Designed by function, lived by journey
47%of customers leave after two or three poor interactionsGENESYS ’26
CTO / CIO Briefing · 2026–2027
Written for the CTO and the CIO
They promised the experience.You have to make it runin production.
Every AI commitment made in front of a customer lands on your side of the house as an integration, a retrieval accuracy number, a run cost, and an audit trail. This is what changed, what you are measured on, and where f/cx.ai plugs in.
71%
face budget cuts or freezes if AI targets are missed
CIO SURVEY ’2685%
say explainability gaps delay or stop production
CIO SURVEY ’2619%
of AI initiatives are meeting or beating their goals
STATE OF CIO ’2682%
say staff build AI faster than IT can govern it
CIO SURVEY ’2674%
regret a major AI vendor or platform choice
CIO SURVEY ’2631%
of CX infrastructure is fully cloud-based today
GENESYS ’2600
The problem f/cx.ai is solving
Bought by function · lived as one journey
Every function is putting AI into the customer path on its own plan, budget and scorecard. The customer meets all of it as one experience. The seams between functions are where value leaks, trust breaks and AI spend loses its return — and no single function owns them.
From the CTO / CIO seatPlatforms get bought before the journeys they serve are designed, so data and integration debt surface after signature.
Six plans, six AIs, six numbers. One customer falls through the seams. One experience graph under every function. The seams become designed handoffs.
MarketingOwn AIShared modelPipelineOne composite
ProductOwn AIShared modelRelease velocityOne composite
InnovationOwn AIShared modelPilots launchedOne composite
ServiceOwn AIShared modelHandle timeOne composite
TechnologyOwn AIShared modelUptimeOne composite
OperationsOwn AIShared modelCost per caseOne composite
No shared model — every function argues from its own data
One experience graph · one journey model · one evidence base · visible weights
Customer journey
Promise ≠ deliveryContext lostPilot strandedData re-keyedHandoff dropped
Six dashboards report green. The customer still leaves.
One composite the board reads: return on experience, beside ROI.
Three failures underneath it
Decided by opinion, not evidence
61%name executive misunderstanding of AI as the top cause of misalignment
Measured after the spend
31%have a measurement frame for agentic AI once it is liveADOBE ’26
What f/cx.ai does instead
f/cx.ai is a customer experience design and innovation agency built to own the seams — one direction, one model, one measure.
01
Design the journey end to end
Across every function that touches the customer — strategy, research, design and AI in the path, from one agency.
02
Decide on one shared model
The ToolKit puts the same evidence and visible weights in front of every seat.
03
Prove it before it is funded
Gates and a measurement frame agreed up front, so year two survives.
The model stays with your team when we release, so the next cycle starts ahead
01
Six challenges that land on the technology seat
AI adoption × customer experience
Pick a pressure on the map — each gauge is the figure behind it
The CTO / CIO agenda6pressures, one seat
01 / 06
Prove the value or lose the budget
19%of AI initiatives are meeting their goals
71% of CIOs face cuts or freezes if AI targets are missed, and 85% now have compensation tied to measurable AI outcomes — while only 19% of AI initiatives are meeting their goals.
What it costs
A platform investment argued on faith in its second budget cycle.
02 / 06
Data debt is the actual bottleneck
50%higher AI failure rate where data debt waits
Fewer than half report adequate data quality and access for AI. IDC expects CIOs who defer data debt remediation to see 50% higher AI failure rates by 2027.
What it costs
Model underperformance that is really a knowledge and pipeline problem.
03 / 06
Explainability blocks production
85%say traceability gaps stall production
85% say traceability and explainability gaps delay or stop projects from reaching production, and 70% expect new AI audit or explainability requirements within the year.
What it costs
A build that passes QA and still cannot be launched.
04 / 06
Shadow AI outruns governance
82%see employees outrun IT governance
82% report employees standing up AI faster than IT can govern it, and 54% have already found unsanctioned tools in use inside the business.
What it costs
Customer data in systems nobody scored, owned, or logged.
05 / 06
Platform bets made before the journeys
74%regret a major AI platform decision
74% regret at least one major AI vendor or platform decision from the past eighteen months, and 62% have had a CEO question one of them.
What it costs
Integration debt discovered after signature instead of before it.
06 / 06
The customer path went real time
31%of CX infrastructure is fully cloud-based
82% of CX leaders expect autonomous agents to orchestrate the experience within three years, while only 31% of CX infrastructure is fully cloud-based today.
