Last spring, we began to suffer from mass mailings every Friday: letters piled up in line and left very late. These were ordinary gray Fridays. Already then, six months before Black Friday, it became clear that autumn would not be easy. I had to scale. I tell you how it went.
Spoiler: everything worked out, and we are well done
Was black Friday in Mindbox flawless? No, it didn’t. Was there at least one defect related to sending newsletters? No wasn `t!
For several weeks we digested peak volumes of letters, and nothing accumulated anywhere, did not linger, did not interfere with transactional ones, did not double or beat.
It is important not to forget that the product is perceived from the outside as a whole, and stable email sending is not enough. At the same time, it is important to pay attention to both weaknesses and what turned out well. It is very tiresome to be a fool and learn from your mistakes, sometimes it’s nice to find local success and try to replicate it.
In short, then:
- sent more than 1,300,000,000 letters in a month (this is one billion three hundred million),
- it's 59 terabytes (without pictures),
- in one day - 71 million,
- at a peak speed of 200,000 per minute.
Below are a few lessons for the future, some of them with pictures.
Black Friday does not exist, there is a black half of November
, : (29 2019 ) — , - . — !
, , .
— , - . 40 — , .
28 69 175 .
, , .
1 300 000 000 — , :
( 71 ) — 26 , , , :
- — , 18:00. - 200 ( — 2 ):
? - email-? , .
№1:
CDP (, ). 2019 , , , .
2017 «, , ». , .
, : , .
№2:
: , , , :
, , . , , . .
, : , , , .
, ? — , ?
, .
№3:
, - : MTA , — . : .
, ( , ). MTA — , 100 , , , AMP- .
, : ( ), — .
,
:
« , ?» CIO : «, ». , .
, «» - , - .
, ? , :
, :
, - , , , .
,
2013 , , 5–15 . , 1 .
, 8–10 . «… , , ».
, , 100 ! , .
, , . , .