Housecall Pro · Test results and evidence

How We Tested Housecall Pro: Test Results and Evidence

Housecall Pro test results by scenario: each run on record with its timings and observations, what was not completed and why, and what remains untested.

On this page

This page is the evidence record behind the Housecall Pro review: each test run on record, what it measured, what was observed, and what was not completed and why. It is not a buyer assessment and carries no verdict, no plan comparison and no prices. For the assessment, read the review; for plans and costs, the Housecall Pro pricing guide; for the rules these tests follow, how we test and evaluate software.

Testing status

Testing status Testing in progress: 11 of 18 core scenarios complete Latest test

11 of 18 core workflow scenarios completed on a Housecall Pro trial account, in 22 recorded passes. Not completed: create recurring job (not run), configure reminder / follow-up (not run), configure review request (not completed), export / report data (not run), inspect mobile / responsive workflow (not run), inspect relevant integrations (not run) and inspect online booking where supported (not run).

Driver confirmation: the person who drove the sessions has confirmed the written observations on 22 of 28 recorded runs; 6 are pending. A pending confirmation withholds any statement of how the product felt to use; it does not change what a run did, which the recording and the action log establish.

Scenario coverage

Coverage of the core suite: 11 of 18 complete with a repeat pass · 1 not completed · 6 not run. Only a completed, successful repeat pass counts a scenario as complete; a first pass alone is learnability evidence.

  • A-01: The product requires a ten-digit phone; the fixture's seven-digit number was refused and is stored with an area code added, with the rest of the fixture as typed (runs 001 and 002).
  • A-05: The new-job form offers no field to link an existing estimate; the criterion's service half was met and both jobs carry the price-book line (runs 009 and 010).
  • A-07: Step 5 (a notification on assignment) is conditional; the assignment showed no notify option and nothing was triggered on camera, so it does not apply. The fixture technician was assignable while the invitation was pending, so no workaround was needed (runs 013 and 014).
  • A-08: The attached photo was a test image from the driver's Pictures folder rather than the fixture's grey placeholder file, in both passes (runs 023 and 024); the note is the fixture text, and the criterion (a note and an attached photo on the job) was met.
  • A-09: Runs 015 and 016 (first sitting) were left without sending and created no invoice record; the retest (runs 019 and 020) sent the invoice by e-mail to the synthetic non-deliverable address, which created the record, and completed the scenario. The first-sitting runs stay on record as not completed.
  • A-10: The payment was a manual cash record, the scenario's safe path: the product offers no sandbox or test payment mode, and card processing needs a connected bank (runs 025 and 026).
  • A-13: Not completed: the review-request setup asks for a Google or Facebook account connection before automatic requests can be configured, and the test rules do not allow connecting an outside account (runs 027 and 028); whether the wizard lets the connection steps be skipped was not tried.
  • A-14: Runs 017 and 018 (first sitting) had no invoice to show because none existed; the retest (runs 021 and 022) listed the jobs, the estimate and both sent invoices on the customer record and completed the scenario. The repeat pass opened the wrong one of the two identically named customers first.
Coverage record: steps logged, criteria met, runs and the latest pass for each scenario
Scenario Status Steps logged Criteria met Runs Latest
A-01 Complete 4 of 4 1 of 1 001, 002 Oct 11, 2026
A-02 Complete 4 of 4 1 of 1 003, 004 Oct 11, 2026
A-03 Complete 5 of 5 1 of 1 005, 006 Oct 11, 2026
A-04 Complete 4 of 4 1 of 1 007, 008 Oct 11, 2026
A-05 Complete 5 of 5 1 of 1 009, 010 Oct 11, 2026
A-06 Complete 6 of 6 1 of 1 011, 012 Oct 11, 2026
A-07 Complete 4 of 5 1 of 1 013, 014 Oct 11, 2026
A-08 Complete 5 of 5 1 of 1 023, 024 Oct 12, 2026
A-09 Complete 5 of 5 1 of 1 015, 016, 019, 020 Oct 12, 2026
A-10 Complete 4 of 4 1 of 1 025, 026 Oct 12, 2026
A-11 Not run — — — —
A-12 Not run — — — —
A-13 Not completed — — 027, 028 —
A-14 Complete 3 of 3 1 of 1 017, 018, 021, 022 Oct 12, 2026
A-15 Not run — — — —
A-16 Not run — — — —
A-17 Not run — — — —
A-18 Not run — — — —
What the counted repeat pass recorded, per scenario.

"Steps logged" and "criteria met" are what the counted repeat pass recorded against the scenario's step list and its success criteria; a step not logged is explained in the notes above, and a scenario stays complete only when every criterion was met. Run numbers are the last digits of the run ids in the registry.

How the tests were run

We ran each scenario on a Housecall Pro trial account ourselves, with a person at the keyboard in a normal browser. Housecall Pro’s terms leave automated access to the product a matter of the vendor’s permission, which we did not request, so no automation touched it. Each session was screen-recorded with a wall clock burned into the frame, and a timestamped action log was kept alongside; the timings on this page come from that clock, and the review of each recording is ours. The data entered is a synthetic fixture set: no real customer, property or payment detail was created or kept.

