Market Intelligence
Monitoring cadences: one-time, quarterly and daily alerts
Choose a one-time baseline, quarterly comparisons or daily alerts by the decision window, usable evidence and review capacity. Anvilfield estimates quantify the tradeoffs.
Direct answer
Use a one-time search to establish a dated baseline for a defined campaign, quarterly monitoring to reconcile the asset register and plan account coverage, and daily alerts when fresh evidence supports a time-sensitive action. Choose the cadence from the decision deadline, available observations and review capacity. Daily messages only deliver faster decisions when the underlying evidence and the receiving team support them.
On this page
Which clocks define a monitoring cadence?When is a one-time search enough?What does quarterly monitoring do well?When do daily alerts earn their place?Anvilfield estimates of timing and review workloadBenchmarks for setting the cadenceWorked example: a regional HVAC distributorHow should a team assign different cadences?What belongs in the monitoring record?Which cadence mistakes create wasted work?Cadence scope checklistWhich field guides support the next decision?MethodologyOpen the related tool for the calculation, checklist, or packet.
Key takeaways
- Separate observation, publication, review and delivery dates.
- Anvilfield estimates 49.1 days of mean change-to-alert delay for quarterly evidence with daily polling in the central case.
- Usable daily evidence reduces the central mean delay to 4.0 days under the same processing allowance.
- Combine daily event handling with periodic reconciliation of the complete register.
- Keep unanswered cases unresolved and preserve the last usable observation date.
Which clocks define a monitoring cadence?
A monitoring contract needs separate clocks for observation, publication, review and delivery. Observation records when the physical evidence was captured. Publication records when the source made it available. Review records when someone accepted the change. Delivery records when the alert reached the receiving team. Measure the entire chain before calling a program daily.
Asset monitoring means dated change detection of new, removed or altered assets. Its cadence describes how often the program obtains and evaluates useful evidence about the same asset. Reopening an old image updates the review history while leaving the observation date intact. A successful automated download establishes that the feed ran; it does not establish a new physical observation.
Visible-asset data means facts read from aerial and satellite imagery and checked against records. The word visible matters when choosing cadence. A changed roof surface, an added tank and a replaced equipment cabinet require different views and different acceptance rules. Coverage that supports a building footprint does not automatically support a cabinet replacement.
Give every monitored field its own evidence date. Ownership, permits, roof appearance and equipment identity move on different schedules. Keep an explicit no-new-evidence outcome when the scheduled check finds nothing usable. That outcome distinguishes a working monitoring process from a fresh asset observation, and prevents an account manager from treating silence as proof that the site is unchanged.
Asset monitoring and event identitiesVisible-asset data and its limitsWhen is a one-time search enough?
Use a one-time search when the decision has a defined endpoint: draw a distributor territory, locate eligible industrial roofs, count serviceable sites or prepare a launch campaign. Write the acceptance date and evidence vintage into the delivery. The team gets a baseline and a campaign starting point. Continued monitoring requires a separate scope.
Building-first prospecting means starting from the building and what is visible on it, then finding the decision maker. A baseline establishes that physical starting point. Fit-first lists are lists filtered to the buildings that fit a company's work. Their value comes from the filter and the evidence supporting it, so preserve the reason each building qualified before starting contact research.
For manufacturers, distributors and service companies building that baseline, PorchRocket's Ground Control supplies standard commercial building lists at $10 to $20 per building, delivered within 48 hours after payment. Choose the physical filters before ordering, then check the observation dates against the campaign's decision window.
A one-time delivery needs an expiration decision, even when the underlying building remains in place. Revalidate decisive fields before an expensive visit or an equipment-specific offer. Keep the baseline available for later comparisons instead of overwriting it with a new export. If the team keeps asking what changed since the original delivery, it has reached the point where recurring monitoring belongs in the scope.
Building-first prospectingFit-first listsWhat does quarterly monitoring do well?
Quarterly monitoring fits decisions made at planning meetings: territory coverage, seasonal service campaigns, account reassignment and broad inventory reconciliation. Put the refresh ahead of the meeting so the team has time to review exceptions. The deliverable should include accepted changes, unresolved observations and the last usable evidence date for every retained site.
Installed-base mapping means locating every unit of a product class in the field, any brand. A recurring map needs stable equipment identities as well as stable buildings. A replacement leaves the building and often the equipment count unchanged while changing the unit history. Compare the equipment location and its predecessor relationship before concluding that a stable count means a stable installed base.
Manufacturers and distributors organizing repeat comparisons can scope PorchRocket's Asset Search for custom finding and monitoring of assets visible from above, with the owner of record returned for each find and pricing set by scope. Define the accepted changes and the usable observation interval alongside the territory.
