AI TCO Worksheet: The Seven-Sheet Model

Seven-sheet TCO worksheet map with a panel showing naive cost 0.25 versus fully loaded 3.03 per ticket, a 12.3 times gap — The seven sheets, and the three numbers a CFO can repeat.

Key takeaways

Most AI business cases price only the model invoice, missing 60-80% of the true operating cost in integration, governance, review time, and infrastructure.
The seven-sheet TCO worksheet maps directly to the canonical seven-layer AI TCO Framework used throughout this site.
A finance analyst and platform engineer can complete the worksheet together in an afternoon per use case, producing three numbers a CFO can repeat: fully loaded unit cost, naive-versus-loaded gap, and attribution status.
The pilot-to-production bridge sheet addresses the most common business case failure: assuming pilot economics scale linearly to production.

Why this matters

In 2026, Marc Benioff said Salesforce expected to spend about $300 million on Anthropic tokens in 2026, largely for coding-related work. Interpretation The spend was fragmented across departments, embedded in SaaS contracts, and categorised inconsistently across finance systems. Interpretation This pattern repeats across enterprises: AI costs accumulate faster than visibility systems can track them. Interpretation Most business cases price only the visible model API invoice, missing the larger costs in human review, integration work, governance, and infrastructure. Interpretation

The hidden cost multiplier: At Uber, the visible API invoice for AI coding tools ran $500-$2,000 per engineer per month. The hidden costs - review time of engineers checking agent-generated code, rework when agent-committed code needs correction, orchestration engineering, opportunity cost of attention shifting to prompt-wrangling - never appeared on the Anthropic invoice but belong in cost per unit of shipped work.

The seven-sheet TCO worksheet provides a structured approach to capturing full-stack AI costs before they become budget surprises. Interpretation It is designed as a working tool, not a framework diagram - something a finance analyst and platform engineer can fill in together in an afternoon per use case. Interpretation

Download the worksheet

A blank Excel version of the seven sheets is available: AI TCO worksheet, version 1.0, 6 October 2026 (.xlsx). It follows the structure below, contains no example data, and calculates the unit economics, bridge multiple, variance status and net position. Input cells are shaded cream.

The seven-sheet structure

The worksheet maps directly to the canonical seven-layer AI TCO Framework: infrastructure; data and context; models; integration and workflow redesign; people and capability; governance, safety, and compliance; operations and portfolio oversight. Interpretation Each sheet captures a distinct cost layer, with Sheet 3 dedicated to the costs business cases routinely forget. Interpretation

Sheet 1: Use case header

Every use case needs a named owner, a clear workflow description, and an explicit stage designation. Interpretation The stage designation - pilot, scaling, or production - determines which cost assumptions apply and which governance requirements trigger. Interpretation

Field	Notes
Use case name and owner	One named person accountable for economics
Workflow description	What the AI does, who consumes the output
Stage	Pilot / scaling / production
Business case reference	Link, or "none" recorded as a finding
Baseline captured?	Y/N, date, where stored
Unit of work	The denominator everything divides by: ticket resolved, change merged, document processed, workflow completed

The unit of work is the most important field on the sheet. Interpretation Without a clear denominator, cost per unit cannot be calculated, and the use case cannot be compared to alternatives or tracked over time. Interpretation

Sheet 2: Direct AI costs (monthly)

Direct AI costs include model and API consumption, seat licences, embedded SaaS AI tiers, fine-tuning runs, and hosting infrastructure. Interpretation This is the cost layer most business cases capture, but it typically represents only 20-40% of the fully loaded cost. Interpretation

Model and API consumption:

Input tokens, output tokens, blended £/unit by provider
Agentic multiplier: tokens per task for agent-driven work versus interactive work (tracked separately; this is where forecasts break) Seat licences and SaaS tiers:
Tool licences for this workflow
Embedded SaaS AI tier allocation (the AI-attributable share of the relevant SaaS contract, even if estimated) Training and hosting:
Fine-tuning or training runs, amortised over expected life
GPU, hosting, vector store, orchestration platform share

Sheet 3: The costs business cases forget

This sheet is the worksheet's reason to exist. Interpretation These are the costs that appear in operating budgets, engineering time, and governance overhead but rarely make it into AI business cases. Interpretation

Human review and rework:

Minutes per output × volume × loaded rate of the reviewer
For coding tools, review of AI-generated changes; for service AI, QA sampling and escalation handling
Share of AI outputs corrected or redone, × cost of correction

When review costs exceed model costs: Starbucks retired its inventory management AI agent after discovering the system's recommendations were being routinely overridden by store managers who understood local context the model missed. The agent consumed resources but delivered no measurable improvement over human judgement. The retirement decision came only after establishing baseline performance metrics and full-stack cost accounting, including the manager review time that exceeded the model API cost.

