid tier so it absorbs adoption rather than the launch month alone. A forecast built on agent count is a forecast that breaks the first time usage matures.
Source: Redress Compliance advisory engagement file, 2024 to 2025.
Agent count tells you almost nothing about cost. The mix of skills, and how fast it tilts toward autonomy, tells you almost everything.
What do the three forecast scenarios show?
The three scenarios show how the same action volume produces very different burn as the autonomous share rises. Holding total monthly actions at 46,000, annual burn moves from roughly 696,000 credits to 1,128,000 as autonomy climbs. The mix, not the count, drives the spread.
What assumptions sit behind the numbers?
Each scenario fixes total monthly actions at 46,000 and varies only the split across retrieval, guided draft, and autonomous skills. Retrieval draws 1 credit, guided draft 2, and autonomous 5. These values are illustrative anchors, not a quote, so replace them with your own rate card.
Three scenario burn forecast, 46,000 monthly actions held constant
| Scenario | Autonomous share | Blended credits per action | Monthly credit burn | Annual credit burn |
|---|---|---|---|---|
| Conservative | 4% | 1.26 | 58,000 | 696,000 |
| Base | 11% | 1.57 | 72,000 | 864,000 |
| Aggressive | 22% | 2.04 | 94,000 | 1,128,000 |
Which line should you size against?
The annual burn row is the number to size against. A launch month understates the total because adoption lifts the autonomous share over the year. Fund the scenario your estate is drifting toward, not the one it starts in.
How do you use the forecast at renewal?
You use the forecast to size the paid tier against observed burn and to negotiate the rate card from evidence. The number that matters is the gap between projected burn and the complimentary allotment.
Size the gap, not the launch month
Compare annual projected burn against the allotment and fund the gap, using the complimentary allotment described on the Workday AI Flex Credits page. Do not extrapolate from a quiet first month.
Set annual projected burn beside the complimentary allotment and read the gap. That gap, not the gross burn, is what you fund and what you negotiate. A generous allotment can absorb a surprising amount of production usage.
Bring the heavy case
Run a second forecast with autonomy dialed up and take both to the table. A tier that only fits the light case is a tier that fails on adoption.
Take both cases into the room. Anchor on the base case for the commercial ask, and use the heavy case to justify headroom in the credit pool. Sizing to the light case alone invites a mid term true up.
- Base case: current skill mix and action volume.
- Heavy case: autonomous share raised to test headroom.
- Allotment line: the complimentary credits you do not pay for.
Suggested reading
- Workday Flex Credits pillar. The full pricing and levers pillar.
- Flex Credits explained. The cost model with a worked example.
- Workday Illuminate AI pricing guide. The wider Illuminate pricing picture.
Which governance guardrails hold agent burn down?
The guardrails that hold burn down live in the Agent System of Record, where every agent is registered, configured, activated, and deactivated. The Agent System of Record and its Agent Gateway meter all agent interactions, including third party agents, from one control point.
What are the four controls that matter?
- Registration. No agent runs unregistered, so there is no unmetered burn hiding in the estate.
- Observability. Usage detail per agent and skill supports chargeback and early drift alerts.
- Deactivation. Retire idle agents before they quietly accrue credits against the pool.
- Gateway metering. The Agent Gateway meters third party agents on the same rate card.
Which contract protections should you demand?
You should demand a locked rate card, a written complimentary allotment, credit rollover, an overage price cap, and metering transparency. Each closes a gap where cost can drift after signature. The rate card lock matters most, because credit values per skill are the variable that moves burn.
Contract protections to demand before signing
| Protection | Why it matters | What to ask for |
|---|---|---|
| Rate card lock | Credit values per skill can change | Fixed credit values for the term |
| Allotment floor | Complimentary credits scale to size | Written allotment, renewed annually |
| Credit rollover | Unused credits may expire | Rollover or true down of the pool |
| Overage cap | Extra credits bought at list | Pre agreed unit price for top ups |
| Metering transparency | Burn is metered on completion | Access to Agent Gateway usage detail |
Put these terms in the order form, not the sales deck. A rate card shown on a slide is not a contractual commitment, and credit values that live only in documentation can move at the next release without notice.
What should a buyer do next?
What to do next is a short ordered checklist: inventory the agents, measure real actions, model three scenarios, and size the tier before you sign.
- Inventory every Illuminate agent and its skills from the Agent System of Record.
- Capture monthly actions per skill during the complimentary window.
- Build the base case burn forecast.
- Run a heavy autonomy case as a sensitivity.
- Size the paid tier against the gap to the allotment.
- Engage independent Workday advisory before you sign.
Frequently asked questions
How do you forecast Illuminate agent costs?
You forecast Illuminate agent costs by mapping each agent to the skills it runs, multiplying expected monthly actions by the per skill credit value, and totaling the credit draw against the complimentary allotment. The complimentary window supplies the usage data.
Why do Illuminate agents vary so much in cost?
Illuminate agents vary in cost because they run different skills, and each skill has its own credit value. An agent that mostly retrieves information draws near 1 credit per action, while one that completes tasks autonomously draws near 5.
What data do you need to build the forecast?
You need the list of active agents, the skills each one runs, and the monthly action volume per skill from real usage. The Agent System of Record supplies the agent and skill inventory, and the complimentary window supplies the volume.
How far ahead should the forecast run?
The forecast should run at least two quarters ahead of renewal so the complimentary window can supply real usage data. A late forecast leaves you sizing the paid tier on a vendor estimate rather than your own numbers.
Is Illuminate a separate SKU from Flex Credits?
No, Illuminate is not a separate SKU. Its agents meter through the same universal Flex Credit rate card as every other Workday agent, so you negotiate the credit pool and rate card rather than an Illuminate price line.
When does an Illuminate agent consume a credit?
An Illuminate agent consumes credits when a task is complete, not per prompt or token. Metering is outcome based, so an experimental action run outside production does not draw from your credit pool.
How many credits does an autonomous action cost?
An autonomous completion draws about 5 credits per action, while a retrieval action draws about 1, per Workday's published rate card anchors. Mid range skills sit between, which is why the skill mix sets the blended rate.
What are complimentary Flex Credits?
Complimentary Flex Credits are an annual allotment included in your subscription and sized to your company. They renew each contract year and let you run agents in production to gather usage data before paying for more.
How do you model three forecast scenarios?
You model three scenarios by fixing total actions and varying the autonomous share across conservative, base, and aggressive splits. Each split produces a different blended rate and annual burn, which brackets the tier you need to fund.
Which contract protections matter most for agent credits?
The rate card lock matters most, because credit values per skill drive burn. Pair it with a written complimentary allotment, credit rollover, an overage price cap, and access to Agent Gateway metering detail.