When will Quicken finally get a major performance overhaul?

TheGolux

I've been a user of Quicken for decades. I'm a fan of the overall approach, and have (currently) no difficulty with how it models my financial data or the interface FUNCTIONALLLY.

The problem is performance. I have done all the hacks. Try to archive accounts, validating, super-validating, blah blah blah. Quicken is fine with small quicken files. But when the files get large, with hundreds of accounts and tens of thousands of transactions (especially investment transactions) it just chokes.

It is annoying that quicken keeps adding feature updates but seems (so far) unwilling to undertake the work to overhaul the un-sexy tech debt problems underpinning painfully poor performance.

As a software engineer myself, and one who often has to deal with tech debt, I can see that there are two major tech debt problems. One is the interface with the Windows OS and in particular the mechanisms for repainting on the graphics driver. The screen flashes on and off, windows and parts of windows are unnecessarily repainted multiple times, and the graphics rendering itself is inefficient and low-fidelity. I'm on a machine that can flawlessly play video games in real time, but Quicken flickers and halts and chokes, even when it is the ONLY interactive process running. The other big tech debt problem clearly is the efficiency of dealing with its own data model and database. It is clear from looking at the metrics that Quicken doesn't scale well with its own data implemented in its own files. Access is slow and cumbersome, and if you look at the disk reads and writes, there is WAY too much going on. My hypothesis (which of course I cannot prove) is that Quicken is using its own database technology for file access written years ago, and not up to current technology standards and not employing efficient data design and data normalization.

I realize that fixing this kind of tech debt is expensive and time consuming, and difficult for QA, and many customers won't notice that the upgrade adds a feature or fixes a bug. In fact in the short term, it is likely to add some regressions. But it is really time for Quicken to get a major infrastructural overhaul to deal with the incredibly inefficient and poorly performant issues. We are paying subscriptions now for this software. Please divert some resources into that.

mshiggins

Hundreds of accounts?

TheGolux

Yes, hundreds. Some closed. Many are property accounts. Many investment accounts.

Chris_QPW

In my opinion this is never going to happen, at least not more than maybe addressing some localized problems as they are identified.

Quicken Inc doesn't have the resources to take on what you are suggesting. What's more it is my belief that in fact you are targeting the wrong parts of Quicken that have the performance problems. People keep stating they think the problem is in the database, I don't think so at all. Personally, I believe 90% problem is in the GUI, and another part of it might be in that they should do less calculations on the fly. Just like you I can't prove that, but I do have plenty of indications that is a fact. And note on the GUI one of the reasons that "window scaling on high resolution displays" is such a hard problem for a program like Quicken is because Quicken has so many different GUI APIs in it.

For instance, where you might find slowness in investment transaction entry of a single transaction, Quicken has no problem moving thousands of transactions to another register in the blink of the eye. Not to mention it can go through years of transaction history and produce a report in a pretty fast fashion. That isn't a sign of a slow database. I think the problems in the investment account are the GUI and perhaps them calculating all the security lot information every time anything changes.

And from what I know yes, the database is very old, but it wasn't created by Intuit, some have suggested it came from Oracle. But what it definitely is, is a database engine that has long ago had support dropped for it.

Some people might think that you can just rip out the old database and put in a new one. I seriously doubt that. I would bet that it is not nearly that clean of an interface with the rest of the system. The same goes for the GUI. I'm almost positive that a good portion of the GUI code has system functionality and database query code mixed into it.

What's more people (including Intuit) way underestimate the complexity of Quicken's feature set. Intuit rewrote Quicken Mac starting in 2007 and put out Quicken Mac Essentials in 2010, and it was so defeatured that no would buy it. To this day that project isn't completely finished.

Quicken Windows has tons of feature only because it has been around for so long and because they didn't take the time to go back and change everything every time Microsoft decided that there should be a new GUI standard. Personally (as software engineer) I find this kind of behavior to be a pain, but I also understand it from a business standpoint, people pay for features, not bug fixes that they can't really see much change in. This has a big dark side, like having lots of features that haven't got the proper support in many years. But I haven't seen anything to suggest that users have stopped requesting new features. Yes, you get requests like yours that go like "They should stop putting in useless features and fix Quicken." That are usually followed quickly with statements like "But XXX feature I want has to be implemented NOW."

