Previously I posted about Google Maps alternatives since they introduced a pricing structure that affected us in a big way! They have now seemingly taken on the comments of a number of developers, start-ups and established businesses and has now re-negotiated it’s pricing plan for Google Maps for those who it affects, including ourselves.
While the Maps API remains free for the vast majority of sites, some developers were worried about the potential costs. In response, we have lowered the online price from US $4 per 1,000 map loads to 50¢ per 1,000 map loads.
— Google Developers Blog
That’s means that Google Maps is now 87.5% cheaper than before for standard map developers. This now makes Google Maps a lot more competitive with other paid for offerings such a MapBox out there. But how exactly does that stack up with a highly visible site?
Working It Out
Working out some basic numbers using one of our own stats on map views on a site Fubra’s Director/Co-Founder, Paul, and myself have worked out a cost comparison.
As you can see it’s a considerable drop in cost from over $36k (~£23k) to $4.5k (~£2.8k) but still a little costly for a site that only uses a map to provide additional visual feed back on data as apposed to it being a main factor. It’s hard to argue that Google should be offering Maps for free as they are expensive to run, but it does show that their original offering of $4 is somewhat excessive and it goes to show they know it all to well by the slash. We still feel that $4.5k for a site that makes very little money is too much, and if we did have to pay for a solution we would rather partner with a smaller business. But we aren’t charging access to use the maps on our sites, they are in fact open to all to see with Google’s logo in the bottom always on show.
Our solution to use Open MapQuest still proves to be free for us, and on the basis of cost-benefit analysis, for us free is good for a great quality crowd-sourced solution.
It will be interesting to see if people would move back to Google Maps when our workings showed they are cheaper then MapBox; but MapBox have the start-up feel that a lot of other individuals and smaller companies would look towards as apposed to Google.
It’s Not All About Costs
It has to be said that Google are well established with servers that can take the load of a huge number of users accessing their maps and they are very well documented. Being possibly one of the biggest mapping providers out there, you can’t argue that you won’t get what you pay for. But saying that, if they are one of the biggest mapping providers then surely majority of sites maps look the same and their implementations are very copy & paste? Of course you can have styled maps and there is typically more than one way to skin a cat, but through some experimentation we’ve been doing here, to move away from the base model that Google Maps provide with their API was very more of a hack and didn’t quite fit in with what we wanted. Suffice to say, Google Maps isn’t an open source solution and sometimes that’s what you need more than something that quick and easy to implement.
I mentioned in my previous post that one of things we found with our choice solution of Open MapQuest gave us details that previously where missing from Google Maps, but it’s more than that that keeps us using Open MapQuest and the Leaflet API. There is a sense of freedom in what we are now developing with maps, Google Maps caused us to be very rigid and most of the time forced us into a box that we didn’t like the look and feel of and the box was all too familiar to that point it was boring; nobody likes a boring box and boring websites aren’t that great to use.
Six Months On
Having used a free, open solution for over six months now we aren’t in a position to look back to Google Maps, even with the massive reduction in price. That initial charge we were faced with has opened our eyes in regards to mapping solutions and thus caused us to solve problems differently. It’s also allowed us to see that although a product is popular and well supported, it might not fit your solution; the case of Google Maps not showing airports on the map yet OpenStreetMap’s data showed us what we needed made us wonder why we didn’t do this before! For the majority of locations we check, the data is more complete, and if it’s not then we are able to update the data either in a number of ways, helping contribute to the OSM community and making it grow.
Google Maps is a quick product and allows ideas to develop very quickly, but it’s more of a prototyping tool for us that somewhat akin to front-end coding solutions like Twitter’s Bootstrap in that it helps get an idea off the ground and into the public but not something that you’ll develop any long term plan. As we have all the code for Leaflet API available to us, we can edit it in any way we wish to suit our means, and access to the CSS is a must have in this day-and-age of web apps. The API for Leaflet in my opinion is much more of a pleasure to use that allows us to move in directions we may not of thought to previous.
We no longer think that Google Maps is the best out there, it’s by far the most complete with a number of solutions built in, but saying it’s the best is subjective and we’re not in that camp. Open MapQuest API offer resources that will help in regards to getting a more complete solution, sometimes offering more than what you get with Google Maps API.
At a later date I will post more on the experience of making a web app (I’m in-between writing about our RoutePlanner app for PetrolPrices.com) with open source solutions. But I thought I’d write my response on Google Maps latest change and how we are still committed to our open source solution to the point that we are sponsoring State Of The Map 2012 so be sure to check that out! Here’s a look at our sponsor slide…