UX for Spaces™

This document defines UX for Spaces™, a design method for how the technology layers of a space support what the people in it need. The method was authored by Jimmy Powers, Principal of Blend Technology, and Blend applies it on its own projects.

The method is called UX for Spaces™. Its working instrument is the UX for Spaces™ Framework, currently version 15.5, published as two workbooks, residential and commercial. This document explains the method so that architects, designers, builders, owners, and anyone researching it can understand it, apply its vocabulary, and cite it.

Four kinds of statement appear here, and they are labeled so the reader can tell them apart:

Every space is built for something. A primary suite is for rest and recovery. A study is for focus. A guest suite is for arrival and sleep. A dining room is for conversation. UX for Spaces™ begins by stating that intention plainly, before any product is chosen, and then designs every technology layer, and the building assemblies those layers depend on, to hold it. Two intentions show how the method runs.

A space for rest, recovery, and sleep. The intention is that a person lies down and the space holds dark, quiet, and a stable temperature until they choose to leave it. Light at the eye is held under 1 lux melanopic EDI, which means blackout that seals, indicators that go dark, and a night path from amber or red sources only. Sound at the pillow is held to a steady floor of NC-25 with no start, stop, or speed change audible above it, which is a mechanical design decision made before the equipment is sized, not a setting. Temperature holds within 1 degree of the sleeper's choice and corrects without a fan spinning up. Nothing in the space announces itself: no chime, no glow, no notification. Where isolation from an adjacent space is needed, the partition, the door, and the floor are specified for it on the construction documents, because no scene can add mass to a wall after it is closed.

A space for work and focus. The intention is that a person sits down and the space supports attention for as long as the work takes. Every layer has a job:

• Shading. Privacy shades close the space to sightlines from outside, and glare control keeps direct sun and reflected glare off the task plane and the screen, while daylight is kept where it helps. Both are automatic once the focus scene is called.
• Lighting. Spectrum and level are set to keep the nervous system alert rather than drowsy: 250 lux melanopic EDI or more at the eye through the working day, from sources whose spectral data is on file, no bare source visible from the seated position, and electric light that compensates as daylight fades so neither the task illuminance nor the melanopic level sags through the afternoon.
• Climate. Temperature holds inside a tight band and corrects silently, with CO2 used as the ventilation trigger so the air is refreshed before it dulls attention.
• Acoustics. The space is designed so that speech from adjacent spaces is not intelligible at the desk, and its own noise floor stays below NC-30. Where an adjacent space cannot be quieted, masking covers what leaks in, at a tuned level under the focus ceiling of 40 dBA.
• Attention. During a focus scene or a video call, the doorbell does not sound in the space. The arrival is shown by a light indicator, a soft pulse on the keypad and the entry fixture, so the person sees it when they look up and is not pulled out of the task by a chime. Notifications hold until the scene ends.
• The drawings. If the space must be isolated from sound, that isolation is an assembly, and the assembly is on the MEP plans, not only on the architectural set. An isolated shell cannot be penetrated by a duct, a conduit, or a plumbing run without losing what it was built for, so it changes where the mechanical pathways go, where the return air comes from, and how the supply is lined and silenced. Those decisions are made at Schematic Design and confirmed at Construction Documents, or they are not made.

Both spaces follow the same chain: the stated need becomes a criterion with a unit, the criterion becomes coordinated behavior across every layer, each behavior is assigned to a trade and a drawing at a construction stage, the built result is measured against the criterion, and the record is adjusted in use. The rest of this document defines each link.

UX for Spaces™ exists because the layers that shape how a space feels are designed by different people who rarely design them together. Lighting, acoustics, HVAC, air quality, shading, and controls each pass their own specification. Nobody is responsible for how they behave in combination for the person in the space. The result is a space that meets every individual spec and still fails at what it was for.

The method fixes the responsibility gap. It defines what each space must do for the people in it, writes that as behavior the building can carry out, routes each behavior to the trade and construction stage that can deliver it, and then measures whether the built space does what was written.

It applies to residences, hospitality properties, workplaces, learning environments, retail and experience spaces, and performance spaces such as cinemas, studios, and listening spaces. The UX for Spaces™ Framework v15.5 covers all of these in seventeen behavioral categories, described below.

The method covers:

• vocabulary
• turning human needs into a space brief
• writing behavior specifications
• resolving priorities and conflicts
• coordinating across trades
• timing decisions against construction
• documentation
• commissioning
• adjustment in actual use
• ownership and support
• version governance

It does not cover equipment selection, brand comparisons, or pricing. Equipment is chosen to satisfy the behaviors, after the behaviors are written, and it is not the subject of this document.

A UX for Spaces™ designer can carry the method through construction, calibration, and ongoing care, as Blend does on its own projects, or hand a completed design to a qualified integration partner. The method is the same either way.

These terms are used exactly as defined here throughout UX for Spaces™ documents. All are method definitions unless noted.

Every requirement in a UX for Spaces™ project travels the same six links. If a link is missing, the requirement is not yet designed.

Read left to right: a need becomes a measurable or checkable criterion, the criterion becomes a set of physical and programmed decisions, each decision is assigned to a trade and a drawing at a construction stage, the built result is checked against the criterion, and the record is adjusted in use.

The chain is also the test for whether a document is complete. A behavior with no criterion is a wish. A criterion with no trade assignment is a hope. A trade assignment with no verification is an assumption.

The brief comes first. Method: no behavior is written until every space it could touch is named in the brief, because every behavior must bind to a real space in the project.