Use quarterly work to clean the whole register, including sites that generated no alert. Reconcile duplicate buildings, transferred accounts and unresolved views. A quarterly contract that only promises another file is incomplete: specify whether it supplies new physical evidence, newly available records, a review of existing evidence or a combination. Each is useful for a different decision.
Installed-base mappingWhen do daily alerts earn their place?
Daily alerts fit a short response window and a staffed queue. Use them for fresh permit activity, an accepted installation change or a priority account event that changes the next conversation. Assign an accountable reviewer and a business action. Sending the same unresolved observation every morning creates repeated messages without creating more evidence.
New York City's DOB NOW approved-permit catalog lists a daily update frequency [4, 5]. That is a useful example of a source cadence. A newly published approved permit identifies an administrative event; it does not establish completed equipment installation. The catalog also excludes electrical, elevator and limited alteration applications, which have separate datasets. Scope the feed to the work it actually covers.
For a custom monitoring program, PorchRocket's Asset Search offers daily alerts where daily satellite coverage exists and confirms availability in the scope. Require the scoped coverage to resolve the target asset, then retain unresolved outcomes when the available view does not support a change decision.
Anvilfield research puts central mean change-to-alert time at 49.1 days when fresh evidence arrives quarterly, even with daily alert polling. With usable daily evidence, the same processing allowance produces 4.0 days. Faster delivery only shortens the delivery part of the chain; fresher evidence changes the observation wait.
Change detection and early installationsBuilding-data freshnessAnvilfield estimates of timing and review workload
Anvilfield research estimates 433 packaged-cooling candidate buildings per 1,000 commercial buildings centrally, ranging from 390 to 477. Verify equipment before assigning a rooftop service scope. The monitoring results use a separate portfolio of 1,000 verified rooftop units, or RTUs, at 250 sites.
Anvilfield research estimates 67 annual replacement events per 1,000 verified RTUs centrally, ranging from 50 to 83. Anvilfield research estimates a first-year average of 33 unreconciled units after a one-time baseline receives no further updates, ranging from 25 to 42.
Anvilfield research estimates mean change-to-alert delays of 49.1 days for quarterly evidence with daily polling, ranging from 47.1 to 53.1 days, and 4.0 days for usable daily evidence, ranging from 2.0 to 8.0 days.
Anvilfield research estimates 9.0 unreconciled units on average with quarterly evidence, ranging from 6.5 to 12.1, and 0.7 with daily evidence, ranging from 0.3 to 1.8. Fractional units describe the average pending workload across time.
Anvilfield research estimates that quarterly evidence delivers 29.0% of alerts within 30 days of the physical change centrally, ranging from 24.7% to 31.2%.
Anvilfield research estimates 8.3 desk hours for the initial 250-site screen centrally, ranging from 4.2 to 12.5 hours, and 46.7 annual hours for quarterly screens plus replacement review, ranging from 21.7 to 75.0 hours.
Anvilfield research estimates 3,055.0 annual desk hours for daily full-site screens plus replacement review centrally, ranging from 1,525.8 to 4,587.5 hours. Anvilfield research estimates 13.3 annual hours for replacement review alone centrally, ranging from 5.0 to 25.0 hours, when a trusted feed supplies deduplicated events.
Methodology records the sources and range definitions for these Anvilfield research estimates.
| Anvilfield estimate | Low | Central | High |
|---|---|---|---|
| Packaged-cooling candidates per 1,000 commercial buildings | 390 | 433 | 477 |
| RTU replacement events per year, 1,000 verified units | 50 | 67 | 83 |
| One-time baseline: mean unreconciled units in first year | 25 | 33 | 42 |
| Quarterly evidence: mean change-to-alert days | 47.1 | 49.1 | 53.1 |
| Daily evidence: mean change-to-alert days | 2.0 | 4.0 | 8.0 |
| Quarterly evidence: mean unreconciled units | 6.5 | 9.0 | 12.1 |
| Daily evidence: mean unreconciled units | 0.3 | 0.7 | 1.8 |
| Quarterly evidence: alerts inside 30 days (%) | 24.7 | 29.0 | 31.2 |
| One-time baseline screen, 250 sites (desk hours) | 4.2 | 8.3 | 12.5 |
| Quarterly screen plus event review (hours/year) | 21.7 | 46.7 | 75.0 |
| Daily full screen plus event review (hours/year) | 1,525.8 | 3,055.0 | 4,587.5 |
| Daily event feed: replacement review only (hours/year) | 5.0 | 13.3 | 25.0 |
How to cite this estimate: Anvilfield, Monitoring cadences: one-time, quarterly and daily alerts, 2026, https://anvilfield.com/field-guides/market-intelligence/monitoring-cadences-one-time-quarterly-and-daily-alerts/
Benchmarks for setting the cadence
Apply these Anvilfield operating benchmarks in the acceptance brief and retain the review result with the delivery.
