Is the data freshness a joke? 2 DAYS to get data into the data warehouse? Stitch, Fivetran, Segment, and more can do way better than that without any internal hooks.... :/
> Additional processing time–up to 48 hours–is required to make your account’s transactional data available in Sigma. This means that it does not reflect your account’s most recent data and should be considered a couple days behind. The Sigma interface displays the date and time of the last update to your data.
I've seen plenty of reporting pipelines that are that slow over the years. If this was built on the cheap, so it just uses existing pipelines, and instead of working through streaming, it regenerates the world every night, a 48 max failure with some is not out of the question once you add some CYA magic. That would make this pretty cheap to make: Some website work, boxes for queries, and some security work to make sure data from other customers doesn't leak.
Given how Stripe seems to build products in a lean way, it'd not surprise me if they are just launching like this and measure customer reaction. If the main reason it doesn't get the traction they want is the 48 lag, they'll just rewrite their reporting pipeline to use streaming, and the product gets faster for free.
It's payments. Miss a single row and customers will balk and scream at the missing $.45 that they think they are owed. Better to buffer with a longer SLA until you feel very confident that your pipeline is bulletproof before you try to approach 24 hours or less.
... hile we were at it, AWS announced AWS Lambda, which looked like a potential remedy to our challenge. Upon further investigation, however, it became clear AWS Lambda is centered on an asynchronous programming model that did not quite meet our needs. The custom logic we run in Auth0 is an RPC logic: it accepts inputs and must return outputs at low latency. Given the asynchronous model of AWS Lambda, implementing such logic would require polling for the results of the computation, which would adversely affect latency...
Definitely. Outbound sales was how we validated our business model before we wrote a single line of code. There is so much you can learn.
Outbound sales, be it cold calls, cold email or warm intros, are very powerful. Keep in mind "With Great Power Comes Great Responsibility". Don't waste people's time. Ask for permission to speak. Listen.
This is really interesting. Validating your idea is so important, and unless you know a bunch of people who have X problem or are in X industry and aren't going to just 'be nice' about your idea, then you do need to ring around.
Hey Patrick, stoked to hear your response! Solid advice all around, I definitely agree! Unfortunately, I was looking for something more than just "do a lot of work and do it well"... but I might just be wishing for something too good to be true!
On a related note, one of my favourite features of the Stripe webhooks is the UI for showing logs/events/etc. Makes developing SO MUCH easier when it's not just some magic black box.
Sounds like a great idea for a follow-up blog article. In fact, I just saw a presentation that said that before the referral program, Dropbox only had a K = 0.1. I'll try to put a list together.
Most of the models that we've seen are rather simple -- they assume that you just "have 1000 users" and then just show what happens from viral growth.
While those are illustrative of first principles, they don't really hold up when it comes to modelling your userbase over time.
> Additional processing time–up to 48 hours–is required to make your account’s transactional data available in Sigma. This means that it does not reflect your account’s most recent data and should be considered a couple days behind. The Sigma interface displays the date and time of the last update to your data.