A scenario is run twice. The first-use pass measures how findable the path is for someone who has not used the product for that task, with help allowed and logged. The repeat pass measures the workflow once the path is known, on the exact fixture, and it is the pass that counts toward the testing status. A run that did not complete the fixture as specified stays on record as not completed: it feeds no result beyond the behaviour it exposed.

Three things are kept apart on this page: whether a run completed its scenario as specified (protocol), what the run observed (evidence), and whether the person who drove the session has confirmed the written observations (validation, which gates any statement of how the product felt to use, never a statement of what was done). A scenario completed with a recorded workaround counts as completed and says so; a scenario inspected without a repeat pass does not count.

Protocol on record for these runs

  • Account: a Housecall Pro trial started by the driver on October 11, 2026, with the feature set of Max; no plan was purchased; the trial banner stayed on screen through the first sitting and showed on the dashboard in the second.
  • Driver and instrumentation: one person at the keyboard in an ordinary desktop browser; a continuous screen recording with a wall clock burned into the frame, and a typed, timestamped action log.
  • Data: the synthetic fixture set, version chimney-v1.0; no real customer, property or payment detail.
  • Benchmark and catalog: benchmark version v1 of the core scenario catalog, A-01 to A-18.
  • Times: Eastern Daylight Time, the evenings of October 11 and 12, 2026.
  • Review of the evidence: the observations were written from the frames and the log on October 11 to 13, 2026; the driver’s confirmation state is printed under the testing status above.
  • Retest trigger: a change in a tested workflow, the benchmark window, or an open scenario; the two scenarios not completed in the first sitting were retested in the second sitting on the same trial, and the review-request scenario stays open.

Runs on record

Full run registry: 28 recorded runs
Run id Scenario Pass Completed Result Recording length Steps Help Admissibility
STS-TEST-HOUSECALLPRO-CORE-2026-001 A-01: Create customer first use (discovery) October 11, 2026 completed as specified 282 s 4 none First-use pass: learnability evidence, not a benchmark result on its own
STS-TEST-HOUSECALLPRO-CORE-2026-002 A-01: Create customer repeat (known path) October 11, 2026 completed as specified 173 s 4 none Admissible known pass
STS-TEST-HOUSECALLPRO-CORE-2026-003 A-02: Create service / item first use (discovery) October 11, 2026 completed as specified 811 s 4 none First-use pass: learnability evidence, not a benchmark result on its own
STS-TEST-HOUSECALLPRO-CORE-2026-004 A-02: Create service / item repeat (known path) October 11, 2026 completed as specified 130 s 4 none Admissible known pass
STS-TEST-HOUSECALLPRO-CORE-2026-005 A-03: Create estimate first use (discovery) October 11, 2026 completed as specified 205 s 5 none First-use pass: learnability evidence, not a benchmark result on its own
STS-TEST-HOUSECALLPRO-CORE-2026-006 A-03: Create estimate repeat (known path) October 11, 2026 completed as specified 151 s 5 none Admissible known pass
STS-TEST-HOUSECALLPRO-CORE-2026-007 A-04: Approve / convert estimate first use (discovery) October 11, 2026 completed as specified 183 s 4 none First-use pass: learnability evidence, not a benchmark result on its own
STS-TEST-HOUSECALLPRO-CORE-2026-008 A-04: Approve / convert estimate repeat (known path) October 11, 2026 completed as specified 137 s 4 none Admissible known pass
STS-TEST-HOUSECALLPRO-CORE-2026-009 A-05: Create job first use (discovery) October 11, 2026 completed as specified 331 s 5 none First-use pass: learnability evidence, not a benchmark result on its own
STS-TEST-HOUSECALLPRO-CORE-2026-010 A-05: Create job repeat (known path) October 11, 2026 completed as specified 145 s 5 none Admissible known pass
STS-TEST-HOUSECALLPRO-CORE-2026-011 A-06: Schedule / reschedule job first use (discovery) October 11, 2026 completed as specified 216 s 6 none First-use pass: learnability evidence, not a benchmark result on its own
STS-TEST-HOUSECALLPRO-CORE-2026-012 A-06: Schedule / reschedule job repeat (known path) October 11, 2026 completed as specified 123 s 6 none Admissible known pass
STS-TEST-HOUSECALLPRO-CORE-2026-013 A-07: Assign technician first use (discovery) October 11, 2026 completed as specified not measured 4 none First-use pass: learnability evidence, not a benchmark result on its own
STS-TEST-HOUSECALLPRO-CORE-2026-014 A-07: Assign technician repeat (known path) October 11, 2026 completed as specified 58 s 4 none Admissible known pass
STS-TEST-HOUSECALLPRO-CORE-2026-015 A-09: Create invoice first use (discovery) October 11, 2026 not completed as specified 255 s 4 none Not admissible as a benchmark result: the fixture was not completed as specified; its observations stand as recorded
STS-TEST-HOUSECALLPRO-CORE-2026-016 A-09: Create invoice repeat (known path) October 11, 2026 not completed as specified 146 s 4 none Not admissible as a benchmark result: the fixture was not completed as specified; its observations stand as recorded
STS-TEST-HOUSECALLPRO-CORE-2026-017 A-14: Find customer service history first use (discovery) October 11, 2026 not completed as specified 176 s 2 none Not admissible as a benchmark result: the fixture was not completed as specified; its observations stand as recorded
STS-TEST-HOUSECALLPRO-CORE-2026-018 A-14: Find customer service history repeat (known path) October 11, 2026 not completed as specified 114 s 2 none Not admissible as a benchmark result: the fixture was not completed as specified; its observations stand as recorded
STS-TEST-HOUSECALLPRO-CORE-2026-019 A-09: Create invoice first use (discovery) October 12, 2026 completed as specified 154 s 5 none First-use pass: learnability evidence, not a benchmark result on its own
STS-TEST-HOUSECALLPRO-CORE-2026-020 A-09: Create invoice repeat (known path) October 12, 2026 completed as specified 73 s 5 none Admissible known pass
STS-TEST-HOUSECALLPRO-CORE-2026-021 A-14: Find customer service history first use (discovery) October 12, 2026 completed as specified 64 s 3 none First-use pass: learnability evidence, not a benchmark result on its own
STS-TEST-HOUSECALLPRO-CORE-2026-022 A-14: Find customer service history repeat (known path) October 12, 2026 completed as specified 81 s 3 none Admissible known pass
STS-TEST-HOUSECALLPRO-CORE-2026-023 A-08: Add job notes / photos first use (discovery) October 12, 2026 completed as specified 144 s 5 yes, logged First-use pass: learnability evidence, not a benchmark result on its own
STS-TEST-HOUSECALLPRO-CORE-2026-024 A-08: Add job notes / photos repeat (known path) October 12, 2026 completed as specified 73 s 5 none Admissible known pass
STS-TEST-HOUSECALLPRO-CORE-2026-025 A-10: Record / attempt payment workflow first use (discovery) October 12, 2026 completed as specified 118 s 4 yes, logged First-use pass: learnability evidence, not a benchmark result on its own
STS-TEST-HOUSECALLPRO-CORE-2026-026 A-10: Record / attempt payment workflow repeat (known path) October 12, 2026 completed as specified 48 s 4 none Admissible known pass
STS-TEST-HOUSECALLPRO-CORE-2026-027 A-13: Configure review request first use (discovery) October 12, 2026 not completed as specified 138 s 1 yes, logged Not admissible as a benchmark result: the fixture was not completed as specified; its observations stand as recorded
STS-TEST-HOUSECALLPRO-CORE-2026-028 A-13: Configure review request repeat (known path) October 12, 2026 not completed as specified 48 s 1 none Not admissible as a benchmark result: the fixture was not completed as specified; its observations stand as recorded
Every run allocated for this product, from the run records.

