Having worked at Intuit, I'm always amazed at the number of people that look at how crap Quickbooks is and think that beating them is a matter of just making better software. There's some really important psychology that most competitors completely miss.
First, Quickbooks is frequently not chosen directly. Many, many small businesses choose their bookkeeper first and use whatever they use. One trick Intuit uses both in Quickbooks (bookkeepers) and TurboTax (CPAs) is to put a lot of effort/money into focusing on those relationships knowing that they have a huge indirect impact on acquiring and keeping customers.
Second, the fact that Quickbooks is so awkward and inconsistent to use isn't a UI problem, it's a feature. This is an important lesson in designing software that people use for their jobs. When you make elegant, intuitive software that almost anyone can use in short order, it stops being an impressive item to put on a resume. And it stops being a barrier to entry for competing professionals. Software like Quickbooks that's hard to use correctly becomes a selling point for bookkeepers and almost accomplishes the same purpose (though to a lesser degree) than professional licensing organizations...it limits the competition and keeps the rates they can charge high.
So many Quickbooks competitors have failed to knock them off their perch by not realizing that their customers aren't really their target audience and what should be their target audience doesn't want easy.
If I may, your point is exactly what the taxi drivers in NYC were making about Uber et al.
Still I think there is a misunderstanding on the concept of difficulty here. When the first Quickbooks version came out there were probably not many intuitive bookkeeping software. People/SMB/Accountants loved Quickbooks. And then there were demands to add more features. Now Intuit could either redesign the flow and ask people to re-learn or just tack it on with another new page. Guess what? Most people are resistant to change and learning, especially enterprise level. They are more about saving their jobs and asses than efficient process and software. So, Intuit just tacked on the feature and called it a day. All parties were happy - Intuit making money out of a low effort solution and users being happy of not having to re-learn everything.
But...there is a problem here and Intuit is stuck in a bad place. As newer breed of SMB/Accounts etc get into the market they will be frustrated by how the software behaves. It will take couple of newbies, even maybe someone on this very thread, who grow up into well-known CPA or bookkeepers etc to reverse the tide. And Intuit will go the RIM way, not knowing what hit them because their customers will always say that the difficult to use software is the best. Or even Oracle's way, scrambling to introduce a quickbooks for the newer generation, so to speak.
> They are more about saving their jobs and asses than an efficient process and software.
the less cynical view is that moving from a known process to an unknown one presents costs, and the movement could actually result in a less efficient process.
The inverse of this is always trying to move to the most efficient thing all the time, in which case you're going to spend all your time trying to "upgrade" your process, and losing the time too that coordination. You are supposed to do this for your core compentencies! Always improve the things that need improving! But you know what? Most companies don't have book-keeping as their core competency. Book keeping is a thing a company does just like brushing your teeth is a thing you do. When was the last time you optimised your toothbrushing process?
The role of book-keepers and accountants isn't just about "be efficient at accounting", it's almost more about making sure all of this happens transparently to the rest of the company in an efficient way. If you find some thing that cuts the accounting work in half but you have to inform 20k employees on changes, you've just introduced a lot of work for something that (if you're improving all the time) will change soon anyways!
The lesson is that if you're making software you need to actually make people's jobs easier in the entire context and not just optimize on some technical metric. The reality is most people in these roles are overworked, and this accounting software won't mean you don't need an accountant just like Word doesn't make you not need a lawyer. Until it _does_ mean this of course.... but for the most part you need to actually make something that helps people without generating more work.
I have sometimes worked with accountants, overworked ones too, and agree to the larger point that change in processes are painful and costly. My experience has been that accounting law changes nearly every year. And, I find that instead of doing it right, most of them tend to go for the easier option - add a page. Now, this eventually means the software is heavier and cumbersome to use. So, year-over-year software is just adding to the cost of teaching a new person how to use the software while avoiding the cost of changing a process.
Hence the need to write an optimized software. It seems that not many fields have a concept similar to technical debt.
I fully agree with your first point. However, I don't think Quickbooks purposefully designed their software to be hard to use. I think that just happened when they grew their software organically without spending much effort and rework with good UX designers.
It's more important that B2B "work" software works accurately, correctly and ticks all the required features, than that it is intuitive or beautiful. A steep learning curve is a problem for B2C software, not so much for B2B software that is used intensely.
They may not have purposefully designed it originally, but they purposefully keep it hard to use now.
Internally, they have Quickbooks training courses that new employees and employees who work on other products can take. When I took the course, I asked why deleting a transaction was a completely different UI interaction from deleting a contact in QBO. Both pages were, at their core, list widgets so both should share UI, right? But, I was told, they'd done UX tests with experienced users who had strong reactions to making UI interactions like that consistent.