How the brief is built. Discovery begins with people, not systems. The UX for Spaces™ process calls the first stage Listen and Map: who is in the space, how they move, where they slow down, where things become frustrating or uncomfortable. The output is a map of daily life, the moments that matter most, and an emotional target for each area. The brief converts that map into rows.

What each row holds.

From target words to criteria. A performance target is a word; a criterion is a condition with a unit or a checkable function. Method: each target on a space is expanded into criteria in the behavior register, not in the brief. The brief stays readable by the client. The register carries the numbers.

The expansion draws on the method's practice targets. A few examples, each a practice target unless marked:

Recovery is one target set, whatever the reason. When a space is tagged Recovery, the framework sets every environmental factor at the most demanding published criteria available for healthcare environments and measures the space against them: for sound, the WHO Guidelines for Community Noise values for hospital ward rooms (30 dB LAeq indoors at all hours, night-time events not above 40 dB LAmax, treatment and observation areas not above 35 dB LAeq), which are more demanding than the FGI Guidelines' patient-room range of NC 30 to 40 and therefore govern on level, while FGI Table 1.2-3 supplies the partition criteria (STC 45 between patient rooms on the same floor, 50 floor to floor, 50 to public space, 60 to a service area) and Table 1.2-4 the speech-privacy criterion for confidential rooms (STI 0.12 or less at the unintended listener), with the method's steady floor below NC-20 and the mechanical-transition limits of Acoustic Design row 10; for air, ASHRAE Standard 170-2021 Table 7-1 for a patient room (2 air changes per hour of outdoor air minimum, 4 total, MERV-14 minimum filtration), with HEPA active in the space and the WHO 2021 air quality guideline values for particulates; for thermal, the same table's patient-room design range of 70 to 75 F (21 to 24 C) and relative humidity at or below 60 percent, held within 1 degree of the occupant's chosen value, inside the ASHRAE Standard 55 comfort zone (PMV within plus or minus 0.5, about 10 percent predicted dissatisfied, humidity ratio at or below 0.012); for light, the Brown et al. 2022 melanopic recommendations, and ANSI/IES RP-29-25 Table A-1 for the patient zone of a general patient room: 200 lux general, 4 lux night-light, 30 lux night observation, 400 lux reading at the head of the bed, with night-lights from amber or red sources or below 2800 K and every target reached through at least three-level control. Where no published criterion exists for a factor, the method sets a practice target from the best available evidence and labels it as such. The same targets apply whether the occupant is recovering from surgery, illness, travel, training, or a demanding day. The values are checked against the current edition of each source at project start; editions change, and the framework records the edition each value came from. This is a statement about where the targets come from; it is not a claim that the space is clinical, and the medical judgment stays with the occupant's physician.

What the brief does not do. It does not choose products, and it does not promise outcomes. A space tagged Sleep is a space the building will hold in a sleep posture to defined conditions. Whether a particular person sleeps well depends on many things the building cannot control. The brief commits to conditions.

A behavior is one row. When a stated condition is true, the building responds as specified. That sentence is the whole grammar of the framework, and everything else on the row is routing.

The seventeen categories. The UX for Spaces™ Framework v15.5 organizes behaviors into seventeen categories, in this order: Interface and Control; Acoustic Design; Biological Rhythm; Cognition; Recovery and Wellness; Restoration; High Performance Spaces; Safety and Trust; Access and Presence; Gathering; Preservation; Arrival and Departure; Transitions; Building Health; Sustainability; Learning Environments; Retail and Experience. Each of the two v15.5 workbooks, residential and commercial, holds 272 populated behavior rows across those categories; the two differ in wording where the setting differs, not in structure. Not every row applies to every project; a project record is the subset that binds to its spaces.

The fields on every row.

How a row is written. Method, in order:

1. Write the trigger as a condition a sensor, a clock, a person, or a project milestone can actually produce. "The occupant feels tired" is not a trigger. "Uninterrupted stillness is detected in the focus space for 60 minutes" is.
2. Write the response for every layer it touches, in one place. If the response involves lighting, shading, HVAC, masking, and a notification hold, all five appear on the row. The point of the method is that no layer is specified alone.
3. State transitions as durations. Autonomous lighting transitions fade over a minimum of ten minutes; wind-down transitions run 30 to 45 minutes; a scene change a person calls may be faster but is never instantaneous. Shading moves at speeds below the ambient noise floor. HVAC corrections do not produce audible fan speed changes. These are practice rules from the framework's operating principles.
4. Bind the row to space, user, moment, and priority.
5. Tag intervention depth and Impacts Build. This is where the row becomes a work item for a specific trade at a specific stage.
6. Flag the conflict profile if the row involves sleep, security, privacy, recovery, energy, or safety, and write the conflict entry.

Three example rows from the framework, condensed. These are framework rows, not project results.

The third row shows something the method insists on: many behaviors are not automation at all. They are design reviews with a deadline. The framework treats "a decision must be made before this drawing is issued" as a building behavior, because if the decision is missed the space cannot behave as intended no matter what is programmed later.