Quality gates and governance:

Senior sign-off, compliance review steps that didn't exist pre-AI
Security and compliance assessments (one-off and recurring)
Evaluation harness and monitoring: building and running the thing that tells you quality held Integration and enabling infrastructure:
Integration and pipeline build: one-off, amortised over an assumed life (state the assumption)
Data preparation and retrieval infrastructure: the enabling layer - cleaning, permissions, indexing Organisational costs:
Change management and training
Vendor management overhead (meaningful once tool count grows)

Failure-state consumption:

For agentic workflows, tokens burned by retries, loops and abandoned runs - tracked as its own line, not blended into the average

Sheet 4: Unit economics

Unit economics is where the worksheet produces its most important output: the fully loaded unit cost versus the naive unit cost most business cases quote. Interpretation

Calculations:

Fully loaded monthly cost (Sheets 2+3) ÷ units of work = fully loaded unit cost
Same calculation with Sheet 2 only = naive unit cost
The gap between the two numbers is itself a finding worth reporting Comparators:
Pre-AI unit cost from the baseline, or the human-only alternative at its full loaded cost
Threshold: the unit cost at which this use case is agreed to be working (set in advance, signed by finance and the use-case owner)

Sheet 5: Pilot-to-production bridge

The pilot-to-production bridge addresses the most common business case failure: assuming pilot economics scale linearly to production. Interpretation They rarely do. Interpretation

For each assumption, pilot value → production value → reason it changes:

Volume and case mix:

Volume (and whether case mix changes with it - the hard 20% usually arrives at scale) Consumption patterns:
Tokens per unit (retrieval depth, context growth, agentic intensity)
Review rate (pilots skip the quality gate production requires) Edge cases and build costs:
Edge-case share and cost
One-off build costs not incurred in pilot

Output:

Production fully loaded unit cost versus pilot unit cost, as a multiple. If the multiple is unknown, the scaling decision is being made on fiction.

Sheet 6: Forecast and variance

Forecast and variance tracking prevents the budget surprises that force reactive rationing. Interpretation

Forecast structure:

12-month consumption forecast by line, with stated adoption and intensity assumptions
Actual versus forecast, monthly, with variance triggers
Suggested triggers: investigate at 15%, restrict at 30% Sensitivity analysis:
Vendor price change ±25% (see Cursor, June 2025)
Agentic intensity ×2
Adoption ±20%

Sheet 7: Value side (paired, so the sheet never ships cost-only)

The worksheet must never ship cost-only. Interpretation Sheet 7 pairs cost with value, ensuring every TCO analysis includes the outcome side. Interpretation

Value tracking:

The measured outcome metric for this use case and its current value versus baseline
Value realised versus business case, quarterly
Attribution status: outcome-linked / activity-linked / unattributed Net position:
Value evidenced minus fully loaded cost, with confidence noted

Worked example: Customer service AI agent

A concrete worked example shows how to use the seven-sheet model for a specific use case. Interpretation This example uses a customer service AI agent handling tier-1 support queries.

Use case header (Sheet 1)

Use case: Customer service AI agent for tier-1 support
Owner: Head of Customer Operations
Workflow: Agent handles incoming support tickets, escalates complex cases to human agents
Stage: Scaling (pilot complete, moving to full production)
Baseline: Pre-AI resolution time 4.2 hours, cost per ticket £18.50
Unit of work: Ticket resolved

Direct AI costs (Sheet 2) - Monthly

Model consumption: 2.5M input tokens, 800K output tokens at £0.015/1K tokens = £49.50
Agentic multiplier: 3.2× tokens per ticket versus interactive chat (agent retrieves context, checks knowledge base, formats response)
Orchestration platform: £2,400/month (shared across 8 use cases, allocated 15% = £360)
Vector database: £180/month
Total Sheet 2: £589.50/month

Hidden costs (Sheet 3) - Monthly

Human review: 12% of tickets reviewed, 8 minutes per review, 2,400 tickets/month × 12% × 8 min × £45/hour loaded rate = £1,728
Rework: 4% of tickets require human correction, 15 minutes per correction at £45/hour = £1,080
Quality gate: Weekly QA sampling, 4 hours/week × £55/hour = £880
Integration build: £24,000 one-off, amortised over 24 months = £1,000/month
Data preparation: Ongoing knowledge base curation, 20 hours/month × £50/hour = £1,000
Monitoring and evaluation: £400/month
Change management: Training and adoption support, 10 hours/month × £60/hour = £600
Total Sheet 3: £6,688/month

Unit economics (Sheet 4)

Volume: 2,400 tickets/month
Fully loaded unit cost: (£589.50 + £6,688) ÷ 2,400 = £3.03 per ticket
Naive unit cost: £589.50 ÷ 2,400 = £0.25 per ticket
Gap: 12.3× (£7,277.50 ÷ £589.50: the naive cost understates the true cost by a factor of 12)
Pre-AI baseline: £18.50 per ticket
Savings: £15.47 per ticket (84% reduction)

Pilot-to-production bridge (Sheet 5)

Volume: Pilot 200 tickets/month → Production 2,400 tickets/month (12× increase)
Case mix: Pilot handled curated easy cases → Production includes full case mix (hard 20% drives review rate from 5% to 12%)
Review rate: Pilot 5% → Production 12% (quality gate requirement)
Tokens per ticket: Pilot 2.1× → Production 3.2× (deeper context retrieval at scale)
Production effect: Fully loaded unit cost rises materially from pilot to production, driven by the higher review rate and deeper retrieval. Complete this sheet with your own pilot figures to quantify the step up.