The problem isn't lack of good UX designers. When I was there, they hired Frog design to completely rethink many of the UI interactions. This was in addition to many talented UI/UX people that work on the product full time. It's a product that brings in billions of dollars a year...the resources are there to fix it. But the problem is an existing product and a user base that's resistant to change and one that views their proficiency in the product as an achievement that they'd like competing bookkeepers to also have to attain.
This is strongly embedded in the accounting culture (CPA, in particular tax preparation business): I want every man, woman and child to understand how close we are to chaos. I want everyone to remember why they need us!
> But, I was told, they'd done UX tests with experienced users who had strong reactions to making UI interactions like that consistent.
Had they strong reactions to those being consistent (and thus easier to learn), or did they have strong reactions to those being different from what they were used to?
These guys aren't a Quickbooks competitor. They use Quickbooks online to do their customers' bookkeeping. They're a technology-enhanced service business.
I really hope you weren't a product manager at Intuit.
>Second, the fact that Quickbooks is so awkward and inconsistent to use isn't a UI problem, it's a feature. This is an important lesson in designing software that people use for their jobs. When you make elegant, intuitive software that almost anyone can use in short order, it stops being an impressive item to put on a resume. And it stops being a barrier to entry for competing professionals.
A good professional product does not have "awkward or inconsistent" UI. It has consistent, useful UI that can expand as user's sophistication grows. i.e. it has a learning curve
The curve for learning an "awkard and inconsistent" UI you highlighted is useless and a detriment to SMB's everywhere. A proper SMB accounting approach would to have standard patterns for doing simple accounting with the ability to grow and extend as sophistication of your org grows.
I was a dev manager for a team that was working on a Quickbooks-adjacent product. I had a lot of meetings with core Quickbooks teams and met many of their PMs. They are, for the most part, really good at their jobs.
Your comment comes off as incredibly naive. It's the same naivete that I had before seeing the internal perspective and it's what my comment was addressing. There are psychological phenomena in play that make an existing, market dominant piece of software more successful and harder to displace when it's objectively harder to use when that software forms the basis for someone's job.
And it's important to realize this because this is counterintuitive to what those of us that create software for a living have been taught. We're taught to think like you. We're taught to make software that delights our users and is as easy as possible to use. But understanding a specific set of circumstances when that approach is destined to fail can be important.
>There are psychological phenomena in play that make an existing, market dominant piece of software more successful and harder to displace when it's objectively harder to use when that software forms the basis for someone's job.
For the record and from the GP's context, I'm going to assume by 'harder' you mean 'difficult to use UI'.
I'd assert 'difficult to use UI' is the incorrect type of 'hardness' or 'difficulty' that should be present in an enterprise piece of software. Mainly because fighting a UI is non-value add to the goal of the user. The user's goal, in this case is to manage their books.
Because of this tension of QB _purposefully_ being at odds with the user goal, I'd expect start-ups and new software products to gobble up QB market share over time. If this is the case, all QB can do is make it more difficult for current users to switch off them (.QBW file extension?) as their market share slowly dwindles and their channel sellers slowly disappear.
Good 'difficulty' should come from the learning curve it takes to master (i.e. accurately, quickly, and reliably) the task at hand.
FWIW I'd expect purposefully building difficult-to-use products has many negative externalities for the company culture and this would be a big red flag for any future Intuiters.
You're still not getting it. A Bookkeeper's goal isn't to keep the books. That's their product, but they have their own small business that has more fundamental goals. Accounting software that's easy to use makes it easier to keep the books, but it runs counter to that more fundamental goal of running a successful small business. You're right that there are negative externalities. Small business accounting costs more than it needs to and, by extension, small businesses need to charge customers a bit more. But the people deciding which accounting software to use are the ones benefiting from that negative externality, not the ones paying for it. And as a software vendor, you always target decision makers over users or you go out of business.
I don't think anyone is misunderstanding your point, I just think its ridiculously backwards. It's the equivalent of the government breaking roads so they can create jobs fixing those roads.
If software can simplify bookkeeping and make the profession obsolete, and you deliberately make your software in a way that keeps that inefficiency at play...you guys are gonna wake up one day and realize you spent all your time serving the bloated middleware instead of the true end user, and your market will evaporate before you know what happened....
We get that from a short term perspective, it may make sense to keep it too complicated for the end users, but efficient markets will prevail short of forced intervention.
If that is how Intuit thinks, I cannot wait for the day when they get displaced.