What it costs
Orchestration promised on a stack that cannot decide in the moment.
Where the gap sits on the technology side
expectation on the left of every pair, capability on the right ExpectationIn placeAgentic AI on next year’s plan
Expect autonomous orchestration70%82%
Expect autonomous orchestration70%82%
AI initiatives meeting goals
CX infrastructure fully cloud19%31%
CX infrastructure fully cloud19%31%
Expect new audit requirements
Comp tied to AI outcomes70%85%
Comp tied to AI outcomes70%85%
Have defined AI KPIs
Adequate data quality for AI47%<50%
Adequate data quality for AI47%<50%
Read it as one sentence. The expectation is now contractual and the substrate is not ready, so the constraint on the 2027 experience is knowledge quality, orchestration capability, and evidence a regulator would accept — decided journey by journey, before the platform is chosen rather than after. That is a design and decision problem, which is the work f/cx.ai does.
02
What the CTO and CIO have to deliver again in 2027
Five books, one composite
Uptime and delivery stayed. Data readiness, AI in production, and audit-grade traceability were added, and the technology seat is now measured on business outcomes it does not fully control. These are the measures the board reads together, and the constraint that breaks each one.
Pick a book on the ring for its measures and the constraint that breaks it
5books20 measures · one composite
Book 01 / 05
Delivery
- 1Pilot-to-production rate
- 2Release throughput and lead time
- 3Change failure rate
- 4Time to restore service
Breaks when
Throughput is high and none of it reaches the customer path.
Book 02 / 05
Data & platform
- 1Data quality and access for AI
- 2Retrieval accuracy on real journeys
- 3Integration debt retired
- 4Real-time orchestration coverage
Breaks when
Knowledge is treated as content rather than as a system requirement.
Book 03 / 05
AI in production
- 1Share of contact resolved end to end
- 2Answer accuracy and escalation quality
- 3Containment with resolution
- 4Cost per interaction and per inference
Breaks when
Deflection is counted as success and the customer calls back.
Book 04 / 05
Trust & risk
- 1Traceability and audit readiness
- 2Eval-set and red-team coverage
- 3Shadow AI surfaced and governed
- 4Incidents in the customer path
Breaks when
Evidence is assembled for the audit instead of captured during the build.
Book 05 / 05
Value & cost
- 1Total cost to serve
- 2Run cost per workload
- 3Benefit realized against the case
- 4Payback on the AI investment
Breaks when
The value case was written by someone who will not own the run cost.
03
What f/cx.ai does about each one
Challenge → engagement → artifact you keep
Pick a challenge to trace its wire to the engagement that answers it
ChallengeEngagement
Agentic Service Design & Deployment Plan
CX AI Readiness & Opportunity Assessment
CX Governance, Trust & Enablement
CX Strategy & Operating Model Design
Contact Center Modernization & Agent Assist
Fractional CX Leadership
The artifact you keep
CX AI Readiness & Opportunity Assessment — A ranked backlog with a modeled return per line, and an investment narrative finance can defend.
The artifact you keep
Contact Center Modernization & Agent Assist — Knowledge rebuilt to the standard retrieval actually needs, with cited, confidence-gated assist in the flow.
The artifact you keep
CX Governance, Trust & Enablement — Guardrails with enforcement and named owners, evaluation sets, and red-team evidence before build.
The artifact you keep
CX Strategy & Operating Model Design — Decision rights, funding routes, and the forums that turn scattered activity into scheduled work.
The artifact you keep
Fractional CX Leadership — A vendor evaluation scorecard weighted against your journeys, and commitments pressure-tested before signature.
The artifact you keep
Agentic Service Design & Deployment Plan — What AI resolves, what it must never touch, and escalation that carries full context to a person.
04
Four ways a CTO or CIO starts
Start where the decision is
Pick a line to ride it into the gates
Four lines in, one gated trunk
every move ends in a decision, and nothing reaches production without oneStart hereEvidenced baselineContact drivers, journey mapping, and the data, knowledge, and platform audit behind them.
Gate 01Criteria before buildAccuracy, safety, and escalation thresholds agreed and written down before anyone builds.
Gate 02Safe to launchEvaluation sets and red-team evidence, rehearsed with the approvers who can decline.
OutcomeBoard-ready roadmapInvestment by workstream, the dependencies that actually threaten it, and a measurement frame.
Line 01 · Working session
Executive Briefing
Not a presentation. One honest baseline for what AI can and cannot do across your experience, and a ranked shortlist the room built itself.
