I’ve built everything on Airtable, but I realize that I’m paying $120 a year for that. Recently when I tried to have volunteers help update stuff (and invited 6 of them to Airtable) I got a bill for $120 for EACH of them. AirTable was really cool about it and refunded the money when I deleted the volunteers and asked for money back. However that isn’t a huge drawback (more my mistake) and I created a better work around.
However, I’ve wondered - are there PERKS to using AirTable vs Google Sheets or Google Sheets vs Airtable?
Google Sheets has been working so-so for me, with a few minor glitches - mainly action elements not functioning properly (delete row failing consistently), but I’m not sure if that has to do with the Sheets API or if it’s on Glide’s side.
Have you noticed any performance issues with Airtable?
Newbie. But not so far. Everything working as planned so far. I posted something about not being able to get connections to work but I suspect thats because I’m not using various tabs in a single spreadsheet (i.e. Airtable Base) but am instead making a separate spreadsheet for Users, Office locations and so on.
Your absolute most reliable option will be Glide Tables. Especially if you don’t need any functionality that’s exclusive to Airtable or Google Sheets. The only downside is that there is no easy way to pre-populate a glide table initially.
The end user’s app will only ever be connecting to glide servers. There is no direct connection between the end user and Airtable/Google. Glide keeps it’s own synchrized copy of the data that’s in Airtable and Google Sheets. So by using an outside database, you are introducing an extra path for data to travel and synchronize, which can lead to additional lag if data is being further manipulated in that outside database. You are also dependent on the reliability and uptime of that third party database. By using Glide Tables, you keep data only on Glide servers, and data doesn’t need to travel to other outside services.
So in the end, the user experience should (in most cases) not differ, regardless of which database is used, since the app is only communicating with glide servers and glide has a copy of the database. What becomes critical is how well glide servers and the outside third party databases are communicating with each other to keep data in sync.
There’s advantages and disadvantages regardless of whichever you choose. Glide Tables are kept within the glide ecosystem. Airtable/Google have exclusive functionality that’s not in glide, but it’s a third party database and a separate account to keep track of.
My primary app is exclusively Google Sheets, because it’s 3 years old and was built long before we had any choices other than Google sheets (or a data editor in glide, for that matter). I also have automated backups of my Google sheet, so I have assurance that I always have my own backup copy. Most new apps I build are exclusively Glide Tables because it eliminates a potential point of failure if the outside service goes down for any reason.
Choose what works best for you. In most cases, there should be little to know performance difference.
Just remember these points:
Separate copies of the database are kept on Airtable/Google, on Glide servers, and on each and every end user device. Glide keeps each and every copy of that database in sync.
Whenever a user makes a data change in an app, that change is saved to the local database → then to glide → then from glide to airtable/google → any sheet specific calculations modify the data → then syncs back to glide → and then glide syncs back to all end users.
To specificly answer your question, I don’t think there should be much difference if you choose to use Airtable or Google Sheets based on the way Glide handles the data. What’s important is which one you are more comfortable with. Also keep in mind that Glide was founded on using Google sheets as a database and has had that option for over 3 years. Airtable is a relatively newly added data source for glide, so there still may be some bugs for glide to work out as far as data synchronization.
I appreciate the thoughtful answer, Jeff! It looks like using a Glide table exclusively is best for performance, as it’s only a two-hop relationship instead of a three-hop, in the case of Google Sheets and Airtable.
With your Google Sheets apps, are there any specific performance bottlenecks you’re seeing arise that should be avoided?
How do GlideApps respond to multiple requests happening in parallel, and how well would it scale to 10-20-50-+ users?
Quick tip: For Airtable or Excel tables, once you create a project in Glide with one of these sources, their tables can be duplicated in Glide and then they will show up as Glide Tables and be available for this and other projects in your Team folder. This is unique to the new Airtable and Excel data sources and will not behave the same way for Google Sheets (duplicating a Google Sheet table just creates another Google Sheet table).