Yes, it's ridiculously backwards. But it's reality. Intuit, at various times, has tried to make the kinds of changes you guys are suggesting. And every time they've run smack into reality and had to backpedal. Reality is like that...you can try to deny, and fail, or accept it and adjust your thinking. It isn't a how Intuit thinks issue, it's the way their customers think after decades of getting comfortable using Intuit software.
Another example...Intuit has wanted to transition customers away from the desktop version of Quickbooks for many years now. If you think Quickbooks online has it's learning curve, just spend some time using the desktop product. It's a maze that takes years to learn to navigate and many of the "features" are horribly buggy from years of neglect. But pry it from their cold, dead hands Intuit cannot, so you can still buy new versions of Quickbooks Desktop.
That's why what you guys are saying is so ridiculous. You're denying a reality in a field you're not familiar with. You're saying something should be different because it's inefficient or other industries work differently. But you haven't done the user research. You don't have decades of experience dealing with the people who actually spend money on the software. You're in no position to critique the way that Intuit thinks.
> And every time they've run smack into reality and had to backpedal.
I am all for being brutally aware of reality...but, there is also a time horizon to these things. Most great startups had to break through resistance. Maybe the resistance you guys saw was simply resistance to change...
Maybe tweaks to UI isn't what wins this game...maybe its tools like Pilot or services like bench, or maybe its something like the car solving the horse manure problems of the turn of the century...
And to be fair, there are some studies that show cognitive strain being linked to better focus or something like that (Disclaimer: I have no idea what im talking about, but am pretty sure Thinking Fast and Slow discusses it.) so, maybe a harder UI is better...
What I do know is that I want software to make my life easier and more efficient. And I don't mind if that takes away part of a means of my livelihood, I'll figure out the next step the same way I got to where I am today.
If I understood you correctly, the folks in charge of this are very good at their jobs and I am sure they are optimizing for their quarterly earnings, or bonuses, or whatever else they are incentivized to do...and I get the idea that they see bookkeepers as their customers, not business owners, and are serving them as best they know how...
but this feels like the lazy way out of innovating.
- government managers are evaluated on budget and headcount because the core services they provide are hard to measure. Thus they don’t want new software to help them be more efficient, unless the alternative is catastrophic, visible failure to provide their service.
- Lawyers charge by the hour. They don’t want new software to dramatically reduce the time spent in discovery. Then they would have to bill for outcomes (cases won) and there are too many things out of their control.
Basically nobody wants to charge for value, because value is hard to quantify accurately and quickly. You wind up showing costs to prove value, so nobody wants to reduce costs.
> government managers are evaluated on budget and headcount because the core services they provide are hard to measure.
That’s... untrue. Government managers are judged (in terms of evaluation of job performance) largely on minimizing both costs and negative public attention.
OTOH, managers generally (and not just in the public sector) like to have big budget and headcount numbers on their résumés as proof of scope of responsibility.
Thanks for sharing this story. This is something I suspected and have actually complained to customer support about repeatedly (things are buried within menus of menus, strange link redirects that don't make sense, functionality that is not available to business owner accounts but is available to the supporting accountant-logins!!). These terrible patterns that keep users from more easily managing their books and the awful lobbying of congress to keep taxes complex are why I think people should actively boycott Intuit software.
Just a note, not to argue any of your substantial points, but Intuit announced about a year ago that they will no longer develop a desktop version of Quickbooks for Mac (leaving v. 2016 as the last version, officially supported through May 2019). They continue to make a version for Windows.
> If software can simplify bookkeeping and make the profession obsolete
The GP's point isn't that software can simplify bookkeeping but that bookkeeping itself is not simple.
Any company (especially startups) that gets to scale eventually outgrows Quickbooks. You don't see any F500 or any large scale startup ($100M+ in revenue) using QB because it simply isn't built for those type of needs. But there are swaths of bookkeepers who understand how it works and how to migrate their profession to NetSuite/SAP/Intacct/etc.
QB is the bookkeeping equivalent of Excel. It does it's job well and power users benefit massively from it. Pitch and Bench are competitive, but the aren't great fits for someone who will eventually need more professional finance or uses a tax accountant that is more efficient with their tool of choice.
In isolation that sounds fantastic. When looking at the greater market dynamics, it's not at roses. Professionals tend to form opinions in the interest of their self preservation, whether consciously or otherwise.
By shaving off the time it takes to keep the books, a percentage of clients will decide that the time it takes to keep the books isn't worth outsourcing. The shorter you make the time commitment, the more the cost/benefit analysis leans that way.