Operating principles that govern every row. Six rules from the framework sit above all categories:

• Circadian override. Circadian logic is always overridable, in any space, at any time, by any authorized user. A manual scene, guest preference, or occupant request suspends it for that space until released. The building never imposes biology on someone who chooses otherwise.
• Sleep overrides circadian. When a sleep or rest scene activates, the space leaves the circadian schedule and holds its sleep posture regardless of the time of day, until explicitly released. A 7 am sunrise does not override a guest who needs to sleep.
• One circadian standard, timed per occupant. The evening and night light targets are the same for everyone, set at the level the most sensitive occupant needs; there is no separate rule for children. What varies is timing: each space's wind-down is anchored to the sleep time of its occupant, whoever that is, and shared spaces follow the earliest sleeper present or the occupants' stated choice. A space whose occupant sleeps before sunset either blacks out at wind-down start or follows outdoor light until daylight falls below the evening target; an electric sunset in a daylit space is not attempted. Where an occupant needs more light to move safely at night, pathway output is raised within the amber or red spectrum so the night ceiling still holds.
• Focus cycle. Focus scenes run on a one-hour cycle with a built-in break: at 60 minutes, a peripheral warm light shift for five minutes, no audio, no alert.
• Technology disappears. Devices, speakers, sensors, and controls recede into materials, surfaces, and architecture. No visible hardware unless activated.
• Materials are systems. Material choices are evaluated for acoustic impact, light reflection, and thermal contribution to the space. They are not specified independently of system design. Materials and systems are one specification, not two.

Two categories will target the same space with different requirements. The method expects this and writes the resolution down before construction, so the rule makes it into specification rather than being improvised at commissioning.

The priority scale. Method:

Priority does two jobs. It decides which behavior wins when two collide, and it decides what gets cut when budget or ceiling depth runs out. A row at priority 5 is the first thing removed from a specification; a row at priority 1 is the reason a wall gets thicker.

The conflict register. Every row flagged with a conflict profile has an entry that names the space, the two categories and moments in tension, what each requires, the resolution rule, and the priority of the rule. The framework ships with illustrative patterns that a project adapts or replaces with the household's or operator's own logic. Four of them, condensed:

What programming cannot fix. Method, stated plainly because it is the most common misunderstanding about this work: a resolution rule can decide which scene wins, but it cannot manufacture physical separation or mechanical capacity that was never built.

• If the workout space shares a partition with the sleep space and the partition was not designed for the isolation the sleep space requires, no rule about audio levels makes the sleep space quiet. The conflict is resolved at Construction Documents by the partition assembly, or it is not resolved.
• If the kitchen hood cannot clear cooking at its lowest speed, dimming the task lights for guests does not help the air. The conflict is resolved by hood selection and duct design, or it is not resolved.
• Masking covers intermittent intrusion by raising a steady, tuned floor. It cannot lower a background noise level that already exceeds the target. If the sleep space's HVAC noise is above NC-25, the fix is in the mechanical design, and adding masking on top of it moves the space further from its target, not closer. Masking is still the right supplemental tool where isolation cannot reach the target, as on a rehab, which is why the framework gives every space type a masking ceiling as well as a build target.

Masking ceilings. Practice targets, measured with the tuned spectrum at the occupant position. The build target is for the steady floor; the ceiling is the most masking the space will carry before the source is routed to correction. Anchors: WHO Guidelines for Community Noise (1999) bedroom values of 30 dBA continuous and 45 dBA maximum; Basner et al. 2014 on steady versus fluctuating noise during sleep; ANSI/ASA S12.2-2019 for NC. The FGI Guidelines cap electronic masking in healthcare spaces at a 48 dBA background; every ceiling below sits under it.

A person asleep is woken by change more than by level, so a steady 35 dBA floor that buries a cycling compressor is a better sleep environment than a quieter one that fluctuates. That is the trade masking makes, and it is made only after the source and the transmission path have been addressed: the method's evidence review, Human Performance as a Building Design Requirement, notes a 2026 polysomnographic trial in which the tested pink noise itself altered sleep structure, which is reason to treat continuous masking as the last layer rather than the first. A 45 dBA floor in a bedroom is outside the WHO envelope and is not a trade the framework allows.

This is why the conflict register is written during design and why each entry carries an intervention depth. A conflict whose resolution is "structural" or "architectural" is a drawing change. A conflict whose resolution is "programming only" is a rule. The register says which, before anyone assumes the software will handle it.

A behavior row is a coordination instrument. Its intervention depth tells each discipline whether the row is theirs, and the response text tells them what the row needs from them. The UX for Spaces™ designer holds the space's intent and the register; each trade holds its own drawings and its own professional responsibility.

Who owns what. Method. The table maps intervention depth to the disciplines most often responsible. Real projects assign by contract; this is the default routing.

What each discipline gets from the register. Instead of a technology package that arrives after MEP is complete, each discipline receives its rows early:

• Architecture receives the rows tagged Schematic / DD and Architectural or Structural: proportions of performance spaces reviewed against modal clustering below 300 Hz; STC targets per boundary with a box-in-box recommendation where standard construction cannot reach them; rack location as an architectural decision with thermal, acoustic, power, and service clearance requirements; daylight geometry and orientation documented as inputs to the lighting specification.
• Mechanical receives NC targets per space before duct sizing, the rule that HVAC corrections in sleep and focus spaces must not produce audible fan speed changes, demand-controlled ventilation logic with CO2 thresholds, filtration standards, and the humidity envelope for preservation spaces.
• Electrical and lighting receive the lighting specification gates: minimum CRI 90 in occupied spaces, flicker-free operation per IEEE 1789-2015 at all dimming levels including below 20 percent, constant-current-reduction drivers rather than pulse-width modulation, published SPD data and M-DER for any luminaire in a circadian, sleep, recovery, learning, or wellness space, color rendition by the named TM-30 specification for the space's intent (see Alignment tags and external standards), and glare and uniformity limits.
• Plumbing receives leak and flow monitoring points, space and whole-building shutoff valve locations, and the rule that plumbing runs adjacent to sleep and performance spaces are evaluated as vibration paths.
• Interiors receives the rows that make materials part of the system: RT60 targets that constrain finishes, matte or honed finishes on horizontal surfaces in work, dining, and reading spaces to avoid veiling reflections, surface reflectance values as inputs to daylight reach, and the rule that fixture layout is not finalized until interiors has been consulted.
• Technology receives everything, because the register is its native document, plus the acoustic rules that apply to its own hardware: active equipment in or adjacent to a space with an NC target must operate below that target at every state including idle; every motorized element must operate below the NC target of the space it serves through its full travel; every cable penetration through a rated assembly is sealed, baffled, and fire-stopped and shown on the construction documents.