Every run: UI-H, a person at the keyboard; benchmark v1. Recording length is the length of the screen recording, including setup before the task marker, never a task time; task timings are in the observations below. Only a completed, successful repeat pass with a person at the keyboard counts toward the testing status. "Admissibility" is about that benchmark: a run that is not admissible contributes no timing and no completed scenario, while what it observed is preserved and published as recorded.

Runs 001 to 018 come from the first sitting, recorded on October 11, 2026, and runs 019 to 028 from the second, recorded on October 12: each sitting is a continuous recording cut into its runs afterwards from the markers typed during it, so the run ids follow the evening’s order. Six runs are on record as not completed as specified. Runs 015 and 016 (create invoice) found that the invoice screen offers print, download and send and no save; both were left without sending, and no invoice record existed in the account afterwards; runs 017 and 018 (find customer service history) found the customer’s jobs and estimate on the customer record but no invoice, because none existed. Both scenarios were retested in the second sitting and completed: sending the invoice by e-mail to the synthetic test address is what created the record (runs 019 and 020), and the customer’s record then listed both invoices (runs 021 and 022). Runs 027 and 028 (configure review request) stopped at the setup wizard, which asks for a Google or Facebook account connection our rules do not allow. Two scenarios count with a recorded deviation: customer creation (runs 001 and 002), where Housecall Pro requires a ten-digit phone number so the fixture’s seven-digit number was stored with an area code added; and job notes and a photo (runs 023 and 024), where a test image from the driver’s computer was attached in place of the fixture’s placeholder file, with the note entered as written.

Observations by scenario

The scenarios follow the catalog order, A-01 to A-18, whatever order their runs were recorded in; each scenario and each run has a stable link, and the review’s findings point here.

The observations below were written from each recording and its action log at processing time. The person who drove the sessions confirms them run by run (how testing and verification work): confirmed on 22 of 28 recorded runs, pending on 6. Recorded runs include the ones not completed as specified. A pending confirmation does not change what a run did; it withholds any statement of how the product felt to use.