You've also made it easier for other bookkeepers. Now they have more room in their day for more clients, plus the barrier of entry for new bookkeepers is powered by the easier software. All of a sudden you have negative pricing pressure driving your rates down and next thing you know your client base has gone up but your rates have gone down and you're making the same amount of money but with more client management overhead due to additional clients being serviced.
Then at some point, the ease of use shrinks to the point where a critical mass of the client base have decided to do it themselves, and you start finding it hard to source enough clients to fill all 8 hours a day, and have yet more negative pricing pressure to make up for it. What started as making your job more efficient evolved into making your job so efficient you weren't needed.
Quickbooks isn't slow to use, it's unintuitive. That's a crucial difference to your calculation. An experienced Quickbooks user can still work very quickly by internalizing the wonkiness. The inefficiency is suffered by novice users.
In the UK the SMB accounting market is owned by Freeagent and Xero, which are both really easy to use. We used QB for a short period of time. It was a terrible piece of software and we couldn't dump it fast enough. They are still entrenched in the US but I wouldn't expect their so-called competitive advantage of "hard to use" to last.
I agree with this. Good software updates that I've resisted in the past have often surprised me at how much I prefer them once I acclimate to them.
The alternate line of thinking, that nothing big about the UI or system can ever change unless a customer specifically asks for it, seems to conveniently pop up in teams with enormous technical debt that make big changes unthinkable for managers. I'm guessing this is the case for QuickBooks - given the years they've spent supporting an inconsistent system, apparently with a goal of preventing it from becoming consistent, while also inserting customer-requested features.
I have a hunch that if you're afraid of inconveniencing customers by making the software better, then the real problem is that you're afraid users won't stick around, and it's because you think that you haven't taken good enough care of them to weather the bump of re-acclimation. Compare the unchanging-system strategy to Snapchat's. Snapchat regularly improves their software (and users complain about it until they re-acclimate), because they are confident that users will stick around through updates; this confidence comes from the fact that they know they do a good job serving their users. So, my hunch is that when this unchanging-system strategy emerges, it's likely because you've already spent a long time disappointing users and eroding customer confidence.
I work in accounting and completely disagree with your statements.
Quickbooks is VERY easy to learn and use. It is almost like word where you can kind of just jump into it and learn as you go. That's how easy it is to use. Compared to other "higher end" accounting software packages, Quickbooks is a dream come true. That's why people prefer to use it. It is easy and straight forward. Same story for turbo tax. Another big reason so many companies use it is because the available labor pool of people who know how to use Quickbooks is huge which helps to keep labor costs low. If you have some obscure accounting software then you will struggle to find a qualified candidate for a fair wage. Then it eventually becomes a moat for Intuit because everyone is using it and people are hesitant to move away from it - similar to Windows.
Most CPAs aren't using Turbotax, they are using the higher-end professional tax prep software programs. Turbotax is mainly focused at people who want to do their own taxes quickly without having to pay a professional.
Quickbooks isn't that hard, but it's got a lot of inconsistencies. For instance, when I'm viewing a list of customers, I have to drill way down to the individual customer edit screen to make the customer inactive, but when I'm viewing a list of transactions, I can delete one simply from the right-hand menu on the list item. Both are similar list widgets, but the actions you take to interact with the widgets are completely different. It's not hard once you learn it, but you have to learn how to use each screen individually instead of learning patterns that can be applied across different screens.
Intuit's pro tax offering is called Lacerte and it has many points of integration between TurboTax and Quickbooks. To Intuit, the relationships they have with CPAs and bookkeepers don't apply to a single product. Not having worked in the tax group, I tend to be careless about saying TurboTax when I'm referring to the larger tax group, but you're right that I probably should have been clearer that most of the CPAs they deal with are not connected with TurboTax.
I am a small business owner that does my own bookkeeping. After 9 years on Quickbooks - gritting my teeth the whole time - I switched my 4 entities to Xero and couldn't be happier. I am having a hard time understanding your justification of a poor UX and a closed platform. That's not what I'm seeing as a customer.
I use QuickBooks so someday other small businesses / investors / industries could easily review my books. Some areas ripe for disruption: better serialized inventory and warranty tracking, barcode scanner support, QBAR reporting documentation, adding custom columns to printed item receipts, integrating with HR providers (Gusto), importing bank statements for reconciliation in PDF, any marketing or CRM integration, running performance comps, mileage and GPS tracking (Corrigo is step in right direction), verifying customer addresses in invoices.