The seat. The UX for Spaces™ designer's coordination role sits alongside the architect, lighting designer, MEP engineer, and interior designer during Schematic Design and Design Development, in its own seat. It does not replace any of them. The lighting designer still designs the lighting; the register gives the lighting designer the space intent, the gates, and the deadline.

Every active row carries an Impacts Build value: the latest construction stage at which the decision still has to land. Miss it and the behavior becomes a retrofit, or becomes impossible. Method: pick the stage where the work or decision must still be made, not the stage where it would be convenient.

The eight stages.

Enforcement states. A row that is not simply assigned to a stage carries one of these instead, so that no row is silently dropped:

Why the stage matters more than the technology. Most of what makes a sleep space quiet, a focus space private, or a cinema accurate is decided at stages 1 through 3, before any technology is purchased. The UX for Spaces™ process stages Listen and Map, Define Experiences, and System Logic are aligned to run before and during Schematic Design for this reason. The seven process stages and the eight construction stages are different sequences: one describes the designer's work, the other describes the building's. A row's Impacts Build value is always a construction stage.

The project record is a set of documents that together carry the chain from need to tuning. For teams that work in commissioning terms, the Spatial Performance Brief plays the role of the Owner's Project Requirements and the behavior register plays the role of the Basis of Design: the first says what the building must do for the people in it, the second says how each discipline will produce that. Method: each document below is produced at the stage named, and later documents reference the rows in earlier ones by number rather than restating them.

The experiential sign-off. The method closes a project against the experience, not against a device count. The sign-off template lists twelve items in eight groups: Discovery (discovery summary); Experience (experience definitions by space and moment, emotional intent targets by space, movement and flow mapping); System (experience-to-system mapping, system behavior and logic, scene and automation definitions); Interaction (user interaction model); Wellness (calm, WELL, and biological alignment); Performance (performance criteria for light, sound, air, and thermal); Validation (edge cases and exceptions); and Sign-Off (final stakeholder approval). Each carries an owner, a status, and a signature.

Row numbering. Every row in the register keeps its framework category and number (for example, Restoration 6) plus a project suffix when a project-specific row is added. Drawings, commissioning entries, and post-occupancy log entries cite that number. This is what lets a builder ask "which row requires this" and get an answer.

Commissioning is not about devices turning on. It is about the space behaving correctly. Every scene, automation, and response is tested against the intended experience, and every numeric target in the register is measured at a defined position and recorded. Method: a target that was never measured is aspirational, and the commissioning record says so.

Three kinds of check. The method distinguishes them because they answer different questions and use different instruments.

What is measured, per space type. From the framework's commissioning rows. These are practice requirements within the method.

• Circadian and sleep spaces: M-EDI at the eye plane in daytime, evening, and night states, compared to the practice targets (250 lux or more daytime, 10 lux or less in the evening state, 1 lux or less on night pathways and in sleep spaces during sleep). Flicker modulation depth and frequency at all dimming levels including below 20 percent. Illuminance uniformity across the task plane, 0.6 minimum-to-average or better. Where daylight through low-emissivity glazing is relied on, spectral transmission at the glazing plane.
• Acoustic spaces: RT60 per space, background noise as NC at the defined position, STI at defined positions where intelligibility is a target, STC verification where isolation is critical. Acoustic commissioning is a required deliverable on any project where acoustic performance was specified.
• Air: CO2, PM2.5, VOC, and humidity readings from installed monitors validated against reference instruments; ventilation response confirmed at the register thresholds.
• Thermal: setpoint hold within the specified band in sleep, focus, and recovery spaces, with the fan-noise rule verified by sound measurement during a correction.
• Interface: every physical button traces to a justification; every touchscreen traces to a convergence of two or more functions that need a screen; every voice command has a physical fallback; idle screens are dark.

Targets and current conditions are kept apart. The commissioning record has two columns for every criterion: the target, and the measured value. A target that came from intent alone, with no site survey or baseline behind it, is marked provisional until it has been measured once in the built space. A target is never rewritten to match a measurement. If the measurement misses, the record says so and the deviation is routed to the row's intervention depth: a programming fix, a device change, or, where the miss is architectural or mechanical, a documented limitation with the reason.

Mechanical transitions in sleep spaces are a separate requirement. A mechanical system can meet its steady-state noise target and still wake someone when it starts, stops, or changes capacity, because sleep is disturbed by the change above background more than by the level. The method's evidence review, Human Performance as a Building Design Requirement (see References), sets this out from the polysomnographic literature and defines the requirement the method applies. For every sleep space the criterion schedule carries three acoustic limits, not one: the steady background, the absolute event maximum, and the permitted rise above the preceding stable background, with the sampling interval, event window, and instrument time weighting stated so that unlike measurements are not subtracted. The acoustician and mechanical engineer enter the numbers before schematic sign-off; an unassigned number is an open design item, not permission to skip it. The acceptance test is run at each pillow position in the agreed sleep configuration (doors, windows, and shades as they will be), synchronized with equipment-state logs, exercising starts, stops, minimum-load operation, speed and stage changes, dampers, pumps, and defrost where applicable, with representative overnight logging to catch repeated or combined events a short demonstration misses. Raising masking, reducing required airflow, or disabling necessary operation does not close the requirement.

