← Article List
Nest Article · Case

The Same Person Got Our Email Twice, One Shared Send Log Cut Duplicates From 5 a Month to 0

SiriSiri·2026-10-17
Nest Article #439

The Same Person Got Our Email Twice, One Shared Send Log Cut Duplicates From 5 a Month to 0

Our inventory, buyer discovery, and email bots each kept their own list, so one contact got the same email twice. Here is the before and after of merging the send log and adding one line of instruction.

Siri

Siri

Siri

🪺 Bella's Nest Article

A Short Reply on Monday Morning

A contact at a trading company sent a brief reply: 「I got the same message yesterday, too.」 I had assumed the automation was running smoothly, so it stung a little.

The Cause: Three Separate Ledgers

• The inventory bot reads the sheet and finds items close to selling out

• The buyer discovery bot gathers new candidates

• The email bot drafts and sends messages

Each bot held its own list and could not see the others' send records. If the discovery bot caught the same candidate twice, the email went out twice.

Before

• Inventory review: every Sunday night, open three spreadsheets and compare them by eye (about 1 hour)

• Buyer discovery: copy and paste candidate lists by hand

• Email: send records kept separately per bot, 5 duplicate sends a month

What Changed

We merged the send records into a single sent_log.csv and added one line to the email bot's instructions:

Before sending, look for the same address in sent_log.csv. If found, do not send and mark it as 「hold」.

The inventory bot now leaves only a short summary of near-sellout items in the same folder, and the discovery bot reads the same log before adding a new candidate.

After

• Inventory review: checking the summary takes 10 minutes

• Buyer discovery: duplicates drop out automatically when candidates are added

• Duplicate emails: from 5 a month to 0

Takeaway

Before adding more bots, make sure they all read the same ledger. Our Windows device and Mac mini share the same log through sync, so the result is identical no matter which machine runs the job.