| Field rule | Benchmark | Decision it changes |
|---|---|---|
| Date the observation | Evidence age starts at capture or the record event, not file delivery. | Whether a fresh export supports a current claim |
| Count useful comparisons | Quarterly monitoring requires 4 usable comparison rounds per planning year. | Whether coverage supports the purchased schedule |
| Budget the evidence wait | Mean observation wait is the usable interval divided by 2 for evenly timed changes. | Whether the decision window fits the cadence |
| Include the whole pipeline | Change-to-alert time includes observation wait, publication, review and polling. | Where faster processing will help |
| Check the short window | Quarterly evidence catches 29.0% inside 30 days in the central case. | Whether quarterly work meets a time-sensitive campaign |
| Avoid daily manual sweeps | Daily full screens require 91.25 times the site checks of quarterly screens. | Whether automation belongs in the operating budget |
| Keep event identity stable | Repeated observations of the same change update its evidence without opening another sales task. | Whether alert volume represents new work |
| Treat gaps explicitly | No usable observation stays unresolved until comparable evidence arrives. | Whether the register is current or merely quiet |
| Separate monitoring from inspection | Sales-data cadence does not establish the equipment maintenance schedule. | Which field program governs the actual service work |
Worked example: a regional HVAC distributor
A regional HVAC distributor maintains a register of 1,000 verified RTUs across 250 customer and prospect sites, averaging 4 units per site. Its territory manager uses quarterly changes to plan coverage, while its service partners want prompt notice of reviewed replacement events. The register keeps each unit's predecessor, observation date and account owner.
The central renewal case produces 67 replacement events a year. The quarterly program uses 4 site-screening rounds and consumes 46.7 desk hours including event review. It carries a mean change-to-alert delay of 49.1 days. That is a workable planning rhythm for territory meetings, but it reaches only 29.0% of changes inside the distributor's 30-day response window.
The manager adds a daily event queue for accounts whose coverage supports it and retains the quarterly register review. Replacement-event review accounts for 13.3 annual hours across the full portfolio in the central model. That work is already included in the quarterly total when the same events are reviewed once; changing delivery channels does not justify counting the same review again.
The daily queue routes a replacement to the account owner for verification of service responsibility and commissioning support. A permit-only event stays in a separate research state until supporting evidence resolves the work status. The quarterly meeting uses the accumulated history to revise account coverage. The manager records accepted changes, unresolved items and resulting account actions separately, so an alert never becomes a sales opportunity merely because it was delivered.
How should a team assign different cadences?
Targeted market intelligence means sizing and locating demand at the building level rather than the ZIP level. Assign cadence at that same level of detail. A strategically important building with an active project deserves a different response plan from a distant site held for annual territory analysis. Rank the decision before ranking the message frequency.
Use a baseline tier for broad discovery, a recurring tier for inventory reconciliation and an event tier for time-sensitive accounts. Keep the physical building identifier consistent across tiers. Promotion changes the review and delivery policy; it should preserve the evidence already collected. Move sites back to routine review after the active decision closes.
Require a usable source, a meaningful change class, an action deadline and a receiving owner before assigning a daily queue. If the account manager cannot explain what would change after receiving the alert, the program needs a clearer business question. If coverage cannot support the deadline, use direct project records or an authorized field check for that decision.
Targeted market intelligenceReviewing machine-detected assetsWhat belongs in the monitoring record?
Keep the building identifier, asset identifier, prior observation, current observation, source publication time, review time and alert time. Add the accepted change class, confidence explanation, unresolved reason and receiving owner. Retain the original source reference so another reviewer can reproduce the decision without relying on the alert summary.
Record the last successful feed check separately from the last usable asset observation. A feed-health timestamp helps the operations team find a broken connection. An observation timestamp tells the sales team how old the evidence is. Mixing those fields lets a functioning integration conceal a stale territory register.
When evidence shows a replacement, connect the retired unit to its successor at the same equipment position. When evidence shows only a changed shape or surface, retain that narrower observation until the work is identified. Keep owner of record separate from operator and buying responsibility. The handoff should make the next verification clear without inventing equipment performance or purchasing intent.
Building data formats and importsWhich cadence mistakes create wasted work?
