Big news: Grammarly is acquiring Coda!

[Stefan_Stoyanov]

I generally agree with your POV and mine is not that different. In business we make decisions, then pivot and deal with it. My issue is with the lack of openness, trust and honesty.

@Bill_French2

you picked the wrong horse (for your interests) long before this merger.

The feeling is that I was deceived. How much? Well, in 2024 I started only 1 new serious Coda doc and stopped promoting the idea of using Coda several months ago. Already started looking for ways to migrate all important docs to other tools.

@Bill_French2

Coda is not impervious to economic realities, and as skilled technologists, we each have a duty to ourselves and our constituents to make rational judgment calls concerning the probability of sustainable platforms we build upon.

100% agree! But if I was a Codan I would not trust a leader who is deceiving the community brothers because he will likely do the same with me on the next pivot.

@Bill_French2

If you think there’s a greener pasture, you need to hop the fence and when you do, please share with us the GPS coordinates of that pasture.

I certainly will because I am thankful to you, Piet, Paul, Max, Richard, Christiaan, Ed, Al Chen and many other community members for helping me grow. I value the community more than the tool, to be fair.

@Bill_French2

What’s your point?

The weak AI capabilities of Grammarly and how limited it is compared to other LLMs. That’s all I meant.

[Bill_French2]

@Richard_Kaplan

what does Coda have to do with Grammarly?

I think I explained it here, and @Piet_Strydom also did a fine job here.

[Nikola_Lajic]

I understand some people are excited for changes something like this might bring. I am an early adopter by default, and if this affected only my personal account I would be looking forward to what happens as well.

But from the perspective of a company that has used Coda as a framework to facilitate business processes, if there is a possibility that costs might go up by triple digit percentages, or stop working for our use cases because Coda decides to pivot, we expect more transparency about the roadmap and upcoming changes.

[Aymeric_Matencio]

Yet ChatGPT can do a much much much better job right at this time with Coda formula helps.
Yet I don’t expect that people at open AI even tried to train their latest consumer model on Coda formula… So I think your argument, while not untrue is not correct.

[Tamerlan_PRO]

@Bill_French2

There are about 200 billion lines of solid open-source code from which to train LLMs about Python. There are near zero lines of Coda’s formula language in open circulation. Many unfamiliar with the data necessary to train LLMs assume it’s a simple matter of providing the CFL documentation, and the LLM magically understands how to use it in millions of cases. It’s not that simple.

So, just a few posts ago, you were fantasizing about how incredibly AI’s potential could unfold after a merger between Grammarly and Coda, yet you justify the poor performance of Coda AI by saying it lacks enough data to train and do its job well, and that it’s not so simple. Meanwhile, ChatGPT handles writing formulas for Coda several times better than Coda AI! So, ChatGPT has enough data to write Coda formulas, but Coda AI, which literally has “Coda” in its name and works within the platform, can’t manage? Is it really acceptable for an AI tool, designed specifically to enhance productivity within the platform, to fail at probably the most obvious task, considering that formulas are literally woven into the essence of Coda?

[Tamerlan_PRO]

@Bill_French2

• Imagine this popup UI with the ability to tap Coda documents, tables, and even CFL inline in every app you use.
• Imagine Coda at your fingertips 24/7 in every task to influence everything you write.
• Imagine the ability to capture content and context from every place you work directly into Coda.
• Imagine not just a grammar copilot everywhere you type – but a comprehensive data and content copilot as well.

All of this sounds like wishful thinking about how to integrate a “smart home” system into a house (Coda) that hasn’t yet been fully built or refined, hoping that it will somehow compensate for all the shortcomings that this house currently has. Moreover, this “house” already has its own “smart home” system called Coda AI, which, as practice has shown, doesn’t work that well.

All these lofty words and reflections on the prospects of a merger are nothing more than your personal interpretation, filtered through your own lens of experience (perhaps rich and highly professional). However, in reality, it will never become true value, for the same reasons why installing a smart home system in an unfinished house is likely to fail.

[Bill_French2]

@Aymeric_Matencio

So I think your argument, while not untrue is not correct.

What sources would OpenAI train on?

[shishir]

Thanks for all the feedback — very much appreciate it! I hear your concerns about the future, and honestly, I’m energized by how passionately you care about Coda.

To that extent let me emphasize a few things:

1. The Coda product is the center of this partnership: It’s not going away, it’s the main part of this deal. We will accelerate (not decelerate) our pace. Coda has an amazing community (50k+ teams, millions of users), and I have some very exciting plans for the core docs-as-powerful-as-apps surface that you know and love. That includes finishing some things you’re already excited about (did anyone ask for sub-doc sharing?) as well as some new things that you may not have thought of yet that I’m really excited about.
2. Re AI — Everything is not AI, and we know it: I’m an AI optimist, and generally believe that it can have a huge impact on work when seamlessly blended into our workflows. But don’t worry that we would “just focus on AI,” I have a super interesting roadmap in mind on core Coda (and Grammarly!) features as well. We know very well that much of the Coda secret sauce lives in our flexible surface, tables, formulas, automations, Packs, and all the features that connect them. I see AI as an addition, not a replacement, for those core competencies.
3. Re Pricing — No changes planned: Reading some of the stories here on how Coda’s unique Maker Billing approach has impacted your ability to scale your Coda usage is heartwarming. Pricing is a hard topic — and not one I take lightly. I can assure you that there’s no changes planned here now.