Some things that strike me as odd: a feature to repair the data file (should I run daily? Annually?). Broken filters that still give reports without warnings they are broken (wrong data < no data). Multi-users in invoices, one saves and the entire system crashes. QBAR takes up to an hour to refresh data and says its on step 141 of 132. OS/2 Warp floppies had more accurate progress meters in the 1990s. If you had a button for “fax this PO to my vendor” it might be revolutionary.
> Second, the fact that Quickbooks is so awkward and inconsistent to use isn't a UI problem, it's a feature. This is an important lesson in designing software that people use for their jobs.
This is an interesting point but isn't there a risk that this will backfire long term? Existing bookkeepers may stick to Intuit, but if a new bookkeeper can learn Pilot.com in half the time, and without a class, won't they be able to get customers at a lower price, and eventually cut into Intuit's business that way?
Edit: And also this doesn't account for DIY customers, for whom any time above 0 is too much spent on bookkeeping, and would definitely value an easier product.
In the world of bookkeeping only accountants apply. Because of the complexity of tax laws you need to be an accountant to figure them out - it isn't worth your time unless you will use it constantly. It is a boring an tedious subject with legal implications if you get it wrong: you are better off focusing on your own business and letting an accountant take care of these details.
Thus if your accountant asks you to enter your spending into some program you buy that program and enter the details. Every few months your accountant files some forms, and once in a while gives you some spending advice.
The last thing accountants want is for business owners to decide that their accountant isn't cost effective vs. do it themselves. Most of the forms an accountant fills out are automated. The only thing left is audit (the numbers add up, but do they make sense?), and advice.
Sounds like a space ripe for innovation then. Even my accountant struggles with the more esoteric parts of the user interface. If I could find a better payroll solution, I would have already moved on by now.
The point about Quickbook being hard to use as a barrier of entry is correct. It is indeed a feature once Quickbook is already widely used.
On the other hand, should we intentionally make a software product that's hard to use? How does it get off the ground to the point of being widely used in the first place?
I completely agree. The observation is less useful when looking at how Quickbooks is/was built and more useful from the viewpoint of building a Quickbooks competitor. Just fixing broken UI interaction patterns isn't going to be successful. If you want to beat Quickbooks, you need to either replace the accountants that small businesses use entirely or offer those accountants a significantly better value proposition than Intuit does. Because what they have now is working and their livelihoods depend on it. They have processes in place that are built around Quickbooks. The status quo works, so you almost need to be revolutionarily better to overcome that level of friction.
Quickbooks is marketed as being easy to use! And it is, have you ever used some old crappy accounting software?
When software like Quickbooks first came out, your other options for a small business were complicated programs like ACCPAC or stuff that only ran on larger systems like IBM’s.
A nice GUI lets you use the mouse and the other parts of Windows/PC that you already know instead of a cryptic keyboard driven character-based interface.
Interesting insight here. It reminds me of a similar case in video games, namely traditional real-time strategy games like Starcraft. The UX design in these games is pronounced and inelegant, in part because it raises the skillcap of high level play. For professional RTS players, not only is your tactical decision making a differentiatior, but also your ability to manipulate the various UI menus as fast as possible.
> For professional RTS players, not only is your tactical decision making a differentiatior, but also your ability to manipulate the various UI menus as fast as possible.
Is it, though? What original StarCraft did right, and what was likely a big part of what made it popular as a pro game, was that you could do almost everything with the keyboard. Storing and recalling groups of units, selecting commands, abilities, moving the screen around, etc. - all of that was just one keypress away, and easily-discoverable one at that. About the only thing you need the mouse for in pro SC play is pointing and clicking to select units or target commands you've activated by keyboard. In particular, you don't navigate any "UI menu" ever.
Pilot customer here; they are amazing. It's pretty different from a regular bookkeeper largely in that Pilot is super accurate and timely. My time spent doing bookkeeping for my small startup went from an hour or two a week to zero hours a week - those are pretty important hours. Highly recommended.
Yep - all of that is automated. Importantly, all the the classification of the transactions is automated as well, so I no longer have to look at a charge like "GUS99PWR-T BILL.COM" on my credit card statement and figure out a) where it is from and b) that it should be categorized as a marketing/promotional expense.
How do they know that "GUS99PWR-T BILL.COM" is a marketing expense? Do they use the fact that other users have previously classified "GUS99PWR-T BILL.COM" as 'Marketing'?
The first time we see something totally unknown to us, just like a normal human bookkeeper, we'll have to ask you about it (assuming there isn't metadata somewhere else that helps us categorize it).
But then once you've told us, unlike a normal human bookkeeper, we won't forget about it later — we'll actually encode a rule in our software that will enforce this check going forward.