Every numeric criterion carries eight fields. Purpose (the activity and need it serves); location and receiver (the position, eye or ear height); operating condition (occupancy, doors, windows, shades, equipment, time mode); criterion (metric, units, target, tolerance, permitted excursion); basis (code, standard, research recommendation, or explicit project preference); test (instrument, method, duration, averaging); accountability (design owner, installer, verifier, and who corrects); operation (maintenance, fault response, retest, override). A number without these fields is not yet a requirement.

Three result categories stay separate. A delivered physical condition (airflow, light at the eye, sound spectrum) establishes whether the requirement was met under the tested conditions. An occupant's reported experience (comfort, distraction, reported sleep) is recorded with its context and never merged with the physical result. A clinical or physiological outcome (sleep stages, a diagnosis, a validated biomarker) is established only by a study designed to establish it, and a consumer wearable does not qualify. The commissioning record and the post-occupancy log keep the three in separate columns.

Two technical rules that come up at commissioning.

• Masking is verified as a covering layer, never as a fix for an excessive floor. The masking level is set at or below the space's ceiling with the tuned spectrum, and recorded. If a space's measured background noise already exceeds its NC target, the record flags the source (most often HVAC) for mechanical correction and notes the residual level. Raising masking on top of it would raise the floor further.
• CCT is not accepted as evidence of biological light performance. Two luminaires at the same correlated color temperature can produce materially different melanopic output. The record uses measured M-EDI from a spectrometer or a melanopic-capable meter, or manufacturer SPD data converted through M-DER, never CCT alone.

Once the space is occupied, refinement begins. Real life introduces nuance the register could not know: a wake time that moves, a guest who reads later than the wind-down assumed, a sensor whose field of view catches a hallway. The UX for Spaces™ process calls this stage Tune in Real Use.

How tuning is done. Method:

1. A report from occupancy is matched to the row or rows that govern it. "The hallway light wakes me" maps to the night pathway row, the sleep space spill criterion, and possibly the sensor placement row.
2. The adjustment is made at the shallowest intervention depth that can resolve it. Programming before device, device before infrastructure.
3. The change is logged against the row with the date, the reason, and the new value. The register is updated so the record and the building agree.
4. If the shallowest depth cannot resolve it, the report is escalated with the row number and the depth it would require, so the owner can decide with the cost in view.

Drift. Buildings do not stay commissioned. LED output depreciates and phosphors shift spectrum. Sensors drift. Filters load and fan noise rises. Furniture, rugs, and curtains change absorption and reverberation. Scene logic drifts from intent after user modifications. Drift is cumulative across systems, and it is silent unless measured.

Annual recommissioning. The method's requirement on any project under ongoing care: once a year, the commissioning measurements are repeated at the same positions as the original record and compared to baseline. The scope covers M-EDI, flicker, and uniformity in circadian spaces; NC and RT60 in acoustic spaces; HVAC static pressure where fan noise governs an NC target; calibration of lux, CO2, occupancy, and temperature sensors; shade motor speed, noise, and limit behavior; and scene verification against the register. Deviations are corrected and documented. The building does not silently degrade.

What tuning is not. It is not a way to reach a target that construction missed. If the sleep space's partition transmits the neighbor's television, tuning can change scenes and masking, and the log will record that the isolation target remains unmet and why.

The intent belongs to the owner. The register is the owner's record of what the building was designed to do for the people in it, and it is handed over at sign-off in a form the owner, a future operator, or a future integrator can read without the original designer present.

Roles. Method:

Support. The UX for Spaces™ process names the final stage Support and Evolve: the way a space is used shifts, expectations change, and the system is kept aligned with real life rather than frozen at the original design. On projects under Blend's care this is delivered through a membership program with scheduled on-site visits, remote support, proactive monitoring where the tier includes it, and programming adjustments within a defined window after any new installation. The register and post-occupancy log are the working documents for every visit.

Handover to someone else. A record written in framework rows can be read by another firm. The method requires handover of the register, conflict register, commissioning record, and post-occupancy log in full at the end of any engagement, so that the owner is never dependent on one company to know what their building was designed to do.

A hotel guest arrives at 1 pm off an overnight flight and needs to sleep until evening. Housekeeping is working the floor. The guest in the next key has the television on. The property needs the space back in service by 6 pm for the evening. This example is illustrative and is built from the hospitality rows in the framework.

The study example had one person and one boundary. This one has four classes of user, an operator, a business schedule, and a property that will be verified key by key. The method is the same; the register looks different.

Human requirements. The guest needs dark, quiet, and a stable temperature at a time of day the building would normally treat as midday. Housekeeping needs to know the guest is resting without knocking. Engineering needs to reach HVAC diagnostics without entering. The property needs the space to reset for evening service without a manual walkthrough. The adjacent guest needs to watch television without being told to turn it down.

Criteria. In the guest space while rest is active: light at the pillow under 1 lux M-EDI; background noise at or below the space's NC target, with intermittent intrusion from the corridor and the adjacent key below the level that carries into the space; thermal setpoint held within a 1 degree band with no audible fan speed change. Between keys: partition and door assemblies that reach the isolation target from the adjacent television at normal viewing volume. In the corridor: a status the staff can read without entering.

Coordinated design and behavior.