A-01: Create customer

STS-TEST-HOUSECALLPRO-CORE-2026-001 first-use pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • Adding a customer in Housecall Pro took two clicks from the dashboard: Customers in the top menu, then Add customer, which opens a single long form for contact details, address, notes and payment terms.
  • The mobile phone field accepts only a full 10-digit North American number; a 7-digit number is flagged as not valid until an area code is added.
  • The street field offers Google address suggestions as you type, and the city, state and zip can still be typed by hand when the address is not in Google's list.
  • Send notifications is switched on by default for a new customer, and the saved profile showed email marketing consent as opted in.
  • Saving took about five seconds to land on the new customer profile, and the customer appeared in the All Customers list a few seconds later.

Capabilities exercised: crm.

STS-TEST-HOUSECALLPRO-CORE-2026-002 repeat pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • Housecall Pro's Customers page has a Create customer button above the customer table; from the dashboard the new-customer form was two clicks away.
  • Filling the form took about a minute and a half, including about 20 seconds correcting the address after a Google address suggestion replaced the typed street with a real address in another state.
  • After the correction the saved address text was right, but the map on the customer profile still showed the suggested location.
  • Housecall Pro saved a second customer with the same name, email and phone without any duplicate warning; the two records were told apart only by the company field.
  • Saving took about two seconds to land on the new profile, and the customer list showed the new record straight away.

Capabilities exercised: crm.

A-02: Create service / item

STS-TEST-HOUSECALLPRO-CORE-2026-003 first-use pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • Housecall Pro calls a priced catalog item a "pricing form", kept under Price book alongside Services and Materials; on a first attempt the tester went into recurring service plans instead and spent about 8 minutes before finding Pricing forms.
  • Once on the right screen, filling the form took about two minutes: name, description, a fixed one-hour duration, and the $189.00 price entered on a quantity field row rather than in a single price box.
  • Saving goes through a preview of the line item before the final Save, and the new item appeared in the Pricing forms list straight away with a "Pricing form created" confirmation.
  • The pricing form had no category field, and the list shows item names only; the price is visible after opening the item.
  • No help documentation was opened and no plan restriction appeared on these screens during the trial.

STS-TEST-HOUSECALLPRO-CORE-2026-004 repeat pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • Housecall Pro keeps priced catalog items as "pricing forms" under Price book; reaching a new one took three clicks from the main navigation.
  • Filling the form (name, description, a fixed one-hour duration and the $189.00 price on a quantity field row) took a little over a minute.
  • Saving goes through a preview of the line item, and the item appeared in the Pricing forms list about ten seconds after starting the save, with a "Pricing form created" confirmation.
  • A second item with exactly the same name was accepted without any duplicate warning, and the list shows names only, so the two entries look identical until opened.
  • No help documentation was opened and no plan restriction appeared during this task on the trial.

A-03: Create estimate

STS-TEST-HOUSECALLPRO-CORE-2026-005 first-use pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • A new estimate opened from the Home page in one click, and the customer and the service were both added on that single form.
  • Picking the service from the price book brought in its price and duration, but set the quantity to zero, so the line read $0.00 until the quantity was raised to one.
  • The form's only button was Save estimate: the estimate saved as a draft without being sent, and its activity log recorded the creation and the $189.00 total.
  • The customer search showed name and address only, so two customers with the same name and address looked identical in the list.
  • Saving the finished form opened the saved estimate within a few seconds; the whole first attempt, including finding the way, took about three and a half minutes.

Capabilities exercised: estimating.

STS-TEST-HOUSECALLPRO-CORE-2026-006 repeat pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • On the repeat, a new estimate for the customer was open within about 15 seconds of starting, from a one-click shortcut on the Home page.
  • Adding the service from the price book again set the quantity to zero; saved as it was, the estimate total read $0.00 until the line was reopened and set to one, which brought it to $189.00.
  • Save estimate opened the saved estimate about three seconds after the click; it was saved as a draft without being sent, and the estimates list showed it as open at $189.00.
  • The whole repeat, from start to the estimate checked in the list, took about two and a half minutes, including the quantity correction.

Capabilities exercised: estimating.

A-04: Approve / convert estimate

STS-TEST-HOUSECALLPRO-CORE-2026-007 first-use pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • Approval is a step in the estimate's progress bar: clicking it offers Approve or Decline, and the confirmation box warns that approval cannot be undone.
  • The same confirmation box has a 'Copy to new job' option, so approving and converting the estimate into a job took a single confirm, and the new job opened about five seconds later.
  • The new job was created unscheduled, with no prompt to pick a date and no message sent to the customer.
  • Afterwards the estimate read 'Pro approved' with a link to the job, and the job listed the estimate as won and copied to it, with the conversion recorded in both activity feeds.
  • From the start of the task to seeing the estimate approved and converted took a little over two minutes on this first pass, including time spent exploring the estimate's menus and re-reading the task.

Capabilities exercised: estimating.