(And yes, you also do benefit from "If our system has seen this type of transaction somewhere else, we can make a more intelligent guess")
So let's say I have a $29.99 expense from AMAZON.COM and tag it as a marketing expense. Then I buy something else from Amazon tomorrow for $44.99.
Are you telling me that it gets automatically tagged as marketing? Because that seems... naive at best, and hugely problematic at worst (as it can result in a huge amount of expenses getting incorrectly tagged).
This is exactly why the human is in the loop and why the solution isn't purely software — to help come up with processes and mechanisms for resolving this sort of ambiguity.
(Every outsourced bookkeeper today has this challenge, and there are a variety of options for resolving it, depending on what the customer requires.)
My guess...they use quickbooks in the background, which allows you to upload receipts. If the description of your amazon receipt looks like furniture, then they'll mark it as such...if it looks like office supplies...office supplies.
I use Wave, which lets you upload receipts and it will perform OCR to extract vendor info, date, total cost. It's not much of a leap to go 1 step further and look at the line items and categorize...again just a guess.
> And yes, you also do benefit from "If our system has seen this type of transaction somewhere else, we can make a more intelligent guess"
Suppose User A has a bank entry "Payment to Stormy Daniels" and classifies it as "Hush Money".
User B also has a bank account entry "Payment to Stormy Daniels". Will the system suggest to User B that the entry should be classified as "Hush Money"?
Yes to reconciling, generally no to preparing/sending invoices, but it depends. (The Pilot Plus tier basically is "Let's talk to you and really understand your needs and put together a custom quote that makes sense given the work you want.")
Generally speaking we like to take on the work of the "monthly close" — i.e. making sure the books reflect what happened in your business the previous month, and we're less keen on doing stuff like following up with your customers who haven't paid, both because (1) that's something you probably want to have happen more than once a month, and (2) there's actually a lot of on-the-ground knowledge that you or the sales rep have that we may not have.
I see that pilot includes Quickbooks essentials. Wouldn't pilot just reference their customers to use quickbooks to handle the invoicing? And maybe some hand holding if needed...but otherwise it's DIY?
Yep, we generally point people at QuickBooks invoices or bill.com, but we have them do the inputting/sending of the invoices via one of those services.
In NZ, UK and Australia we have been using Xero for years, and there is a vibrant ecosystem around it. And yes - it’s turned bookkeepers and accountants into advisors, completely changing their professions.
Xero is listed, worth $3-4 billion and growing quickly.
So how is this different, and how is it going to integrate with Xero, QBO and other cloud accounting software? And why does the article not mention these elephants in the room?
That was my first thought too. Xero is quite good and very large, and it's absolutely in the "modern era". And it's growing a very large ecosystem of accountants and consultants that are building on the platform to offer the exact service that Pilot is (apparently) offering.
Launching a Xero killer is ambitious but doable. Launching some new bookkeeping software (or...service? what even are they doing?) seemingly without being aware of Xero seems...deeply myopic. Xero is big, their product is solid, and they're able to move a lot quicker than the legacy vendors. And this:
> When a company starts working with Pilot, the actual core experience on the customer side doesn’t really change all that much: they still work with a human on the other end. But the bookkeeper from Pilot is working with the internal tools they have built to bring in the data from the company, organize it and structure it, and produce a set of books that are more accurate than someone might have produced than just doing it by hand.
That's a bizarrely accurate description of how stuff works today. That's not new, that's standard. I'm not sure if the article is terrible or the product is terrible, but something is clearly not coming through.
Similarly, all the stuff about categorization of expenses? Again, that's just a core feature in modern bookkeeping software; Xero allows you to configure very detailed rules. And then since the rules can't be 100% accurate in all cases, your actual bookkeeper (the one who uses the software) is going to ask you about the ambiguous ones.
I'm just mystified as to what the value prop is. "Bookkeeping firm raises $15m to provide the same services everyone else does"? How is this tech news? Or even news? (Caveat: My experience is outside the US, maybe this is less common/standard in the US?)
Pilot is not replacing Xero or QuickBooks (in fact, we do all our work in QuickBooks -- and we're definitely aware of Xero.)
Pilot is a better bookkeeper, and the reason we can be a better bookkeeper is that, even with a tool like QBO or Xero, there's still a bunch of incredibly manual work that happens ON TOP of those platforms, and a lot of that work is very well-suited to being done in software.
Do Xero and QBO get better and smarter with time? Sure, yes, of course they do. But the problem is: no one software company (i.e. a company with the ability to develop software) owns the problem end-to-end, so it's never going to be seamless.