• Arrival. A check-in credential event moves the space from standby to an arrival state: welcome-level light, temperature at the guest's preference or property default, audio off, blackout open. Recovery posture is the default: warm light, thermal precision, quiet.
• Rest override. A single legible control at the bedside puts the space into rest at any hour. Circadian logic suspends for this space. Blackout closes regardless of exterior conditions. Lighting drops to near zero. Temperature holds at the sleep setpoint. Masking activates at its tuned level. The space does not release until the guest releases it. A 1 pm rest is treated exactly like a 1 am one.
• Staff roles. Housekeeping, engineering, and management each have role-specific access and system visibility. A housekeeper cannot override a guest's rest scene. An engineer can read HVAC diagnostics for the key without entering it. The corridor indicator at the door shows the rest state as a peripheral status, not a message.
• Night path. If the guest gets up, path lighting to the bathroom activates below 10 lux, warm, from low sources, with no overhead trigger; the exhaust fan stays at minimum.
• Adjacency. Isolation between keys is specified at Construction Documents to the target that keeps a television at normal volume from carrying. Door assemblies get seals and closers rated for the corridor. This is where the adjacent guest's requirement is satisfied, and it is the one item on this list that no programming can substitute for.
• Reset. When the guest releases rest, or at checkout, the space returns to the evening or vacancy posture: HVAC pulls back or restages, lighting to the period scene, vacancy logged, housekeeping notified where applicable.

Trade and drawing responsibility. Partition and door assemblies: architect and acoustic consultant, Construction Documents. Bedside control, corridor indicator, sensors, and masking device: technology drawings at rough-in, trim at MEP finish. Role-based access: programming, with the role definitions supplied by the operator as an operational policy row. HVAC quiet-correction rule: mechanical design at Schematic / DD plus programming. Housekeeping rule for a rest-active key: operational policy, owned by the operator.

Verification, per key. Isolation between keys measured with the adjacent television at a defined level. M-EDI at the pillow in rest state with corridor doors open and closed. NC at the pillow with HVAC in a correction. Functional verification of the arrival state, rest override, corridor indicator, role restrictions (a housekeeping credential attempting to change a rest-active scene is refused), night path, and reset. Recorded per key, because a property is only as good as its worst key.

Tuning. Guest feedback and staff reports map to rows. A recurring report of corridor noise in a particular key routes first to door seal adjustment, then to the masking level for that key, and, if the record shows the isolation target was never met, to a documented limitation for that key with the assembly change it would take.

Operational data. A property running this register can produce a verified performance summary for a key or a floor: air quality, thermal hold, and acoustic measurements over time. That is a report of measured conditions. It is not a certification and not a health claim.

What this does not promise. The guest may still wake. The method removes the causes the building controls: daylight through a gap, a corridor door that slams, a housekeeper's knock, a fan that cycles. It commits the property to conditions per key and gives the operator a record to manage against.

A tag on a row says which external framework the row's behavior relates to. It routes evidence and coordination: on a project pursuing WELL, LEED, or Calm Tech Certified for Spaces, the tagged register shows the consultant where each behavior and its evidence live. As with those standards generally, certification of a project runs through the program's own process with a certified designer on the project.

The eight CTI principles as the framework uses them. This is the method's application of the Calm Technology principles to spaces. The eight principles were created by Amber Case, building on research by Mark Weiser and others at Xerox PARC in the 1990s; the Institute publishes them at calmtech.institute/calm-tech-principles, where the Institute defines calm technology as a process for designing technology that works with human attention instead of against it. The principle names below are the method's working phrasing for spaces, and the criteria say when a row earns the tag.

Blend and the Calm Tech Institute. Blend is a strategic partner of the Calm Tech Institute and co-presented Calm Tech Certified for Spaces with the Institute's founder at CEDIA Expo 2026. That relationship is why CTI tags are native to the framework. CTI tags roll up to a calm-technology alignment score for the project, and a project designed to the framework arrives at Calm Tech Certified for Spaces with its evidence already organized. As with the other standards, project certification requires a certified designer on the project.

Metrics the framework adopts from external standards. M-EDI and M-DER from CIE S 026:2018. Flicker limits from IEEE 1789-2015. Color rendition by the named specifications of ANSI/IES TM-30-20, Annex E, Table E-2, with CRI 90 kept only as a submittal floor where a manufacturer publishes nothing else. Background noise as NC per ANSI/ASA S12.2-2019. The daytime, evening, and night M-EDI practice targets from the 2022 consensus recommendations published in PLOS Biology (Brown et al.). Each is listed in References.

Color rendition by space intent. TM-30 defines three design intents, Fidelity (F), Preference (P), and Vividness (V), each at three nested priority levels with Level 1 strictest. The framework assigns a named specification to each space intent rather than inventing its own thresholds. Rf,h1 is local fidelity for reds and does the work R9 did in the CRI system; Rcs,h1 is the chroma shift of reds, and a small positive value is what makes skin and food read well. Criteria are evaluated at the space's design illuminance, since TM-30 assumes 200 to 700 lux in a polychromatic environment.

The three templates below are the minimum record for a space. Copy them as they are. The filled example that follows is the study from the work and focus intention at the start of this document, written as a project record would write it. Names, values, and row numbers in the filled example are illustrative.

Blank: Spatial Performance Brief

Performance targets are chosen from the 28-term vocabulary, comma-separated. Notes carry the nuance the terms cannot.

Blank: Behavior row

Blank: Conflict row

Filled example: a study designed for focus. Illustrative.

Spatial Performance Brief

Behavior rows

Conflict row

Post-occupancy log entry (illustrative): 2026-11-12. Row P-01. Occupant found the masking audible during quiet reading. Level lowered one step; STI at the desk re-checked with the great room television at normal volume and found within target. Logged by the designer.

Two things are versioned separately: the framework and each project record. Method.

