Ask HN: What should I do?
I'm a 23 year old solo founder who lives with mom and dad..and that's important because I used to brag about my startup's burn rate which was literally $1 a month (12$ a year for .com domain)...even my heroku server is free.
This was until today where Google decided to limit Google Maps free tier quota down and hard. It used to be free for us, but now it's limited to only 25,000 loads a day. Just to give you an idea, Stay22's traffic needs around 200k loads a day. Now I'm being charged around $10,000 a month at this rate (because not only I have to pay for extra map loads, but using their API like directions and geocoding also charges me per request).
I can't afford 10k a month (esp since I have no funding yet), our revenue is barely half of that.. and changing my whole code to a free alternative like leaflet/OSM would be my worst case scenario (I'd have get my startup down for weeks just to do the transition and idk what can go wrong then (for e.g seo, backwards compatibility, etc)). My secret sauce is all coded using Google Maps's API so it's going to be tough.
What do you suggest? Any ideas? If anyone knows someone at Gmaps who can understand our situation and bill us when we raise money for example that would be great.. Or should I just man up, let my customers lose money (since the hotel revenue is shared) for weeks, and convert the code?
20 comments
[ 0.29 ms ] story [ 54.5 ms ] threadhttp://www.openstreetmap.org/
open source mapping. You'd have to host or find a host. It will take time and effort, but the data is free.
there are other maps as services too.
https://www.mapbox.com/pricing/
Also, are most of your requests coming from unique users, or do you see lots of users making repeated requests over and over? Maybe you could limit the number of requests for a single user. It sucks for them, but it allows you to implement a cap and affect the least number of people possible.
Are there any features of your product you can offload to other services? For example you mentioned directions -- why not link to Google Maps itself and let them handle that? (This might not be an applicable solution because I haven't studied your product to see how it works, but you get the gist.)
Is your code written modularly? Ideally you won't have to port all your code at once in some gigantic all-or-nothing rewrite. You can simply move one component of your code at a time to a free or open-source alternative to Google Maps. This could reduce your burden chunk-by-chunk over time until it's manageable.
Another avenue is to increase your revenue. Easier said than done, I know, but perhaps you could request payment/donations from users, or at least for repeat users. Or maybe you could reach out to the conferences on your site and see if they'll sponsor you.
Finally, I don't know anyone at Google, but it can't hurt for you to try to look up the right people and send some emails. I'd start with their developer advocates. This is a shameless plug, but I started a forum on my site (https://IndieHackers.com) recently as a place for founders to help each other out with problems just such as these. Can't hurt to post there and see if anyone has advice or connections at Google.
Good luck, sounds like a tough position to be in! (By the way, your site is awesome and it'd be very cool to interview you for Indie Hackers.)
1- Caching won't help. It seems I'm being limited on just "map loads", up to 25k https://developers.google.com/maps/pricing-and-plans/#detail... After that I'm billed $0.50 per 1000 requests, and that's still a lot since we have around 200k loads a day (plus not to mention we're growing like crazy expecting to triple that amount by the end of this year). And the costs is just too damn high for us because our widget is like an ad impression..it doesn't always convert and requires lots of loads.
2- If I go OSM, or Bing maps, I might as well do the rest because I'd be spending more time trying to connect the services together. It wasn't meant to be changed or scaled..
I remember IndieHackers when it was launched here in HN, very good reads. I'll give it a try, thanks man!
If they did grant you it, highly suggest opening a waiting list or new users, rate limiting current users, and switching to OSM.
We are currently re-architecting some services and one of the key items in our newest version is creating an api where we can easily swap out map providers.
You say it would take a lot to redo your code to swap service providers, but you don't have a choice. If you "want" to give up, then give up, that is a choice. If you're using this as an excuse, that's fine, but recognize that it is an excuse and not the real reason.
I used to run HearWhere.com (now gone). I was in a somewhat similar position to you shortly before I shut it down (https://techcrunch.com/hearwhere if you're curious). I would have had to do some major code changes due to outside circumstances, and I decided I didn't want to. Not that I couldn't. I just thought that the site had served it's purpose. I learned how to code, it got a bunch of interest, but in the long-term, I wasn't going to be the next songkick and didn't want to be. Mostly just because I didn't want to put in the effort in a very challenging industry.
What I'm saying is there is nothing wrong with stopping if you'd like. But just be honest with yourself as to why you are not going forward. It isn't because Google changed their API pricing, it is because you don't want to put in the effort to change to another service provider that you could afford.
From the way you talk about potential partnerships or help with gmaps it seems you really believe in the value of your business. That's an asset. Try to articulate this value more clearly but understand that gmaps is unlikely to just "help you out" or "invest" in you at this stage since: 1) it's not their mandate and 2) they're the incumbent and potentially you are eating a piece of something google wants to do.
At the stage you're at ... 4-5K MRR 25-100K DAU ( I'm guesssing ), and suddenly a slave to a price gouge, an investor can help you out. Definitely try to find one. There's probably a lot for them to like in your situation, not the least part of it being that you're to some extent desperate for their help right now.
I love your non-defeatist attitude. Don't give up. Keep going. You know what you're doing is somehow close to how Airbnb started right?
To answer you, yes Airbnb does know about us and did show interest. Can't comment more than that.
I've been looking for an investor that believes in us and has the proper connections/industry experience that we are looking for. Not the easiest when you're based in Montreal.
After comparing all our options, MapQuest was the clear winner (100% free). Google/Apple Maps is still used to render the map and pins, but all the business logic uses MapQuest. The transition took roughly 1 day.
Obviously, this won't help with map loads. YMMV.
EDIT: You might be eligible for a grace period until October 12 [1]
[1] https://developers.google.com/maps/pricing-and-plans/standar...
The event planner who is using your widget is not sharing in any of the cost. I think charging them is probably not what you want, so the alternative is to lower your service level to them.
By this, I mean simply display a pre-generated image of the map, and make users who click open up a new tab to your site. That way, you only load a map to users who might convert.
You could try other map sources, multiple ones to get you as much free tier as possible, but eventually you'd run into the same problems again.
I'm not familiar with the google maps API so I'm not sure what a "hacky workaround" would look like. But I'd guess that one of the following would work:
- Rotating API keys
- Loading the map in an iFrame from a rotating set of domains
- Pointing multiple domains to the same IP and rotating which domain you use in the embed