STS-TEST-HOUSECALLPRO-CORE-2026-008 repeat pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • On the repeat pass, approving the estimate took three clicks from its page: the Approval step, Approve, then the confirmation.
  • Converting the approved estimate took two more clicks (Copy to job, then Copy), and the new job opened about four seconds later, unscheduled and with no message sent to the customer.
  • The estimate then showed a 'Pro approved' status with a link to the new job, and the job listed the estimate as won and copied to it.
  • Right after approving, the estimate's header badge still read 'Unscheduled' while the progress bar already showed it approved; the header updated when the estimate was opened again.

Capabilities exercised: estimating.

A-05: Create job

STS-TEST-HOUSECALLPRO-CORE-2026-009 first-use pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • A new job was created from Housecall Pro's own new-job form, reached from the dashboard in two clicks, with the customer, a price-book service line, a date and a private note on one page.
  • The new-job form had no field for linking an existing estimate; the job carries the service line from the price book instead.
  • Adding the price-book service opened it with a quantity of zero, which had to be raised to one before the line showed its price.
  • The customer search showed two identically named customers with the same address and nothing to tell them apart.
  • Saving took a few seconds and opened the new job's page with its job number, status and scheduled window; the job then appeared in the jobs list with its date and amount.

Capabilities exercised: scheduling.

STS-TEST-HOUSECALLPRO-CORE-2026-010 repeat pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • Creating a job from Housecall Pro's own new-job form took a little over a minute from opening the form to the saved job, covering the customer, a price-book service line, a date and a private note.
  • The new-job form had no field for linking an existing estimate, and the saved job's estimates section stayed empty; the job carries the service line from the price book instead.
  • Adding the price-book service opened it with a quantity of zero, which had to be raised to one before the line showed its price.
  • Saving opened the new job's page with its job number, status and scheduled window, and the job's activity feed logged the schedule and the technician dispatch as separate entries.
  • The job then appeared in the jobs list with its status, scheduled date and amount.

Capabilities exercised: scheduling.

A-06: Schedule / reschedule job

STS-TEST-HOUSECALLPRO-CORE-2026-011 first-use pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • Scheduling and rescheduling run through the same control: the job page's Schedule button shows the current slot and opens a full-screen editor with separate start/end date and time fields beside a week calendar.
  • Setting the first slot took under half a minute once the editor was open, and the job page showed the saved time about a second after Save.
  • Moving the job to another day and time took under 20 seconds of editing; the end date followed the new start date automatically, and the job page showed the new slot right after saving.
  • The customer's Jobs list and the reopened job page both showed the new slot, and the notify-customer option was off by default, so nothing was sent to the customer.
  • Right after the reschedule, the product showed a referral offer popup over the job page, which had to be clicked away.

Capabilities exercised: scheduling.

STS-TEST-HOUSECALLPRO-CORE-2026-012 repeat pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • Scheduling a job and moving it to another slot both ran through the job page's Schedule button, which opens a full-screen editor with separate start/end date and time fields beside a week calendar.
  • Setting the first time slot took under half a minute once the editor was open, and the job page showed the saved slot about a second after Save.
  • The reschedule took about 10 seconds of editing: picking the new day moved both the start and end dates, and picking the new start time set the end time an hour later in the same step.
  • From opening the dashboard to seeing the rescheduled slot on the job took about a minute and a half, and reopening the editor afterwards showed the same new slot.
  • The notify-customer option was off by default, so neither change sent anything to the customer.

Capabilities exercised: scheduling.

A-07: Assign technician

STS-TEST-HOUSECALLPRO-CORE-2026-013 first-use pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • Assigning a technician happens in a small 'Edit team' dialog opened from a pencil icon next to the job's current assignee, where team members are ticked and saved.
  • A newly added technician who had not yet accepted the invitation could be selected and assigned right away, and the job then listed them as 'Assigned'.
  • Saving the assignment showed no confirmation message and no option to notify the technician; the change appeared on the job page and in its activity feed as a dispatch entry.
  • Adding the technician as a team member raised no request for payment or an extra seat on the trial.

Capabilities exercised: dispatch.

STS-TEST-HOUSECALLPRO-CORE-2026-014 repeat pass, October 11, 2026, completed as specified; observations confirmed by the driver.

  • Assigning the technician took a few seconds once the job was open: a pencil icon next to the job's current assignee opens an 'Edit team' dialog where team members are ticked and saved.
  • From the dashboard to the job listing the technician as assigned took under a minute.
  • A technician who had not yet accepted the invitation could be assigned, and the job then showed them as 'Assigned'.
  • Saving showed no confirmation message and offered no option to notify the technician; the job's activity feed recorded the change as a dispatch.

Capabilities exercised: dispatch.

A-08: Add job notes / photos

STS-TEST-HOUSECALLPRO-CORE-2026-023 first-use pass, October 12, 2026, completed as specified; observations confirmed by the driver.