The framework. The UX for Spaces™ Framework carries a version number (currently 15.5, dated 2026-09-20). A version is a specific set of categories, rows, tags, operating principles, specification gates, and the construction map. A change to any of those produces a new version. Row IDs are stable within a version; a row that is removed keeps its number retired rather than reassigned, so older project records still resolve.

The project record. Each project's brief, register, conflict register, and commissioning record state the framework version they were built from. A project does not migrate to a newer framework version automatically. When a project is revisited under a newer version, the differences are reviewed row by row and the record states which version governs each row. Post-occupancy log entries reference the row ID and the record version in force at the time.

This document. This document is published with a version and a date. It describes the framework version named in its text. When the framework changes in a way that alters the method described here, the document is revised and the revision is dated. Earlier versions remain available so that citations resolve.

What a version number does not mean. It does not mean an external body reviewed the change. It records that the author changed the method and when.

Provenance. This document was prepared from the v15.4 workbooks and revised against v15.5, the UX for Spaces™ process description, the alignment and construction package templates, the experiential sign-off template, and the evidence validation notes. Where a value is cited from a secondary source, the References say so.

So that the rest of the document can be weighed, here is what the method claims, stated as commitments.

• Authorship. UX for Spaces™ and its framework were authored by Jimmy Powers. This document states them as a method; Blend Technology holds the trademark.
• Conditions. The method commits a project to measurable conditions in defined spaces, verified at commissioning and recorded. What a person does in those conditions, sleep, focus, recover, is theirs; the method's work is the conditions.
• Evidence. Every numeric target is a practice target. Where it adopts a published value, the source is named in References; where no published criterion exists, the value is labeled as the method's own. Claims stay within the cited research, and the evidence review Human Performance as a Building Design Requirement (September 2026) grades that research as established, supported with limits, or emerging.
• Alignment tags route evidence. A WELL, LEED, or CTI tag points the consultant to the behavior and its evidence; certification under each program follows that program's own process with a certified designer on the project.
• Light is specified by spectrum and measured at the eye. Melanopic response depends on spectral composition, intensity at the eye, timing, duration, and the person, and two sources at the same CCT can differ materially, so the method requires SPD data or measured M-EDI.
• Readings carry their context. A CO2, particulate, lux, or sound value is recorded with its position and conditions, and installed sensors are validated against reference instruments before behaviors are trusted to them.
• Separation and capacity are built. Behavior logic cannot substitute for physical separation, mechanical capacity, or a partition that was never built; the conflict register says which resolutions are rules and which are drawings.
• Present practice and future software are kept apart. The services described here (brief, register, coordination, construction package, commissioning, care) are current practice. Any capability of the software Blend is building to carry the method that is not described here as current is in development.
• Examples are illustrative. The rest and focus intentions, the study record, and the hotel guest are constructed to show the method.

External references cited in this document. Each is a published standard, metric definition, or peer-reviewed study. The method's own documents are listed separately.

Light

• Brown TM, Brainard GC, Cajochen C, et al. Recommendations for daytime, evening, and nighttime indoor light exposure to best support physiology, sleep, and wakefulness in healthy adults. PLOS Biology. 2022;20(3):e3001571. doi:10.1371/journal.pbio.3001571. Source of the practice targets of 250 lux M-EDI or more in daytime, 10 lux or less in the evening, and 1 lux or less during sleep.
• CIE S 026:2018. CIE System for Metrology of Optical Radiation for ipRGC-Influenced Responses to Light. International Commission on Illumination. cie.co.at. Defines melanopic EDI and the melanopic daylight efficacy ratio.
• IEEE Std 1789-2015. IEEE Recommended Practices for Modulating Current in High-Brightness LEDs for Mitigating Health Risks to Viewers. standards.ieee.org. Basis for the flicker specification gate.
• ANSI/IES TM-30-20, with Errata 1 (May 7, 2021). IES Method for Evaluating Light Source Color Rendition. An American National Standard. Illuminating Engineering Society. store.ies.org. Annex E, Table E-2 supplies the F, P, and V specifications the framework assigns by space intent.
• ANSI/IES RP-29-25. Recommended Practice: Lighting Hospital and Healthcare Facilities. An American National Standard. Illuminating Engineering Society, 2025. store.ies.org. Table A-1 (patient room, patient zone) and its notes 4, 27, and 28 supply the recovery-space light values; values are adjusted for age 65 and older.

Sound

• ANSI/ASA S12.2-2019. Criteria for Evaluating Room Noise. Acoustical Society of America; approved May 21, 2019. asastore.aip.org. Basis for stating background noise as NC; also carries the A-weighted survey method and RNC curves.
• Basner M, Babisch W, Davis A, et al. Auditory and non-auditory effects of noise on health. The Lancet. 2014;383(9925):1325-1332. doi:10.1016/S0140-6736(13)61613-X. Review of health effects of environmental noise, including sleep.
• World Health Organization. Guidelines for Community Noise. 1999. who.int. Table 4.1 and section 4.3.3: bedrooms 30 dB LAeq and 45 dB LAmax at night; hospital ward rooms 30 dB LAeq at all hours and 40 dB LAmax at night; treatment and observation areas 35 dB LAeq. Anchors the masking ceilings and the recovery sound criteria.
• World Health Organization. Environmental Noise Guidelines for the European Region. 2018. who.int. Source-specific night-noise recommendations that update the 1999 guidance for transportation sources.
• FGI Guidelines for Design and Construction of Health Care Facilities, 2010. Facility Guidelines Institute. fgiguidelines.org. Table 1.2-2 (patient rooms NC/RC(N) 30 to 40, 35 to 45 dBA; NICU 25 to 35), Table 1.2-3 (minimum sound isolation between enclosed rooms; note 5 caps electronic masking at 48 dBA), and Table 1.2-4 (speech privacy: normal STI 0.19 or less, confidential 0.12 or less). A later edition exists; values are checked against the current edition at project start.