Line 02 · Fixed scope
Readiness Assessment
A data, knowledge, and platform audit across CCaaS, CRM, and the content retrieval would have to rely on, ending in a ranked backlog with a modeled return.
Line 03 · Attaches
Governance & Trust
Guardrails with enforcement and named owners, evaluation sets, and red-team evidence — priced inside whatever it governs, never bought on its own.
Line 04 · Embedded
Fractional CX Advisor
Senior CX and AI direction in the room at the moment decisions are made, without a full-time hire or an agency bench underneath it.
05
The ToolKit: 22 modules, one experience graph
Proprietary f/cx.ai IP
Twenty-two working decision-intelligence modules that move an engagement from nobody agrees what is true to a gated roadmap a board can fund. Not a deck — live models your team operates in the room, that change answers when an input changes, and that stay with your people when we release.
22working modulesEvery one a live model, not a slide
4phasesDiagnose · Design · Decide · Deliver
1experience graphChange an input and every view moves
0static decksIt stays with your people when we release
Hover a node, change the phase, or open the whole graph
The graphFour phases, twenty-two modules, one model.
Hover a node, or switch the phase.
What is actually true. The baseline nobody argues with.
Experiences Dashboard — SummaryExperiences Dashboard — ActivePersona ManagerTouchpoint MapperSegmentation ProfilerSentiment AnalysisVoice of the CustomerAI-Driven Experience Audit Framework
You leave with
A ranked value-leak list, each item traced to its evidence.
What we intend instead. Options tested before the expensive commitment.
Journey OrchestratorJourney MappingOmni-Channel ModelerExperience OptimizerExperience Recommendation Scoring
You leave with
A designed AI handoff, including the failure path — and the concepts that did not survive.
What we will and will not do. The phase clients skip.
CX Maturity Assessment ModelerCustomer Experience ScorecardExecutive Alignment MappingExperience Aligned Heat MapCapabilities RankerExperience Rationalization
You leave with
A weighted ranking you set the weights for, and a not-this list with an owner.
What the board can fund. Sequence, gate, measure, hand over.
Customer Experience RoadmapReturn on Customer ExperienceCommunity
You leave with
A gated roadmap, a measurement frame agreed before anything is built, and the model itself.
Inside the ToolKit
Tour all 22 modules, as your team will use them
toolkit.fractionalcx.ai/dashboard-summary
01 Diagnose · module 01 of 22
Experiences Dashboard — Summary
One view of every experience in flight, planned and proposed — volume, cost to serve, satisfaction, and the signals that should change what gets funded next quarter.
You leave with
A ranked list of the decisions worth making this quarter, each with value at risk and model confidence attached.
When we reach for it
Session one, every engagement. The fastest way to find out whether the client’s numbers agree with each other.
01 Diagnose · module 02 of 22
Experiences Dashboard — Active
Every live experience by channel and persona, with the touchpoint events moving the numbers and the four measures the executive team actually asks about.
You leave with
A defensible read on which programs are progressing and which are quietly degrading.
When we reach for it
Steering committees and monthly reviews — it turns “how is it going” into a specific conversation.
01 Diagnose · module 03 of 22
Persona Manager
The people the experience is actually for — each persona traceable to research, with needs, tasks and channel preference, scoping every other module.
You leave with
A validated persona set the whole organization can name, with the research base behind each one.
When we reach for it
Before any journey work. If the client cannot agree who the journey is for, nothing downstream survives review.
01 Diagnose · module 04 of 22
Touchpoint Mapper
The journey as the customer experiences it — every touchpoint, what they are thinking, and where sentiment, effort and cost diverge from what the organization believes.
You leave with
A journey map with evidence attached to every stage and the leak points quantified.
When we reach for it
Discovery workshops. The sentiment-versus-belief gap is usually the moment the room changes its mind.
01 Diagnose · module 05 of 22
Segmentation Profiler
Who the base actually consists of, what each segment is worth, and which submarkets have no owner — click a segment and the whole page scopes to it.
You leave with
A segment model with value index, coverage gaps and behavioral indices per segment.
When we reach for it
Whenever growth targets and service design are being argued in different rooms.
01 Diagnose · module 06 of 22
Sentiment Analysis
Market and product sentiment across tracked dimensions, six months of movement, and the strategic landscape behind the scores.
You leave with
A banded sentiment map showing where a category is still open, with evidence per dimension.