  • Job notes and photos are added on the job page itself, in a private-notes panel and an attachments panel below the schedule and invoice sections; there is no separate documentation screen.
  • The note was committed as soon as it was entered and the photo as soon as it was chosen in the browser's file dialog; neither has a save button, and the job's activity feed logged each addition with the time.
  • Reopening the job from the customer's record showed the note text and the photo thumbnail in place, with the feed entries beside them.
  • The notes panel is private by name and offers no switch to share a note with the customer; the photo upload is a plain file attachment, with no in-product capture, annotation or report step on this path.
  • On this first pass the photo attached was a test image from the computer's Pictures folder rather than our grey placeholder file; the task still completed, and the substitution is recorded as a deviation from our fixture.

Capabilities exercised: photo documentation.

STS-TEST-HOUSECALLPRO-CORE-2026-024 repeat pass, October 12, 2026, completed as specified; observations confirmed by the driver.

  • On the repeat pass the job was open eight seconds after starting, and the note and the photo were added in the same two panels on the job page as the first time.
  • Both items were committed as they were entered, with the job's activity feed logging each addition; the photo tile showed a grey placeholder for a few seconds before the thumbnail appeared.
  • Reopening the job from the customer's record showed the note and the photo in place with the feed entries beside them.
  • The same test image from the Pictures folder was attached instead of our grey placeholder file, as on the first pass; the task completed and the substitution is recorded as a fixture deviation.

Capabilities exercised: photo documentation.

A-09: Create invoice

STS-TEST-HOUSECALLPRO-CORE-2026-015 first-use pass, October 11, 2026, not completed as specified.

  • Opening the invoice from the job produced a numbered invoice straight away, with the job's service line, quantity, price and total already filled in; nothing had to be typed.
  • Getting from the dashboard to the finished invoice took four clicks and about 40 seconds, including a pause to re-read the task.
  • The invoice screen offered three actions: print, download as PDF, and send. It had no save or draft button and showed no status for the unsent invoice, so the screen did not show whether the invoice stays in the account if it is not sent.
  • Downloading the PDF raised a one-time question about including images, then produced a one-page PDF that matched the on-screen invoice.

This run is not admissible as a benchmark result: the fixture was not completed as specified (the run table has the row). Its observations above stand as recorded; it contributes no timing and no completed scenario.

Capabilities exercised: invoicing.

STS-TEST-HOUSECALLPRO-CORE-2026-016 repeat pass, October 11, 2026, not completed as specified.

  • From the job's page, the print icon in its Invoice section produced a numbered invoice with the job's service line, quantity, price and total already filled in; nothing had to be typed.
  • Reaching the invoice took four clicks from the dashboard, and the invoice preview opened in a new tab about a second after the print icon was clicked.
  • The job's Invoice section offered only two actions, send and print; there was no button to save the invoice without sending it.
  • Printing went through the browser's own print dialog, which saved a one-page PDF matching the on-screen invoice in under 20 seconds.

This run is not admissible as a benchmark result: the fixture was not completed as specified (the run table has the row). Its observations above stand as recorded; it contributes no timing and no completed scenario.

Capabilities exercised: invoicing.

STS-TEST-HOUSECALLPRO-CORE-2026-019 first-use pass, October 12, 2026, completed as specified; observations confirmed by the driver.

  • Opening the invoice from the job again produced a numbered invoice straight away, with the job's service line, quantity, price and total filled in; nothing had to be typed.
  • Sending it took two clicks: the Send button opened a dialog already filled with the customer's e-mail address, a subject line carrying the invoice number and amount, a short message and the PDF; the dialog offered e-mail only, with no text-message option.
  • About four seconds after Send, the job page showed a confirmation naming the address, the invoice action read as sent to the customer, and the job's invoice panel carried the invoice number.
  • The invoice then appeared everywhere we looked: the dashboard counted one open invoice, the invoice page listed it with a due date of the sending day, and the invoices list showed it as open for the job with the send time recorded.
  • Before sending, the account counted no open invoices; after sending, it counted one. On this account, sending is what created the invoice record, which the first sitting had left untested.

Capabilities exercised: invoicing.

STS-TEST-HOUSECALLPRO-CORE-2026-020 repeat pass, October 12, 2026, completed as specified; observations confirmed by the driver.

  • On the repeat pass the invoice was reached from the job page's own Send Invoice button, a second route to the same screen as the first pass, and the generated invoice carried the job's line and total as before.
  • Sending took the same two clicks: a dialog prefilled with the customer's e-mail address, a subject naming the invoice and amount, a message and the PDF, with e-mail as the only channel.
  • The job page confirmed the send within a few seconds, with the recipient named and the invoice number shown on the job.
  • The dashboard then counted two open invoices, the invoices list showed both with their send times, and the second invoice's page listed the fixture line and total with the sending day as the due date.
  • From starting the task to the sent invoice took about forty seconds, half the first pass, with no new navigation to learn.

Capabilities exercised: invoicing.

A-10: Record / attempt payment workflow

STS-TEST-HOUSECALLPRO-CORE-2026-025 first-use pass, October 12, 2026, completed as specified; observations confirmed by the driver.