Air

• Allen JG, MacNaughton P, Satish U, Santanam S, Vallarino J, Spengler JD. Associations of cognitive function scores with carbon dioxide, ventilation, and volatile organic compound exposures in office workers: a controlled exposure study of green and conventional office environments. Environmental Health Perspectives. 2016;124(6):805-812. doi:10.1289/ehp.1510037. Basis for treating CO2 and VOC levels as cognition-relevant conditions in focus and learning spaces.
• ANSI/ASHRAE/ASHE Standard 170-2021. Ventilation of Health Care Facilities. ASHRAE. ashrae.org. Table 7-1, Patient room: 2 ach outdoor air minimum, 4 ach total minimum, MERV-14 minimum filter efficiency, relative humidity maximum 60 percent, design temperature 70 to 75 F (21 to 24 C).
• World Health Organization. WHO Global Air Quality Guidelines. 2021. who.int. Particulate guideline values referenced for recovery spaces.

Thermal

• ANSI/ASHRAE Standard 55. Thermal Environmental Conditions for Human Occupancy. ASHRAE. ashrae.org. The humidity ratio limit of 0.012 (dew point 16.8 C) is cited from a published summary of the 2010 edition, section 5.2.2. The PMV limit of plus or minus 0.5 and the 80 percent satisfaction criterion are cited from secondary sources, Wikipedia: ASHRAE 55 and Wikipedia: Thermal comfort, read 2026-09-20; the 10 percent PPD figure follows from Fanger's PPD relationship at that PMV.

Sleep and mechanical transitions (cited through the method's evidence review, Human Performance as a Building Design Requirement)

• Basner M, Müller U, Elmenhorst EM. Single and combined effects of air, road, and rail traffic noise on sleep and recuperation. Sleep. 2011;34(1):11-23. doi:10.1093/sleep/34.1.11. Polysomnographic study of transportation noise; event timing and characteristics matter more than the average level.
• Stanchina ML, Abu-Hijleh M, Chaudhry BK, Carlisle CC, Millman RP. The influence of white noise on sleep in subjects exposed to ICU noise. Sleep Medicine. 2005;6(5):423-428. doi:10.1016/j.sleep.2004.12.004. Source of the observation that the change above background, not the peak alone, predicts arousal; the roughly 18 dB figure is an observation, not a permissible excursion.
• Basner M, Smith MG, Cordoza M, et al. Efficacy of pink noise and earplugs for mitigating the effects of intermittent environmental noise exposure on sleep. Sleep. 2026;49(5):zsag001. doi:10.1093/sleep/zsag001. Polysomnographic trial in healthy adults; the tested pink noise reduced REM sleep and altered sleep structure.
• Zhang X, Wargocki P, Lian Z, Thyregod C. Effects of exposure to carbon dioxide and bioeffluents on perceived air quality, self-assessed acute health symptoms, and cognitive performance. Indoor Air. 2017;27(1):47-64. doi:10.1111/ina.12284. Controlled exposure to pure CO2 versus accumulated bioeffluents; symptoms and task changes followed bioeffluents, not CO2 alone.
• ASHRAE. Position Document on Indoor Carbon Dioxide. Approved February 12, 2025; expires February 12, 2028. ashrae.org. Evidence for direct health effects of CO2 at common indoor concentrations is inconsistent and does not support a universal limit; CO2 is a ventilation proxy.
• ASHRAE. 2023 ASHRAE Handbook, HVAC Applications. Chapter 49, Noise and Vibration Control. ashrae.org. Engineering guidance for background noise; transition limits need a documented project basis.

Frameworks referenced by alignment tags

• International WELL Building Institute. WELL Building Standard v2. v2.wellcertified.com.
• U.S. Green Building Council. LEED v5. usgbc.org.
• Calm Tech Institute. Principles of calm technology and the Calm Tech Certified program. calmtech.institute.
• Case A. Calm Technology: Principles and Patterns for Non-Intrusive Design. O'Reilly Media; 2015.
• Weiser M, Brown JS. The Coming Age of Calm Technology. Xerox PARC; 1996.

UX for Spaces™ documents

• UX for Spaces™ Framework v15.5, residential and commercial workbooks (2026-09-20): Spatial Performance Brief; Standards, Pillars and Build Map; seventeen category sheets; Conflict Resolution.
• UX for Spaces™ Process 2026.
• Calm / WELL / Biological Alignment Documentation Template.
• Technology Construction Package (introduction).
• Experiential Sign-Off template.
• Before You Build: Evidence-Based Validation of Key Claims, v2.
• Human Performance as a Building Design Requirement. Research paper for owners, architects, interior designers, and engineering teams, prepared for Blend Technology, 5 September 2026. Evidence review and design proposal; source of the mechanical-transition requirement, the eight-field criterion schedule, and the three result categories.

UX for Spaces™ is a design method authored by Jimmy Powers, Principal of Blend Technology. Jimmy is an acoustic consultant, a certified UX designer, and has designed automated technology systems for 25 years. Blend Technology is a Strategic Partner and Recognized Partner of the Calm Tech Institute.

UX for Spaces™ is a trademark of Blend Technology. Trademark pending, United States Patent and Trademark Office, application serial number 99789703. Application status at USPTO TSDR.

© 2026 Blend Technology. All rights reserved.