When we reach for it
Competitive and category work — the weakest scores are usually where the opening is.
01 Diagnose · module 07 of 22
Voice of the Customer
What customers are telling you, clustered and weighted, and where in the journey the telling turns into leaving.
You leave with
Verbatim clusters ranked by volume and sentiment, tied to the journey stage they belong to.
When we reach for it
When the client has plenty of listening and no synthesis — which is most of the time.
01 Diagnose · module 08 of 22
AI-Driven Experience Audit Framework
Point it at any experience asset — an interface, a website, a product, a portal, a digital journey or a service touchpoint — and it runs a structured AI audit end to end, scoring against the framework’s criteria and attaching a validated reference to every recommendation.
You leave with
A decision package rather than a findings report: what is wrong, what to do, what it is worth, what inaction costs, and in what order — packaged for a formal accept-or-decline.
When we reach for it
First contact with an existing experience, and the module most leadership teams reach for first. The fastest route from “this needs work” to a sequenced plan an executive can approve.
02 Design · module 09 of 22
Journey Orchestrator
The journey as a policy rather than a map. The engine scores every candidate action across marketing, sales, commerce and service and selects the next best one in real time, for each customer, on every event.
You leave with
An orchestration policy — the weights, the guardrails and the action mix they are willing to run — plus the value gap against their current fixed journeys.
When we reach for it
When the client wants AI to run the journey rather than decorate it. The live race against their fixed journeys is usually what settles the argument about whether adaptive is worth it.
02 Design · module 10 of 22
Journey Mapping
The orchestration behind the journey — every automated step, every handoff to a person, every exit. Drag to rearrange; connections and throughput follow.
You leave with
A working orchestration model with throughput and loss at every step.
When we reach for it
Designing agentic service. It makes the handoff design visible before anyone builds it.
02 Design · module 11 of 22
Omni-Channel Modeler
How the same customer moves between channels — the handoffs that carry context, the ones that do not, and what each switch costs to serve.
You leave with
A current-state and designed-state channel model with the cost of unnecessary switching quantified.
When we reach for it
Cost-to-serve conversations. The switching matrix usually pays for the engagement on its own.
02 Design · module 12 of 22
Experience Optimizer
One idea, every credible way to build it, and the evidence for each — select the variants worth a two-week prototype sprint.
You leave with
A sprint plan with criteria agreed before the build, and the economics of the answer.
When we reach for it
Innovation cadence work. It is how a good idea stops being one person’s conviction.
02 Design · module 13 of 22
Experience Recommendation Scoring
Every finding, what it would take to fix, and what fixing it is worth — select recommendations and watch the composite move.
You leave with
A scored recommendation set with gain per engineering week, ready to defend.
When we reach for it
Building the funding case. The argument gets made here rather than in the meeting.
03 Decide · module 14 of 22
CX Maturity Assessment Modeler
Where each capability operates today, where next year needs it to be, and the honest distance between the two across six maturity levels.
You leave with
A capability-by-capability maturity read with the gap named and sequenced.
When we reach for it
Executive briefings. Most organizations stall at Align — this is where they find out if they have.
03 Decide · module 15 of 22
Customer Experience Scorecard
Ten CX success drivers scored against a peer benchmark, weighted so the drivers that decide outcomes move the composite most.
You leave with
A weighted composite against peer median, with strengths to hold and the constraints that decide next year.
When we reach for it
Assessment delivery. Drag a score in the room and the radar answers immediately.
03 Decide · module 16 of 22
Executive Alignment Mapping
The same portfolio scored twice — once by the function that owns the customer, once by the function that owns the systems. The gaps are the agenda.
You leave with
A value-versus-risk portfolio map plus a ranked list of where the two functions genuinely disagree.
When we reach for it
When a program is stuck and nobody will say why. A two-point gap is a different-outcomes problem.
03 Decide · module 17 of 22
Experience Aligned Heat Map
Where the experience opportunity is hottest set against whether the organization is ready to act on it — the gap between the two is the planning input.
You leave with
An opportunity-versus-readiness map that separates capability investment from experience investment.
When we reach for it
Portfolio planning across business units or sectors.
03 Decide · module 18 of 22
Capabilities Ranker
Every candidate feature scored on the same five criteria with the weights visible — move a weight and the ranking re-sorts live.
You leave with
A ranked, weighted portfolio and a visible funding line.
When we reach for it
The prioritization session. Agree the weights first; the list is then arithmetic, not politics.