  • The payment was recorded from the invoice's own Pay button, which opened a payment dialog with the amount and the customer's receipt address already filled in and four methods on tabs: card, cash, check and other.
  • The card tab asked us to connect a bank account before a card could be charged; the cash, check and other tabs need no processor, and no sandbox or test mode is offered anywhere in the dialog.
  • Choosing cash and clicking the paid-by-cash button recorded the payment in one step: a confirmation message appeared within a few seconds, the dashboard's open-invoice count dropped by one, and the invoices list showed the invoice as paid with nothing due.
  • Reopening the invoice afterwards still showed the full amount due in the printed document's header; the paid state lives on the list and the dashboard, not on the invoice document as issued.
  • From starting the task to the invoice reading paid took about a minute and a half on first use, including two pauses to re-read our safe-path instructions; the record itself took one tab change and one click.

Capabilities exercised: payments, invoicing.

STS-TEST-HOUSECALLPRO-CORE-2026-026 repeat pass, October 12, 2026, completed as specified; observations confirmed by the driver.

  • On the repeat pass the payment was recorded from the invoices list: each unpaid invoice's row carries a Pay action, which opened the same payment dialog as the invoice page without opening the invoice.
  • The dialog opened on its card tab, which asked us to connect a bank account before a card could be charged; the cash, check and other tabs need no processor, and no sandbox or test mode is offered.
  • Choosing cash and clicking the paid-by-cash button recorded the payment in one step; a confirmation message appeared and the row changed to paid with the amount due at zero within a few seconds.
  • From the dashboard to the invoice reading paid took about half a minute, with no help and no prompt in the way.

Capabilities exercised: payments, invoicing.

A-11: Create recurring job

Status: Not run. No run is on record for this scenario, so there is no product observation to show. Next step: a first-use pass and a repeat pass on the current fixture, before any hands-on statement about it.

A-12: Configure reminder / follow-up

Status: Not run. No run is on record for this scenario, so there is no product observation to show. Next step: a first-use pass and a repeat pass on the current fixture, before any hands-on statement about it.

A-13: Configure review request

STS-TEST-HOUSECALLPRO-CORE-2026-027 first-use pass, October 12, 2026, not completed as specified.

  • The review-request feature sits under Marketing in a section called Reviews, reached in about a minute with no help; its entry page is a promotional overview with one button that opens a setup wizard.
  • The wizard lays out four steps before anything is configured: connect a Google account, connect a Facebook account, then configure automatic requests after completed or paid jobs, then choose where reviews go.
  • The automatic-request step the scenario needed is listed after the two account connections, and the page text says to connect a Google or Facebook account to get started; our test rules do not allow connecting an outside account, so we stopped at the overview without setting up a request.
  • No plan or upgrade prompt appeared: on the documentation review management is on every plan, and what blocked the setup was the account connection, not the subscription.
  • Whether the wizard's Continue button lets the connection steps be skipped was not tried, so this run records the setup path as the product presents it, not a confirmed hard gate.

This run is not admissible as a benchmark result: the fixture was not completed as specified (the run table has the row). Its observations above stand as recorded; it contributes no timing and no completed scenario.

Limitations observed

  • A review-request rule configurable without connecting an outside account: Marketing > Reviews reads 'Connect your Google or Facebook account to Housecall Pro to get started' and the 'Reviews setup' wizard lists Connect Google and Connect Facebook before 'Configure automatic requests'; the sandbox rules do not allow the connection, and whether the wizard lets the connection steps be skipped was not tried

STS-TEST-HOUSECALLPRO-CORE-2026-028 repeat pass, October 12, 2026, not completed as specified.

  • On the repeat pass the Reviews section was one click from the dashboard, and it now opened as a dashboard rather than the first pass's promotional page.
  • Its first card said to connect a Google or Facebook account to get started, with connect buttons for each and no review-request setting on the page; we did not connect an account, so no request was set up.
  • No plan or upgrade prompt appeared beyond the trial banner at the top of every page, so the block is the outside-account connection, not the subscription.

This run is not admissible as a benchmark result: the fixture was not completed as specified (the run table has the row). Its observations above stand as recorded; it contributes no timing and no completed scenario.

Limitations observed

  • A review-request rule configurable without connecting an outside account: the Reviews dashboard's 'Recommended next step' reads 'Connect your Google or Facebook account to Housecall Pro to get started' ('Connect accounts', 'Connect Google', 'Connect Facebook'); no request setting is on the page, and the sandbox rules do not allow the connection

A-14: Find customer service history

STS-TEST-HOUSECALLPRO-CORE-2026-017 first-use pass, October 11, 2026, not completed as specified.

  • Housecall Pro's customer record is a tabbed page (Profile, Leads, Estimates, Jobs, Invoices, Attachments, Notes); it was two clicks from the dashboard, and the customer list's Company column told two identically named customers apart.
  • The Jobs tab listed the customer's three jobs with their scheduled dates, amounts and statuses as soon as the record opened, about 15 seconds after starting, and the Estimates tab showed the approved estimate.
  • No invoice appeared on the customer record: the Invoices tab read 'No invoices', because the invoice prepared on the job had not been sent; on the job's own page it sat as an unnumbered invoice section with a Send Invoice button.
  • The profile summary showed a lifetime value and outstanding balance of $0.00, and the dashboard counted no open invoices.