It’s the holiday season and a good time to ask for gifts. So perhaps I could ask for your holiday wishes of the new combined Coda+Grammarly? For the AI optimists, what are your dream scenarios? For the less-enthused-about-AI crowd, what’s on your wish list for the joint companies to focus on instead?

Feel free to keep posting here, I’m listening.

[Jean-Marie_FAVRE]

A huge thanks for the explanation. On this forum I don’t see haters and lovers, but just Coda lovers.

[Bill_French2]

@Tamerlan_PRO

All these lofty words and reflections on the prospects of a merger are nothing more than your personal interpretation

Well, lofty as they may seem, for a year and a half, GrammarlyGO (mobile) has been transcribing my audio notes on iOS and dumping them into the “unfinished house” you refer to as Coda.

From Your Lips to Coda's Ears

Voice notes and composition anywhere, anytime, and from your phone. Recently, I explained how I use Tana Capture to capture voice notes from iOS to Coda. This alchemy of tools makes it possible for me to dash off a voice note and I can manually nudge through a workflow such that my voice appears as transcribed text in a Coda document. There are many issues with this workflow: I need to push notes like this from Tana to Coda via a Tana-based Supertag and into Coda via a designated webhook. Th…

[Jean-Marie_FAVRE]

access control would be a nice feature

[Rohan_Mitchell]

Quite frankly @shishir, more than anything else, this response has me seriously considering leaving Coda … once again. I left before after the incredibly poorly executed introduction of pricing in the first place, and the same is going on again.

Oh, it’s a good time to ask for gifts is it? Well, how about the gift of a roadmap so WE can be excited too?

@shishir

as well as some new things that you may not have thought of yet that I’m really excited about.

@shishir

I have a super interesting roadmap in mind on core Coda (and Grammarly!) features as well.

These are all just meaningless, empty words. So, you have some ideas in YOUR head that YOU’RE excited about. Big deal. Do you really think saying that is going to get users excited also? Particularly when there is so much uncertainty? At the very minimum you could be saying something, anything, about when you’re going to provide a roadmap - but I suspect you can’t say that because you’re not going to right?

As someone else pointed out, if you don’t communicate, then everyone is left to drawing their own conclusions, the majority of which in this thread are leaning to the negative … and the best you can come up with in response is empty non-specific, no-detail fluff.

@shishir

I can assure you that there’s no changes planned here now.

… and so said every politician on the campaign trail. I admit, this a hard one. Quite frankly, you could almost call me an advocate for a pricing increase because I love Coda (the product, still not the company) and I, along with others, see the patterns of behaviour over recent years as signalling its demise, so I would rather pay more and have Coda survive. The trick would be doing it in a way that does it well - both in terms of getting the balance right, but also in communication and consultation (but that seems incredibly unlikely given how poorly Coda still communicates with its users, as demonstrated yet again in the last 24 hours or so).

So now I have gone from being cautiously optimistic to seriously reconsidering whether to continue to use Coda, or look at alternatives. Fibery.io has come a long way and are a genuine competitor in terms of feature set … as a bonus, they seem to have a clear vision of what their product is about, and they communicate with their users. But like Coda they are very poor marketers.

[Bill_French2]

@Tamerlan_PRO

Is it really acceptable for an AI tool, designed specifically to enhance productivity within the platform, to fail at probably the most obvious task, considering that formulas are literally woven into the essence of Coda?

Of course not. And I don’t recall ever saying Coda AI was good per-se.

Coda should endeavor to support CFL generation within, and I believe they were working on it when they lost their chief AI scientist. Since then, Coda AI has not moved forward which is unfortunate.

[Rohan_Mitchell]

That’s right, no “haters”, just users who love and have invested a lot in Coda and quite understandably have reasonable concerns about what such a major change will mean for their investment of time and money. IMO @shishir’s response does little to nothing to assuage those concerns.

[Stefan_Huber]

Coda has always shown immense potential as a “doc-as-app” platform, offering capabilities no other competitors match:

• A database structure simpler than Excel, yet more dynamic than Access.
• An interface precisely adaptable to user roles and permissions.
• A versatile formula language that can make any object programmable and action-ready.
• A robust ecosystem of integrations (Packs) enabling seamless data exchange across services. Not forcing users of specific apps into coda, but allowing doc makers to still work and interact with this data.
• A pricing model that empowers creators to share powerful docs or apps with their teams, clients, or the public.

These strengths could already make Coda the ultimate place to build and distribute interactive business solutions. However, for years the community has requested small but critical improvements, changes that would multiply Coda’s value and usability. These pleas have gone unanswered while development resources were invested in AI features that currently provide negligible benefits to serious Doc Makers. This misplaced focus leaves experienced users discouraged, frustrated, and worried that Coda has given up from its core mission.

