Callback and Service Guarantee Response
Purpose
A callback is four different visits wearing the same name: covered activity that the program should have held, a pest that was never on the covered list, a customer obligation nobody completed, or a defect in how the last visit was executed. Treat all four the same and the shop learns nothing, gives away work it never agreed to, and keeps missing the one category it could actually fix.
This procedure classifies the callback before anything is applied, and it puts the classification in the record so the quarterly read is a real number rather than a feeling about which tech gets complaints.
Scope
Covers an unscheduled customer-initiated report of activity on a live account, from intake through classification, response, verification and the quarterly rollup. Applies to residential and semi-commercial accounts under a service agreement.
Does not cover: the scheduled follow-up interval after an initial treatment, owned by the follow-up and retreat protocol SOP, which sets the per-pest windows this procedure does not restate; what is covered and what the customer owes, owned by the new account inspection and agreement setup SOP; the routine visit itself; and any complaint about drift or off-target application, owned by the drift and neighbor complaint SOP, which has a different clock and a different reporting path.
Roles and responsibilities
| Role | Owns | Handoff point |
|---|---|---|
| Office | Intake, acknowledgement inside the agreement's window, scheduling, pulling the account file to the job record | Hands the tech the covered list, the exclusion list and the obligation status before dispatch, not after |
| Technician | On-site verification, identification, classification with its evidence, the response and the verification date | Escalates any classification the customer disputes rather than negotiating it at the door |
| Service manager | Classification review, approving out-of-scope work as a quote, the quarterly rollup by account and by tech | Owns the conversation with a tech whose callback rate runs above the peer rate, and owns the decision to release an account |
Procedure
Step 1 - Take the report in the customer's words, with a date and a place. Ask what they saw, how many, exactly where, at what time of day, when it started, and whether anything changed in the building. Ask for a photo. Acceptance: the ticket carries the pest as described, a location, a first-sighting date and a count or a frequency, plus the acknowledgement sent inside the agreement's window - the shop default being same business day. Wrong looks like an intake reading "customer sees bugs", which sends a tech out with nothing to verify against. Stop rule: the caller reports illness or a reaction they attribute to an application, this is not a callback; route it immediately to the service manager and treat it under the incident path with the label's first aid statement and the national Poison Control line at 1-800-222-1222. Hazard: none at intake, but this is where a medical report gets miscategorised as a pest report and loses a day.
Step 2 - Read the account before the truck rolls. Pull the covered list, the exclusion list, the status of every customer obligation, and the last two visit records including any monitor counts. Acceptance: the tech has all four on the job record before dispatch, and the on-site visit is scheduled inside the agreement's window - the shop default being 2 business days. Wrong looks like dispatching on the complaint alone, which is how a tech promises a retreat for a pest the agreement excludes. Stop rule: the report names a pest on the exclusion list, do not dispatch it as a covered callback; the office tells the customer it is a quoted job and books it that way. Hazard: none, it is a desk read.
Step 3 - Verify the activity yourself before anything is applied. Inspect the reported location and the harborage around it, read the existing monitors, and collect a specimen into a vial rather than working from a phone photo. Acceptance: live specimens, fresh evidence such as spotting, shed skins, droppings or damage, or a monitor catch, identified to species, with the count and location written down. Wrong looks like applying product because the customer is standing there expecting it. Stop rule: no activity of a covered pest can be verified, do not apply anything; place numbered monitors at the reported location and the likely harborage, set a re-check date, tell the customer plainly what would change the call, and record the callback as unverified. Hazard: pull appliances only with the customer's agreement and the gas or electric connection in view, and never move a gas range on a flexible connector, because a stressed connector leaks into a space you will not smell it in.
Step 4 - Classify the callback, and write the evidence next to the class. Every callback lands in exactly one of four classes: covered activity within the program, out of scope, an outstanding customer obligation, or a shop execution defect such as a missed area, a wrong product choice or a skipped device. Acceptance: one class recorded with the specific evidence that puts it there - a species identification, a line from the exclusion list, an obligation with its date, or a gap in the last visit's own record. Wrong looks like defaulting everything to covered activity, which hides execution defects inside the guarantee. Stop rule: the evidence fits two classes, escalate to the service manager rather than choosing the cheaper one. Hazard: none at this step; it is a call made standing over the evidence.
Step 5 - Respond to the class, not to the pressure. Covered activity gets a labeled treatment aimed at what the inspection found, at no separate charge. Out of scope gets a written quote and no free application. An outstanding obligation gets a re-brief in writing with the date reset, and the treatment only where the agreement still covers it. An execution defect gets the correction plus a root-cause note naming what was missed. Acceptance: the action taken matches the class recorded in step 4, and every product applied is verified against the label for that site and pest. Wrong looks like a broad application made to look decisive, which contaminates monitors and can push a bait-responsive pest away from bait you are about to place. Stop rule: the customer demands a treatment the label does not support at that site, refuse and say why - it would be use inconsistent with the labeling under FIFRA at 7 U.S.C. 136j(a)(2)(G). Hazard: the account's occupant safety data governs re-entry and notification here as much as on a routine visit, and a callback is the visit most likely to skip it.