Forecast and variance (Sheet 6)

12-month forecast: £87,330 (£7,277.50 × 12, assumes 2,400 tickets/month steady state)
Sensitivity: AI vendor prices +25% (all Sheet 2 lines) = £89,099; agentic intensity ×2 (doubles model consumption) = £87,924. Neither moves the forecast by more than about 2%, because most of the cost sits in the human review, rework and quality lines on Sheet 3.
Variance trigger: Investigate if monthly cost exceeds £8,369 (15% over the £7,277.50 monthly forecast)

Value side (Sheet 7)

Outcome metric: Customer satisfaction score (CSAT)
Baseline CSAT: 3.8/5.0
Current CSAT: 4.1/5.0 (8% improvement)
Resolution time: 4.2 hours → 0.8 hours (81% improvement)
Attribution: Outcome-linked (measured before/after with control group)
Net position: £44,400/month pre-AI cost (£18.50 × 2,400 tickets) minus £7,277.50 fully loaded AI cost = £37,122.50/month net value (high confidence). This equals the £15.47 saving per ticket × 2,400 (£37,128 using the rounded saving). The saving is already net of the AI cost, so the AI cost is deducted once only.

This worked example shows the worksheet in practice: the naive cost of £0.25 per ticket would have produced a wildly optimistic business case, while the fully loaded cost of £3.03 per ticket still delivers 84% savings versus the pre-AI baseline. Interpretation

Design rules for implementation

The worksheet must be usable by practitioners, not just comprehensible to executives. Interpretation Three design rules ensure practical utility: Interpretation

1. Label every input cell with provenance

Every input cell must be labelled with where the number comes from: invoice, telemetry, sampled study, estimate. Interpretation Estimates are allowed, but visibly so. Interpretation This prevents the false precision that undermines trust in cost models. Interpretation

2. Produce three numbers a CFO can repeat

The worksheet must produce three numbers a CFO can repeat in any forum: Interpretation

Fully loaded unit cost
Naive-versus-loaded gap
Attribution status

3. Ship with a worked example

Ship the worksheet with one completed worked example using real-shaped numbers, so users see what good looks like rather than staring at a blank grid. Interpretation

Mapping to the AI TCO Framework

The seven-sheet worksheet maps directly to the canonical seven-layer AI TCO Framework. Interpretation

Seven-layer TCO mapping:

Infrastructure → Sheet 2 (GPU, hosting, vector store, compute capacity)
Data and context → Sheet 3 (data preparation, retrieval infrastructure, indexing)
Models → Sheet 2 (model consumption, API calls, fine-tuning)
Integration and workflow redesign → Sheet 3 (integration build, workflow changes, API work)
People and capability → Sheet 3 (review time, rework, training, platform engineering)
Governance, safety, and compliance → Sheet 3 (quality gates, security assessments, monitoring, compliance)
Operations and portfolio oversight → Sheet 3 (monitoring, vendor management, portfolio review)

Lifecycle stage mapping:

Prepare: Sheet 1 (baseline capture, unit of work definition)
Build: Sheet 3 (integration build, data preparation, one-off costs)
Run: Sheets 2, 3, 4, 6 (operating costs, unit economics, variance tracking)
Scale/Retire: Sheet 5 (pilot-to-production bridge), Sheet 7 (value validation for continuation decisions)
Govern (cross-cutting): Sheet 3 (governance costs), Sheet 7 (attribution and value evidence)

What to do next

For finance leaders:

Start with one high-visibility use case and complete the worksheet with the platform team.
Use the naive-versus-loaded gap as the lead finding in the next AI investment review.
Establish the worksheet as the standard template for all AI business cases going forward. For platform and engineering leaders:
Instrument the top three AI workflows so Sheet 2 and Sheet 6 can be populated from telemetry rather than estimates.
Work with finance to establish the loaded rates for review time and rework that populate Sheet 3.
Build the pilot-to-production bridge (Sheet 5) into every scaling decision. For operating-model leaders:
Use the worksheet as the economic foundation for AI governance, not a finance-only exercise.
Track attribution coverage (Sheet 7) as the lead metric for AI value management maturity.
Publish completed worksheets internally to build shared understanding of what AI really costs to run well.

References and further reading