Jeff_Hager… again, yes, thanks for the longer response. What I say below you already know, but for some other reader that happens by (and correct me if I’m wrong)…
For Speed —> Glid Apps only
… but…
For Integration (aka no-code revolution) it seems to me like having stuff in google sheets or AirTable’s opens the world to functionality because so many (Zapier/Integromat,e tc…) work with either of those.
Since I can’t wrap a MailChimp sign up form into GlidePages (the “wrap web page” button that is in GlideApps is NOT available in GlidePages… cough…cough) so you can’t wrap any email signup forms or paypal or donation anything into Glide pages.
Since I use Airtable I was able to set up an integration between a form I built onto the page using GlidePage’s features which dumps data into Airtalbe, and then Automate.io adds a subscriber to MailChimp for me. I don’t think I could do that if i was using just Glide’s built in database/tables.
Nathan, do you feel like there are a large and growing number of integrations with glide tables. It would be unfair (I assume) to expect that there are as many integrations with applications and GlideTables as there are with applications and AirTables or google sheets.
Maybe I’m wrong though. Maybe GLideTables has some easy webhook thing set up so that it can basically work with anything using that one webhook.
I guess I’m timid here. I’d love to move GlideTables for the speed and (keep it all under one roof as you say) but I’d hate to move to GT and then find out (uh oh, I can’t do this essential integration because someone hasn’t built it yet between GT and app XYZ on Integromat or wahtever).
If your project roadmap is still somewhat open and you feel you’ll need a lot of integrations down the road with 3rd party applications, then I’d share your concern, good point.
Personally, my app has just shy of 9k rows. I have some logic that happens in the sheet instead of glide, so there is a delay in waiting for the data to resync to glide. Other than that I haven’t experienced any bottlenecks with my data. I think the theoretical 25k limit that glide imposes is largely due to bottlenecks when synchronizing large sets of data over 25k between glide and google. Also, when trying to sync that amount of data to the end user’s device. You can definitely exceed 25k, but you may start to see delays in the data syncing between glide and google and general app slowness due to large cache size to temporarily store that data.
You shouldn’t have any problems scaling to thousands of users or more. The data on the end user’s device is sync’d to the database on the glide server and I assume it’s processed immediately as it comes in. I then believe glide will send updates to the google sheet in batches. You shouldn’t have any problems. The only weirdness would be if multiple user’s are trying to edit the same row of data at the same time. In that case the last user to make the change would probably win.
For an Airtable-based project, I used to mirror Airtable to Google Sheet using Coupler. When Glide’s native Airtable integration was released, I got excited and thought the days of having to configure an infinite number of webhooks was over.
I rebuilt the app with a similar functionality based on Airtable, though obviously not identical. The performance is really abysmal. By that I mean it takes >30s to use on mobile with load animation freezing, the app hanging when a user switches tabs etc. Each time the app is opened, it loads >10 MB cache onto the end user’s device and, without good connection, get’s stuck in the process.
The Google Sheet based mirror in comparison runs much faster and is fairly smooth. The mistake may be on my side here, but the experience was quite disappointing.
I mentioned yesterday that when I started a new app based on Excel (5 tabs/tables), and then deleted one of the tabs in Excel, Glide removed the link to that Excel tab (all the other tabs remained), and replaced it with a Glide Table that was fully populated (columns and rows). I mentioned that I felt that was a “Feature” – not a “Bug”, and that if the Glide table was not wanted it could be deleted (and I did delete it). But if Glide can confirm whether that is the intended action and whether they will support that as a “Feature”, it could be an easy way to populate Glide tables initially.
@SantiagoPerez, you didn’t answer that point in my post – you simply told me to delete the table to work out the synchronization on the surplus rows on another table.
Hello Santiago, I really need clarification as to whether this is in the works for Google Sheets. You see one of my templates uses Sheets, and Glide now wants me to modify it to use Glide Tables, and I just tried by copying the app and deleting a Google Sheet, and it did not create a Glide Table.
Since everyone who has a template in the Marketplace has to convert their Sheets to Glide Tables, can Glide give us this feature on Sheets also?
Please? It would not be so bad if I could simply point my forms to new tables and everything would work. I have to literally rebuild everything from scratch on a template that took two weeks to write.