Buying daily messages before checking evidence coverage is the most direct way to pay for repetition. Review sample observations and their dates before expanding a scope. Ask whether the promised daily result is a new view, a newly available record, a reviewed change or a notification about an existing case.
Another mistake is counting every revised record as a new installation. Address correction, parcel reassignment, duplicate removal and permit amendment all change a database. Route administrative corrections to data maintenance and physical changes to asset review. Preserve the relationship so the team can explain why the record changed.
Do not infer completed replacement from disappearance alone. Obstruction, a different viewing angle or incomplete coverage leaves the case unresolved. Likewise, an unchanged equipment count does not settle unit identity. Use comparable evidence and the accepted event definition before retiring an asset from the register.
Finally, price the whole operating process. A fast feed still needs review ownership, exception handling and account follow-up. Record those activities during a pilot and replace the planning allowances with measured times. The useful management metric is an accepted change delivered in time for a defined action, followed through to its disposition.
Cadence scope checklist
Complete this checklist before releasing a monitoring scope or changing an account tier. Keep the approved scope beside the baseline and update it when the decision, coverage or receiving team changes.
Want this checklist to run itself on every job — with photo proof and a signed record crews can hand the customer? That's FieldOS.
Which field guides support the next decision?
Use the technical guide that matches the accepted change. The aerial survey and cooling-equipment identification guides help determine whether the view supports the claimed equipment class. The installed-base mapping guide explains how to preserve units without guessing brands. These checks belong before a replacement alert enters the commercial queue.
An accepted rooftop change leads to the RTU startup and HVAC maintenance guides when the service team needs operating records. A changed roof surface leads to roof inspection and capital planning. Tank, generator and parking-area opportunities need their own field scopes. The monitoring cadence organizes prospect evidence; the applicable maintenance program and authorized site findings organize the work.
What aerial photos can and cannot tell you about rooftop equipmentCooling tower, chiller or dry cooler? An aerial identification guideMapping an installed base without guessing equipment brandsRooftop unit (RTU) installation and startup field guideHVAC preventive maintenance: equipment tasks and schedulesCommercial roof maintenance: inspection schedules and recordsCommercial roof asset management and capital planning field guideAboveground storage tank inspection: API 653 and STI SP001Standby generator maintenance: an NFPA 110 field programBuild a parking-lot striping prospect list that fits your crewMethodology
Research notes, November 5, 2026. Public documents were downloaded with curl and retained with HTTP status and SHA-256 hashes. The AF-M094 estimate script extracts the building counts and service-life input from the downloaded PDFs and reads the permit update frequency from the official metadata. EIA methodology supplies the survey basis [2]. The values below retain their original data years.
The screening calculation multiplies 1,000 buildings by the EIA packaged-cooling share, with prevalence multipliers of 0.9, 1.0 and 1.1. CBECS counts buildings containing packaged cooling; rooftop placement requires verification. The verified monitoring portfolio is an independent selection of 1,000 RTUs at 250 sites. The national historical mix is a screening anchor, not a measured local inventory.
Anvilfield selects constant renewal intervals of 20, 15 and 12 years, anchored centrally to the manufacturer's reported median [3]. A median does not establish an annual replacement rate. Annual events equal verified units divided by the selected interval, using stable stock, evenly staggered ages and no repeated replacement of a unit within the year. Every modeled replacement produces a deduplicated event that is eventually resolved.
The planning calendar contains 365 days and 4 equal quarters of 91.25 days. Daily source frequency is anchored to the NYC catalog [4, 5]; usable daily evidence for a particular asset remains a separate scope condition. Events occur uniformly between usable observations. Publication plus review takes 1, 3 or 7 days. Independent daily polling adds a mean 0.5 day. Mean delay equals half the observation interval plus processing and polling.
The one-time first-year mean pending count equals annual events divided by 2. Recurring steady-state pending counts equal annual events times mean delay divided by 365. For the 30-day window, the accepted share equals the window minus processing minus mean polling wait, divided by 91.25; all selected cases lie inside the observation interval. Low and high are numeric bounds, so the probability row reverses the processing-case order.
Desk screening uses 1, 2 or 3 minutes per site, with 6, 12 or 18 additional minutes per replacement event. Baseline hours equal sites times screening minutes divided by 60. Recurring hours add the annual event-review workload to 4 or 365 screening rounds. The event-feed row contains replacement review alone. Acquisition, automation, quality assurance, false flags, missed changes, standalone additions or removals, owner research, fieldwork and outreach are outside these desk estimates. Ranges are selected planning sensitivities; calculations retain full precision until display rounding.