A prime example is Coda AI’s inability to understand its own formula language. If Coda AI were trained on that syntax, it could guide users through complex logic, making it far easier to build robust, app-like docs. Instead, it constantly hallucinates and misfires, failing at even the simplest coda-specific instructions.

Other long-standing community requests remain unaddressed, for example:

• Allowing published docs to function as genuine apps by enabling button-driven actions and edits without requiring login.
• Ensuring hidden pages remain truly hidden, neither appearing in search results nor revealing filtered data.
• Allow doc makers to disable AI where it can give insight into data that shouldn’t be revealed. Data access management is a huge topic for Coda already - I can’t image at the moment where things would go if Coda doesn’t address these issues immediately.

By ignoring these essential features, Coda misses a huge opportunity to unlock its full potential. Addressing them would elevate it from a promising tool to a platform that everyone, enthusiasts, professionals, and new adopters alike could confidently champion.

AI can still play a transformative role if integrated correctly. It’s not about flashy features, it’s about the functionality within Coda that genuinely enhances each level of the platform:

1. Databases: Coda AI already does a great job structuring unorganized data into tables. With deeper integration, it could rapidly convert raw, unstructured inputs into functional, relational databases.

2. Interface: Instead of making users hunt for keywords, AI could let them query the database in natural language and receive precise, context-rich answers. It could also assist in creating new records, streamlining the data-entry process and making the interface more user-friendly.

3. Formula Language: If Coda AI could reliably understand and generate correct formulas, it would dramatically lower the barrier to becoming a Doc Maker, encouraging more users to fully utilize Coda’s programmatic capabilities and stick around after the trial phase.

4. Packs and Integrations: Intelligent AI-driven data transformations could handle mismatched data formats on-the-fly, ensuring smooth data exchange across services. Automating these conversions could make Packs even more powerful and the overall workflow more efficient.

These aren’t pie-in-the-sky features, some of them are well-documented, long-requested necessities. Listen to the community. Honor longstanding demands. And people will look forward to AI!

Oh and before I forget: People build business with Coda. If you wan’t them to continue to do that or winn new customers, Coda has to become more transparent with their technical roadmap. In the last 3 years I started to get more and more uncomfortable with the fact that I wasn’t able to know anything about the future of Coda. I mean I was building app like docs, that people always wanted, loved and now need… What will I do if Coda just decides to pivot completely to AI or suddenly charges for Editors or viewers? Please share your plans in advance!

Thank you anyway for creating Coda - the most amazing App I’ve ever used!

[Piet_Strydom]

Less on the AI side, and more on the “traditional” side.

It would be great to be able to build docs with the following functionality:

1. The ability to create a document that a maker can lease to customers and be able to continue providing content after the customer has made a copy of the doc.
   Current sync pages gets close - I can have a content doc, and then sync the content from that document into the customer’s copy. For example, a habit tracker. In the content doc I can add, over time, say, content over the psychology of acquiring habits.
   What is not currently possible is to allow the creation via a button of a new habit from the content doc into the customer doc. It is not possible to (natively) charge the customer a regular fee.
2. The above was about sharing data with customers. What would also be nice is the ability to upgrade functionality available to the customer in his copy of the doc.
   Currently I can achieve this by giving a customer editor access to the doc. This will allow them to use the functionality, and add their data.
   The downside is that then the owner of the doc has access to their data. So ideally a customer has access to a (copy of a) doc, but changes they make to the doc is not synced back to the source doc. An example here would be a doc to allow people to do financial planning. You want to be able to add additional functionality, e.g. tax or other legislative changes. But the customer does not want to give the maker access to their personal information.

Due to the extremely tight coupling of data and function in Coda, it could be that 1. and 2. are two sides of the same coin.

P

[Jennifer_Biggs]

This announcement has freaked out the Codan Community, which has been expressing its concerns for over two days now. I find it disconcerting that Coda has not offered any reassurances to ease the panic.

[Piet_Strydom]

Shishir shared a blog with announcement, and commented above.

What more do you you want?

[Rohan_Mitchell]

Something that actually speaks to the concerns of the community rather that just non-specific, empty words that don’t actually answer anything.

[SALEM_MALIBARY]

if i wish for something, i would wish 1 seamless frictionless consistence experience across all platforms. this include desktop app, website, tablet, phone and so on…

just to illustrate my wish… see app like : Krita, MediBang. Those are two desktop apps which are mainly for drawing, so in instead of building alterntive apps for tablet, phone they just kept the desktop app with two modes tablet mode, and full desktop mode … so this way whenever the user want full function of the app they just swtich to the desktop mode and when they want light friendly version (which coda’s app now) they just tablet mode. I think by doing this in Coda you will :

• get focus on one seamless experience with all type of users in mind
• get more users since many people want mobile version
• you can drive more sales especially if the app is paid or some aspect of it…

i think the worse thing Coda lack rightnow : is not appealing for the avarage user. anyway thing is a great move. I was expecting this kind of move in future. I hope grammarly work with you with this seamless expierence design and they have something similar not the best but they have one at least. Good Luck guys