This run is not admissible as a benchmark result: the fixture was not completed as specified (the run table has the row). Its observations above stand as recorded; it contributes no timing and no completed scenario.

Capabilities exercised: crm.

STS-TEST-HOUSECALLPRO-CORE-2026-018 repeat pass, October 11, 2026, not completed as specified.

  • On the repeat pass the customer record opened about 25 seconds after starting, most of that time spent re-reading the task, and it opened on the tab last used rather than on the profile.
  • The record's Estimates and Jobs tabs listed the customer's approved estimate and its job, each at $189.00, and the profile summary showed a lifetime value and outstanding balance of $0.00.
  • The scheduled job being looked for sat on a second customer record with the same name; the customer list's Company column was what told the two apart, and the job appeared on that record's Jobs tab.
  • Neither customer record showed an invoice: both Invoices tabs read 'No invoices', since the invoice prepared on the job had not been sent.

This run is not admissible as a benchmark result: the fixture was not completed as specified (the run table has the row). Its observations above stand as recorded; it contributes no timing and no completed scenario.

Capabilities exercised: crm.

STS-TEST-HOUSECALLPRO-CORE-2026-021 first-use pass, October 12, 2026, completed as specified; observations confirmed by the driver.

  • The customer's record opened from the customers list in two clicks and presented the history as tabs: estimates, jobs, invoices, attachments and notes, with a banner above them totalling the two unpaid invoices.
  • The jobs tab listed the customer's three jobs with their created dates, scheduled visits and amounts, and the estimates tab showed the approved estimate as won.
  • The invoices tab listed both invoices sent earlier in the sitting, each against its job number with the send time, the amount and a delivery status, so the full chain from estimate to invoice was visible on the one record.
  • Reading the whole history took about a minute including a pause to re-read the task; the only care needed was choosing the right one of two identically named customer records, which the list tells apart by their company field.

Capabilities exercised: crm, invoicing.

STS-TEST-HOUSECALLPRO-CORE-2026-022 repeat pass, October 12, 2026, completed as specified; observations confirmed by the driver.

  • On the repeat pass we opened the wrong one of the two identically named customer records first, since the list tells them apart only by their company field; going back and choosing the right one cost about sixteen seconds.
  • The right record showed the approved estimate, the three jobs and both invoices on its tabs, and a job page reached from the jobs tab led back to the record through its breadcrumb.
  • The invoices tab listed both invoices against their job numbers with the send times and amounts, and its status column had changed for the first invoice: the product now reported that e-mail as bounced, which is the expected result of sending to our synthetic, non-deliverable test address.
  • Including the detour, reading the customer's full history took under a minute and a half, with no help needed and no prompt in the way.

Capabilities exercised: crm, invoicing.

A-15: Export / report data

Status: Not run. No run is on record for this scenario, so there is no product observation to show. Next step: a first-use pass and a repeat pass on the current fixture, before any hands-on statement about it.

A-16: Inspect mobile / responsive workflow

Status: Not run. No run is on record for this scenario, so there is no product observation to show. Next step: a first-use pass and a repeat pass on the current fixture, before any hands-on statement about it.

A-17: Inspect relevant integrations

Status: Not run. No run is on record for this scenario, so there is no product observation to show. Next step: a first-use pass and a repeat pass on the current fixture, before any hands-on statement about it.

A-18: Inspect online booking where supported

Status: Not run. No run is on record for this scenario, so there is no product observation to show. Next step: a first-use pass and a repeat pass on the current fixture, before any hands-on statement about it.

Limitations and what remains untested

The completed scenarios cover customer intake, the service catalog, estimate creation, estimate approval and conversion, job creation, scheduling and rescheduling, technician assignment, job notes and a photo, invoice creation and sending, a recorded cash payment, and the customer’s service history. One scenario run is not completed: the review request, whose setup asks for a Google or Facebook account connection before automatic requests can be configured, which our rules do not allow. The scenarios not yet run are recurring service plans, customer reminders, a data export, the small-screen inspection, the integrations inspection and online booking. The native mobile apps, card payments through Housecall Pro’s payments product, and every trade-specific scenario have not been run on Housecall Pro. The trial account ran the feature set of Max, so which plan a tested feature belongs to is what Housecall Pro documents, not what these runs show. Observations describe what the product did in this account during these sessions; a later product change can make any of them stale, and a retest is due when the benchmark window or a material product change calls for it.

Evidence preserved, and what is not published

For each run we keep the recording, the timestamped action log, the session sheet with the fixture values, and the extracted frames used in the review of the recording. These stay in the private evidence store: the links on this page and on the review reach the written summaries above, never a recording. Account credentials, the trial account’s identity, raw traces and internal processing notes are not published, and no screenshot appears on this page; Housecall Pro’s terms restrict publishing material from the account, so what we observed is described rather than shown.

For what these results mean for a purchase, return to the buyer assessment.