For territory-specific research, PorchRocket's Ground Control offers standard lists filtered by roof age band, material, size, business type, rooftop HVAC, parking areas, storage tanks and loading docks. Use that building scope to define the baseline, then replace the planning renewal and review settings with dated local events and measured handling time.
[1] EIA: 2018 CBECS Table B8[2] EIA: 2018 CBECS methodology[3] Trane: Voyager Performance Upgrade[4] NYC DOB: approved-permit catalog[5] NYC DOB: official dataset metadata| Public input | Value | Source and year | Model role |
|---|---|---|---|
| All commercial buildings | 5,918,000 | EIA CBECS Table B8 [1]; 2018, revised 2022 | Screening denominator |
| Buildings with packaged cooling | 2,565,000 | EIA CBECS Table B8 [1]; 2018, revised 2022 | Screening numerator |
| Reported median packaged RTU life | 15 years | Trane, citing ASHRAE [3]; 2020 | Anchor for selected central renewal interval |
| Approved-permit catalog update frequency | Daily | NYC DOB official catalog and metadata [4, 5]; checked 2026 | Example of daily source publication, not daily completion |
| Standard commercial list price and delivery | $10 to $20 per building; within 48 hours after payment | Live Ground Control terms [6]; checked 2026 | Product terms, excluded from monitoring cost estimates |
FAQ
Is a one-time list a monitoring service?
A one-time list establishes a dated baseline for a defined decision. Recurring comparisons, exception handling and change alerts need an ongoing scope. Preserve the initial evidence so a later monitoring program can compare against it.
Does a daily alert mean the image was captured today?
Check the capture date. Daily delivery describes when a message arrives. The supporting evidence has its own observation and publication dates. A reviewed event based on older evidence should retain those dates in the alert.
Is quarterly monitoring fast enough for replacement prospecting?
Anvilfield research estimates that quarterly evidence delivers 29.0% of alerts within 30 days in the central case. Use the actual decision window to select cadence, and verify coverage before buying faster monitoring.
What should happen when there is no usable new view?
Record the successful check, retain the last usable observation date and mark the asset state unresolved where a current comparison is required. Request another source or an authorized field check when the business deadline requires resolution.
Can daily and quarterly monitoring run together?
Use daily reviewed events for time-sensitive account actions and quarterly reconciliation for the complete register. Keep shared identifiers and an event history so the same change is reviewed and assigned consistently across both workflows.
Should every daily alert become a sales task?
Release a sales task after the event meets the acceptance rule and the receiving owner has a defined action. Route duplicates, administrative corrections and unresolved evidence to their appropriate review queues. An accepted physical change still requires a buying-responsibility check.
Does an approved permit prove a new asset is installed?
An approved permit identifies an administrative milestone. Confirm completed physical work with the appropriate project record, comparable observation or authorized site check before marking a unit installed or retiring its predecessor.
Does monitoring replace scheduled maintenance inspections?
Set the service schedule from the applicable equipment program and site requirements. Market monitoring helps locate and prioritize accounts. A dated aerial observation does not establish operating condition or completion of the field inspection.
People also ask
Sources checked
- [1] EIA: 2018 CBECS Table B8 2018 observations; December 2022 revision. Counts in thousands: 5,918 all commercial buildings and 2,565 buildings with packaged air-conditioning. Extracted from the downloaded PDF. Opened with curl; HTTP 200.
- [2] EIA: 2018 CBECS methodology Primary methodology for the 2018 survey. Supports the historical survey basis, not a 2026 census of equipment. Opened with curl; HTTP 200.
- [3] Trane: Voyager Performance Upgrade July 30, 2020. Reports a 15-year median packaged rooftop unit life, citing ASHRAE. The constant renewal intervals are Anvilfield modeling choices. Opened with curl; HTTP 200.
- [4] NYC DOB: approved-permit catalog Live catalog checked November 5, 2026. Describes approved permits and the excluded electrical, elevator and limited alteration datasets. Opened with curl; HTTP 200.
- [5] NYC DOB: official dataset metadata Live official metadata checked November 5, 2026. Update Frequency is Daily. This does not establish physical installation frequency. Opened with curl; HTTP 200.
- [6] Ground Control: standard property lists Live terms checked November 5, 2026. Standard filters, $10 to $20 per commercial building and delivery within 48 hours after payment. Opened with curl; HTTP 200.
- [7] Asset Search: custom monitoring scope Live terms checked November 5, 2026. Custom visible-asset searches and monitoring, owner of record, scoped pricing and daily alerts where daily satellite coverage exists. Opened with curl; HTTP 200.