(It's not obviously Xero's problem if the Xero/Gusto sync isn't very good, and it's also not obviously Gusto's problem either. Net result: the bookkeeper does a bunch of work by hand to fix it all up themselves--work that's super-manual and tedious.)
That's where we come in; we're the bookkeeper, but we also have the ability to write software when it's appropriate for solving the problem.
> Pilot is a better bookkeeper, and the reason we can be a better bookkeeper is that, even with a tool like QBO or Xero, there's still a bunch of incredibly manual work that happens ON TOP of those platforms, and a lot of that work is very well-suited to being done in software.
Xero is a platform with an extensive API, partner program, app marketplace, etc., and a large ecosystem of partners, including bookkeepers, that offer what sounds to me like exact thing. If you want a better bookkeeper that uses automation to provide a slick experience on top of a modern software platform...I have no reason to think Pilot can't do a great job of that, but that's totally the value prop of the Xero partner program too...right?
If the sync isn't very good and your doing some manual work categorising stuff, why not...write some tools to automate the categorising, and then feed it back into their API?
I'm sure I'm just being obtuse, but every explanation of what Pilot is doing just sounds very "standard". Are you trying to do this cheaper, or better, or...?
> That's where we come in; we're the bookkeeper, but we also have the ability to write software when it's appropriate for solving the problem.
Okay, right, but...Xero has a pretty great API. I've written software that solves problems by working with the Xero API; I didn't need to write my own bookkeeping software to do it. Xero is outright targeting "bookkeepers that can write software" for their partner programs and developer APIs, no?
(Sorry, if I seem like a Xero fanboy, but I do like them, and their offices are right down the road, so I suppose I might feel some local pride.)
As a founder in the UK I was using Xero and a London based virtual-CFO about 8 years ago, and it was an absolute life saver not having to deal with the cruft.
The accountants did the books for most of the good startups in London at the time, so had a good range of experiences from funding to tax credits. Of course, they did this without raising $15M :)
Interesting to see how this seems to be a "new" thing even now in the US market.
We do all of our bookkeeping in QuickBooks Online (and only QBO) today.
In the US, QBO is still definitely the market leader, and we think there are a bunch of benefits for our customers in using QBO. (Bank integrations and other third-party software integrations tend to be better, there's a bigger ecosystem of e.g. tax and accounting firms that are better-equipped to deal with it, etc.)
(And then from an engineering perspective, we can be a lot more efficient if we can focus on targeting one platform.)
Xero is probably my favourite peice of time saving software we use, possibly just my outright favourite. It works flawlessly and the fact that our accountants can login and work with us on something is a huge bonus.
If you're new to being a startup founder, you might be surprised by how much time you must waste on this area. Getting it right isn't optional, but most technical founders have a weak finance background. Spend money on tools like this to get your time back.
Agreed...started using Wave but I feel like I should have used quickbooks.
Though they have payroll, they do not pay your taxes for you (only in a few select states). You need to do it manually...and at times I've forgotten and had to pay penalty. Now I use square - and they do it all automatically. I didn't know I had to 'turn off' payroll in wave and was charged the $15 a month even though I didn't run payroll for any employees.
Payments...their recurring invoices are good, but you need to do some manual work. They create the invoices and it adds to your sales. When they deposit your money to your bank it adds to your sales as well. So you need to find the one they deposit and manually delete. I asked if there's like a report they generate that tells me everything they've deposited...it doesn't exist. So you could mistakenly delete something, or forget to delete something...and it throws everything out of whack. You could be underpaying or overpaying taxes.
Their 'ACH' requires people to log into their bank. I'm not going to try to convince my customers to click on a link on my invoice and put in their bank credentials.
I have other gripes...there's definitely room for improvement. I haven't seen that much changed in the last few years...not sure if they cut back on their development team or what. I probably should have looked more heavily into Quickbooks since that's the industry standard. I will look into pilot, though.
I want Wave to succeed but can't recommend their accounting product to anyone.
Ran into major issues with their accounting product (especially around reporting, and dealing with multiple currencies). Reached out to their team to explain the problem, they agreed it was a significant problem, but didn't really have a desire to fix it. Fast forward 12 months, core problem is still there. My only thought is when your product is free its not always easy to figure out how to prioritize what you work on.
I have found Xero to be the best intersection of function and price for SMB - Quickbooks online also had some significant quirkyness (i.e. one thing Intuit doesn't really highlight is Quickbooks online is not feature complete when compared to Quickbooks desktop - hence there are many desktop features and reports which just don't exist in the online product...)