One person's useless feature is another one's essential feature.

TheGolux

You may be correct Chris - and mostly I agree with your analysis. (Notice I led with the Graphics issue). Whether they have the resources or not, and whether they are willing to address their tech debt is a different issue. The company I work for has similar tech debt issues. And we are working to address them. I would hope quicken would do the same.

Chris_QPW

I hope so too, but I'm not too optimist about it happening.

NotACPA

https://community.quicken.com/discussion/comment/20383068#Comment_20383068

Q simply isn't the right product for you with that kind of volume. You need a commercial, not PERSONAL, finance manager.

TheGolux

Do you have a suggestion for the "right" product? I'm not a business. I'm a person. Maybe I've moved alot. But point me to something you think will work, I'm happy to consider it. That said, Quicken markets itself to people at my volume. I believe my biggest issue is not the number of accounts, but the length of time on my investment accounts. Quicken doesn't have a good way to archive investment accounts to maintain lot and cost basis information over decades.

NotACPA

"many property accounts" isn't Q's market. And what Q build are you running? The more recent versions have an Investment Archive function.

TheGolux

Yes, I have worked extensively with support, and yes I do have the latest version, and no one at Quicken has even remotely suggested I am outside their target market. Including trying the new(er) investment archive, which still can't archive transactions if you want to keep cost basis and lots in check. I have pointed out ways that the software COULD do that archiving, but it doesn't.

But the point of this post isn't about "how can I get quicken to limp a little faster". The point is that the both the graphics and the database technologies employed are decades old and I would encourage quicken to improve them.

Chris_QPW

Archiving works based on a few long-understood attributes of Quicken. Quicken investment accounts with a lot of securities/transactions/security lots will be slow (Like I said above, I think this is a combination of the GUI and maybe them recalculating things for every change). And Quicken doesn't read in an account until you access it (and once it does it caches in memory, one more reason to believe that the problem isn't the database.

So, before Archive showed up the recommend workaround was to use the "Shares Transferred Between Accounts" action. This isn't exactly ideal because what it does is put remove transactions in the old account and adds (with the cost basis set) in the new account. Fine for taxes, but not for any of the "return" fields and the others based on them in the portfolio view because they work off of the transaction in a given account. Not to mention all the extra transactions introduced with this method.

That is where Archiving comes in. When they finally found a way to implement Move Transactions, Archiving is just a couple of steps past that, and so they put out both features together. Since the actual transactions are moved the "return" fields should now be correct for those security lots. But this only works on closes security lots, and as such if one is the kind of person that never sells anything, then this wouldn't help at all. The whole point is to transfer the bulk of the transactions to a register that you will not open very often.

By this same token Quicken SuperUsers have always had the viewpoint that "active traders" will not be happy with the performance of Quicken, they just accumulate too many transactions too quickly. On the other end of the spectrum is one that I mentioned above, but almost no one talks about and that is a person that never sells anything and has been using a given account for a lot of years (no closed lots to transfer).

As for the number of accounts, other than there being a limit of 499 (document says 500, but recently a person hit this limit) the number of accounts shouldn't affect the performance at all. The number of accounts isn't what I would use for judging of Quicken is right for someone or not. Quicken is database driven, and as I said a given account isn't read until it is accessed, so the number of accounts really doesn't matter in most cases. What's more in this day and age if the person is using an SSD, I doubt the amount of data that has to be transferred isn't important at all. It is more about how slow it is to load into the register. Let's face it, my data file isn't huge, but it isn't small either, 166Mb, but one could read that whole file into memory before you could blink an eye. Just for reference I have 133 accounts and have no performance problems. But I also note that I don't have that many transactions/securities/security lots in any given investment account.

On the other hand, if you are looking for Quicken Inc to tell you that you don't fit well in what Quicken was designed for you are fooling yourself. There are definitely built-in assumptions about their user base, and it won't work for everyone, but they aren't going to tell you that.