03 Decide · module 19 of 22
Experience Rationalization
Every initiative scored against the same alignment criteria — customer, business and technology — so the portfolio is argued on evidence rather than enthusiasm.
You leave with
An alignment matrix with a defensible score per initiative and a not-this list that holds.
When we reach for it
Whenever a client is running experience programs in isolation from the strategy.
04 Deliver · module 20 of 22
Customer Experience Roadmap
The sequence, the decision gates and the dependencies — a roadmap the board can fund and the organization can execute in order.
You leave with
A gated roadmap with investment by workstream and the dependencies that actually threaten it.
When we reach for it
The last third of every engagement, and the artifact clients circulate most.
04 Deliver · module 21 of 22
Return on Customer Experience
The measure a CFO will accept — value per customer moved by things the organization controls, with the employee data that predicts it sitting alongside.
You leave with
A defensible RoX index and the leading indicators behind it.
When we reach for it
Value realization and renewal conversations.
04 Deliver · module 22 of 22
Community
The practice network — what other people working the same problems are publishing, and the threads worth following.
You leave with
A live channel between the client’s team and the f/cx.ai practice after the engagement ends.
When we reach for it
Handover. Capability that lives with their people needs somewhere to keep growing.
01Diagnose8
02Design5
03Decide6
04Deliver3
One experience graph
Change one input. Every view moves.
01Diagnose
Persona Manager
Voice of the Customer
Touchpoint Mapper
AI-Driven Experience Audit
02Design
Journey Mapping
Experience Optimizer
Journey Orchestrator
03Decide
Executive Alignment Mapping
Capabilities Ranker
Experience Rationalization
04Deliver
Customer Experience Roadmap
Return on Customer Experience
Every module reads and writes the same model, so nobody re-keys a number or argues from a different spreadsheet. That is what “one experience graph” means in the room.
06
Four modules to ask for by name
Built for the technology seat
Every panel is live — pick two seats, move a weight, drag a driver
Decide
Executive Alignment
One portfolio, scored by two seats at the table. Choose which two, and the distance between their scores is what the next meeting is actually about.
The gaps are the agenda.
Illustrative
DisagreementPick two seats. Hover a candidate for the distance between their scores.
Value × risk, scored by two seats
Design
Omni-Channel Modeler
How the same customer moves between channels — the handoffs that carry context, the ones that do not, and what each switch costs to serve.
The switching matrix often pays for the engagement.
Illustrative
SwitchRed is a switch that drops context. Click one to price it.
Red = context lost on switch
Design
Journey Mapping
The orchestration behind the journey — every automated step, every handoff to a person, every exit. Rearrange it and connections and throughput follow.
Throughput and loss at every step.
Illustrative
StepHover a step for throughput and where it goes.
Throughput and loss at every step
Deliver
Return on Customer Experience
Value per customer moved by things the organization actually controls, with the leading indicators that predict it sitting alongside.
The measure a CFO will accept.
Illustrative
IndexMove the one thing you control and the index answers.
Index + leading indicators
07
What the CTO and CIO gain
Eight outcomes, not eight deliverables
01
Platform chosen against journeys
A vendor scorecard weighted on your real journeys, not on the demo.
02
Integration debt known early
Surfaced in the audit, before the contract instead of after it.
03
Knowledge built to spec
Rebuilt to the standard retrieval needs, so the accuracy gate is passable.
04
Gates agreed before build
Accuracy, safety, and escalation thresholds set while they are still cheap.
05
Traceability that survives audit
Every score and recommendation carries its evidence, six months later.
06
Shadow AI given a forum
Decision rights, funding routes, and named owners instead of a ban.
07
AI that resolves, not deflects
Containment scored with resolution, escalation carrying full context.
08
The capability, kept
The experience graph and scoring model stay with your people when we release.
Another seat at the table
Sources. 2026 CIO AI accountability survey; Foundry State of the CIO 2026; IDC 2026 CIO Agenda Predictions; Adobe 2026 AI and Digital Trends; Genesys 2026 State of Customer Experience.
f/cx.ai is a customer experience design and innovation agency for the AI era, working with enterprise organizations to design, prototype, and prove the journeys, service models, and AI agents their customers actually move through. Market figures cited are drawn from published industry research and are shown for context, not as a projection for any one organization. Engagement timing is set by scope and complexity and confirmed at scoping. ToolKit module outputs shown in any f/cx.ai material are illustrative and are not any named client’s results.