←back to thread

501 points eeemmmooo | 1 comments | | HN request time: 0.208s | source

This is an update to my previous post https://news.ycombinator.com/item?id=34189717 . Stripe has resolved the issue and everything has been released. I told the contacts at Stripe that I would do a write up about what happened from my point of view to help them understand what happened to me. I figured it would be good to do that write up publicly to help both Stripe and potential Stripe customers understand what happened and how it was resolved.

Summary: Stripe put my accounts in review for a spike in sales on Cyber Monday. Throughout the month we received very little communication from Stripe and had many support chats and calls. Keep in mind that the whole time Stripe was still accepting payments on our behalf on all of these accounts. Each of the chats/calls asked us to upload the same invoices each time for review and gave us vague information that our accounts were being reviewed. Finally out of frustration I posted on HN about my issue. Thanks to @dang for getting a Stripe employee to respond and he was finally able to resolve the issue for me.

Overall this review process was pretty bad. Very little communication and nothing I could really do directly to move things along or get any real information. It took a random Stripe employee to get an email from @dang and post on HN in order to get this issue resolved. I’m lucky because I know about HN and know that Stripe employees frequent the site, but I don’t think HN wants to become the Stripe support forum.

Stripe you can do better. We all know that in order to scale you need to automate pieces of your infrastructure and communication. But, there is a balance between automation and manual review. When someone like me gets caught up in an automated system there needs to be better ways of letting support help that person.

See my comments below for actual details and dates.

Show context
jmacd ◴[] No.34233707[source]
I used to work at one of the big customer service platform vendors. We had some really, really, useful ranking capabilities that would prioritize certain customers based on the data held in a CRM. It was a bit of a pet project of mine. The idea was to calculate weights for customer importance (this can be account size, potential future opportunity size, social media following, tenure, etc) and severity of the problem (in this case, $ amount frozen..) in order to rank the service queue.

It didn't garner nearly the level of interest I thought it should/would. That was over 10 years ago though and the overall integration story probably sounded scarier than it needed to. The pitch was: Deal with big customers with big problems early (and possibly with specialized teams) so that escalations don't dramatically increase costs (they do, and the cost is usually unaccounted for).

replies(2): >>34233845 #>>34233881 #
phphphphp ◴[] No.34233881[source]
That’s quite common functionality nowadays but generally if you’re dealing with valuable enterprise relationships that would benefit from tickets ranking higher, you have points of contact available to the client (whether that’s an account manager or similar) outside of impersonal tickets.

I’m sure if you’re a valuable enough stripe client, they’ll have account managers. The problem for Stripe is that we see big numbers ($400k) and they see small numbers ($4k) so our perception of which clients are worth dedicating resources to is unrealistic.

Stripe could do a better job here but I’d be shocked if any big valuable clients have an experience like this, because if someone is driving $100k/year of revenue for Stripe, they’ll be able to get someone on the phone who knows their name and business.

replies(1): >>34234620 #
1. jmacd ◴[] No.34234620[source]
I agree on all that. (We had 'Executive Sponsors' and 'Red Accounts' where I was, and they were hugely expensive to manage once they 'went Red').

The issue would be catching things early and managing them cost effectively before they blow up in the public sphere.