FIELD NOTE / TIKTOK
The salesperson built a scheduler. Leads waited.
The complete written thought and the evidence behind it. The video edition will follow its public release.
The written argument is here.
This approved TikTok edition is on the journal now. Its video player and original platform link will appear after each public release is verified.
The salesperson built a scheduler. Leads waited.
Video caption
The salesperson built a scheduler. Leads waited. Someone now owns product operations. Use AI on the differentiator. #EricFieldNotes
Full written post / accessibility read
In this fictional team, a salesperson gives a coding agent a prompt and gets a meeting scheduler by lunch. It looks clever. During the same week, leads wait for follow-up. Then a prospect cancels, the calendar does not update and nobody owns reminder delivery.
The feature list grows: time zones, no-show recovery, two-way calendar sync, consent, permissions, logs and customer support. None is impossible. But the business did not decide to staff a scheduling product. It quietly converted selling time into maintenance time without a deliberate investment decision.
Before writing a replacement, test a maintained scheduler with representative cancellations, permissions and integrations. Record the unsolved requirements. If the gap is one unique workflow, build the bridge. If the tool truly needs to be owned end to end, assign support, security and a year-two budget.
My rule: do not turn every employee into an accidental SaaS team. Buy commodity capabilities that meet the contract, and spend agent-building time on work the company uniquely needs. Do this because a visible demo is a poor substitute for the role's real outcome and a maintained service.
#EricFieldNotes
Four-beat scene transcript
1. The salesperson built a scheduler. Leads waited.
In this fictional team, a salesperson gives a coding agent a prompt and gets a meeting scheduler by lunch. It looks clever. During the same week, leads wait for follow-up. Then a prospect cancels, the calendar does not update and nobody owns reminder delivery.
Visual: A fast artifact can displace the actual job.
2. The hidden backlog arrives later.
The feature list grows: time zones, no-show recovery, two-way calendar sync, consent, permissions, logs and customer support. None is impossible. But the business did not decide to staff a scheduling product. It quietly converted selling time into maintenance time without a deliberate investment decision.
Visual: Someone now owns product operations.
3. Run the one-week market test.
Before writing a replacement, test a maintained scheduler with representative cancellations, permissions and integrations. Record the unsolved requirements. If the gap is one unique workflow, build the bridge. If the tool truly needs to be owned end to end, assign support, security and a year-two budget.
Visual: Try a maintained service on real cases.
4. Pay for maintained work when it fits.
My rule: do not turn every employee into an accidental SaaS team. Buy commodity capabilities that meet the contract, and spend agent-building time on work the company uniquely needs. Do this because a visible demo is a poor substitute for the role's real outcome and a maintained service.
Visual: Use AI on the differentiator.
Research and claim limits
- Claude Code overview (S48)
The examples identified as illustrative or simulated are design probes, not reported incidents. Vendor specifications do not establish workload performance.