Step 6 - Set the verification date and say out loud what resolved will look like. Leave numbered monitors in place, record their starting count, set a re-check date, and tell the customer the number that ends this: zero live catches on every monitor at the re-check, with no new sightings reported in between. Acceptance: monitor positions, starting counts and the re-check date are all on the ticket, and the customer can repeat the standard back. Wrong looks like closing the callback the day the treatment is made, which converts a two-visit fix into an open dispute. Stop rule: the re-check finds any live catch of the covered pest, the callback stays open and goes back to step 3 rather than being closed as complete. Hazard: place monitors where a child or pet cannot reach them, and use a tamper-resistant station for anything carrying bait.
Step 7 - Roll the classifications up quarterly, by account and by tech. Count callbacks as a share of routine visits, per tech and for the shop, and compare a tech against the rate of their peers rather than against a blend that includes their own work. Acceptance: a rate per tech and a peer rate computed with that tech's own visits and callbacks removed, plus a list of accounts carrying a third callback in one term. Wrong looks like ranking techs against a shop blend they are inside, which flatters the outliers and understates the gap. Stop rule: a tech's rate runs at or above 1.5x the peer rate, that is a training and ride-along conversation, not a reassignment. Hazard: none, it is a report.
The record this produces
Per callback: intake in the customer's words with the first-sighting date, the account's covered and excluded lists as they stood, obligation status, what was verified on site with species and count, the class with its evidence, the action taken with any label details, the monitor positions and counts, the re-check date and its result.
Per quarter: callbacks by class, the rate per tech, the peer rate, and the accounts carrying repeats. The class breakdown is the payload. Mostly out-of-scope callbacks is a selling problem, mostly obligation callbacks is an onboarding problem, execution defects are a training problem. Without the class field all three look identical on a callback count.
One callback, filled in
Day 0: quarterly residential account, customer calls reporting "bugs in the kitchen at night, five or six, started about a week ago." Acknowledged the same business day. Step 2: covered list includes German cockroach and oriental cockroach; no exclusion applies; all three customer obligations current; last visit's two kitchen monitors read zero.
Day 2, inside the 2-business-day window. Step 3 fails. No live German cockroach, no fecal spotting, no shed skins, and both existing monitors are still clean. Stop rule taken: nothing is applied. Four numbered monitors go in - under the sink, behind the range, at the floor drain and at the dishwasher toe kick - all starting at zero, with a re-check set for day 9. The customer is told plainly that a catch on any monitor changes the call, and that a blanket spray today would only make the next week's evidence unreadable.
Day 9: monitor 2, behind the range, holds 3 specimens, identified as oriental cockroach, which is on the covered list. The floor drain by monitor 3 has a dry trap. Step 4 classification: covered activity, with a conducive condition, evidenced by the identification and the dry trap. Step 5: a labeled treatment aimed at the drain and the harborage rather than a general kitchen application, plus a written instruction to the customer to pour water down that drain weekly. Step 6: the four monitors stay, re-check set for day 23, and the standard stated as zero live catches on all four.
Day 23: all four monitors read zero and no sightings reported. Callback closed as resolved.
The quarterly rollup that follows: 240 routine visits across the shop with 18 callbacks, which is 18 over 240, or 7.5 percent. Tech A ran 60 visits with 9 callbacks, which is 9 over 60, or 15 percent. Comparing 15 percent against the 7.5 percent blend is unfair to the blend, because tech A's own work is inside it. Removing tech A leaves 240 minus 60, which is 180 peer visits, and 18 minus 9, which is 9 peer callbacks, so the peer rate is 9 over 180, or 5.0 percent. Tech A at 15 percent against a 5.0 percent peer rate is 3.0x, well past the 1.5x gate, so the stop rule fires and a ride-along is scheduled.
When the callback does not fit the procedure
The customer refuses entry to the room they reported: verify what you can, record the refusal, and do not classify from the doorway. The activity is verified but the pest is on neither list: that is a gap in the agreement, so treat what the label supports as a quoted job and send the scope back for amendment. The same account takes a third callback in one term: stop retreating and re-inspect from scratch under the initial inspection SOP, because three callbacks means the plan is wrong rather than the treatment. And where a customer is abusive or the site has become unsafe, leave and report it.
References
- FIFRA, 7 U.S.C. 136j(a)(2), on use inconsistent with the labeling, which gates the refusal in step 5.
- The account's own service agreement, which supplies the covered list, the exclusion list, the obligations and the response window this procedure enforces.
- See related: follow-up and retreat protocol; new account inspection and agreement setup; initial inspection and treatment plan; drift and neighbor complaint response.