Lead quality did not drop. On 11/2/2026 a second source called "LSA" started logging every LSA call, including junk and repeat calls that were not recorded before. After duplicates are removed, November books 26% of leads (13 of 50). April to June 2026, the last months when every call was logged, booked 23% to 33%.
What is happening
- The old call line mostly logged booked leads. From July to October 2026 it shows 84% to 97% qualified. That is too high to be a real booking rate. It means unbooked calls were mostly not entered in the CRM. Only April to June 2026 logged every call, and those months show 22% to 31%.
- The new "LSA" source logs everything. It started on 11/2/2026 and records every LSA call and message, whether or not it books. In November it logged 54 rows. 3 booked.
- Both sources run at the same time. The call center still enters booked calls under the old call line (9 of 9 booked in November). The new source then logs the same person again. 14 of November's 64 rows repeat a lead already in the CRM (marked DUPE, the same phone number, or the agent notes name the lead).
- So the two rates are not comparable. The old line shows only the winners. The new source shows everyone. Combine them and remove duplicates, and November is in line with April to June.
November, with duplicates removed
64rows logged (both sources)
50unique leads after removing 14 repeats
13booked an appointment
26%qualified rate (April to June: 23% to 33%)
Why the other 37 did not book
Glass, mirror, screen or window repair18
Other job we do not do (interior doors, commercial doors, roofing tin, cellar doors)6
Not a customer (job seeker, employee, warranty, wrong number, accidental, spam)6
No reason given or below $2k minimum4
Still trying to reach3
Grouped from the call center notes on each lead. Half of the leads that did not book wanted glass or window repair work, which Southern Exteriors does not do. All 37 came in through the new "LSA" source.
What to do next
- Dispute the junk leads in the LSA dashboard. Google credits leads for job types the business does not serve, and calls that are not customers. The 24 repair and wrong-job leads and the 6 non-customers are the ones to check.
- Check the LSA job types and service areas. The glass and repair calls suggest the profile is showing for window repair searches. Turn off any repair or glass job types.
- Pick one source for LSA in the CRM. Log every LSA lead once, under one source. Then the qualified rate can be tracked month to month without duplicates.
- Confirm the history. Ask the call center whether unbooked LSA calls were entered from July to October 2026. If not, those months' rates should not be compared with November.
Monthly detail by source
Qualified rate by month
Old call line (678-218-6809)
SMS
New "LSA" source (from 11/2/2026)
Months where every call was logged
Most months show 80% to 98% because the CRM held mostly booked leads then. Only the shaded months (April to June 2026) include unbooked calls, so they show the true rate. Months with fewer than 10 leads in a source are less reliable.
Leads logged per month
Old call line
SMS
New "LSA" source
Lead volume jumps in April to June and again in November, the months when unbooked calls were logged. In November many new "LSA" rows repeat a person the old call line already booked (marked DUPE in the CRM).
Month by source
| Month | Source | Leads | Qualified | Qualified % | Junk % | Sold |
|---|
Junk = dead, spam, not qualified, duplicate, wrong number, out of area, do not call, not interested, or went with another company. Source file: DUO Disposition Report exported 11/10/2026, 1,092 LSA rows. November is month to date.