They're ok for recurring monthly invoices that you run through credit card. I use Intuit Merchant Servics for ACH (traditional checking account + routing id). It's only $.50 per transaction. As I said my caveat is they count your transaction twice (once for your invoice and again when Wave deposits $ into your account). The 'description' is pretty obvious and you just need to delete it manually. I gripe every time I delete such a large transaction. If you use like square for your monthly credit card, then of course Wave's bookkeeping will just count the transaction once...the deposit into your bank from Wave. But at this point, Wave isn't making money off me and I feel a little guilty...but that's their fault. Their invoicing is simple to use so i'll stick with it, for now.
The bookkeeping...besides mint, I don't have any other experience with bookkeeping. I will look more into Quickbooks, though. Business is growing and need to make everything simple.
I've been using Wave for my bookkeeping for years with no regrets. With that being said, I'm a single-person S-corp with no employees and my income is made up of directly-deposited checks in response to weekly/monthly invoices. I typically have one client at a time, as I do software development on a contract basis. In comparison to others, I imagine that my needs are very basic.
”Wave jumps significantly in the online accounting standings this year ... the free service only charges you for payments, payroll, and premium support (at $19 per month) ... Not only is Wave the best free small business accounting website, it's one of the best online resources period for its target small business audience: freelancers, contractors, and sole proprietors.”
I use two Intuit products at my startup: Online Payroll because my BofA account easily integrates with it and Quickbooks to manage the books (because my accountant suggested it). Both products are easy to use.
Online Payroll looks like a '90s UI but works. Some complications like if you have some non-standard paperwork to file with a tax payment then you have to do at irs.gov.
Quickbooks UI is actually pretty slick (perhaps recently updated?) The automatic classification is ok, nothing special. One issue is duplicate payroll transactions due to sync'ing with a bank account and Intuit Online Payroll.
On the whole I spend about 5 min every pay cycle creating paychecks and issuing tax payments. Once a quarter I classify expenses in Quickbooks - about an hour. And once a year I spend a couple hours double checking things with my accountant for the tax return.
These numbers may get more complicated when we're > 3 employees and have revenues.
As someone who's currently looking at Bench and Pilot, it seems that Bench is a similar product, albeit much cheaper once one's revenues start to grow. How does Pilot compare to Bench?
* Our books are stored in QuickBooks Online, and you have access, so you're never locked in. If you need to work with another provider or just don't like us anymore, you can hand your data to any bookkeeper in the world, because it's just QuickBooks (vs. being handed an Excel export from a custom accounting system)
* All we do is startup company bookkeeping, so we've developed super-deep expertise in it (vs. aiming to solve it for, e.g. the bakery or yoga studio)
* We do accrual-basis bookkeeping (vs. only supporting cash-based bookkeeping)—this is a pretty inside-baseball thing, but it matters as the startup gets larger
I don't know how much automation Bench has already built, or if it's all just humans under the hood, but one structural advantage of #1 for us is that it means our engineers can spend all their time focusing on automating error-prone work (rather than needing to build out a QuickBooks clone).
Since you're focused exclusively on startups, is there any chance y'all can write up a "tools" guide for people just starting out, not just for bookkeeping? i.e. "Here's stack that new companies have a lot of luck with, that minimizes headaches and saves time: Pilot, Expensify, Gusto, Stripe etc." Speaking of which, any recommendations for things I can do now that will save me headaches down the road?
Yes! "Pilot recommended financial stack" is definitely at the top of the list of blog posts we want to write because it's just super-nice to get set up on the right stack from day one, rather than having to worry about it when things are more complicated.
The short, spoiler version for now is: Gusto, Chase corp card, Stripe, Expensify, bill.com if you do a lot of invoicing.
(And then the more 101 stuff, like, "Please don't mix business and personal expenses")
Ian Crosby from Bench, here are a few things I think differentiate us:
* Bench launched five years ago and we’ve scaled to be the largest bookkeeping service in America, with hundreds of employees and thousands of customers. We’re not a brand new startup and you can trust that we’re not going anywhere!
* We have lots of happy customers that can back us up that we’re really, really good at bookkeeping.
* We built our own software because Quickbooks breaks when serving clients at scale. Serving bookkeeping to 100 clients is just a very different game from serving bookkeeping to 10,000 clients, and you need very different systems to make sure everything runs smoothly. Far from locking you in, if you ever want to leave Bench, we’ll work with your new bookkeeper to get onboarded to any system of your choice, at no charge.
* We’ve spent years perfecting our hiring and training programs, and building an incredible culture of client service. If you’re hiring a service, the people that you’re dealing with on the other end really, really matter. We do this part right.
* At the end of the day, it’s now the industry standard for any bookkeeping service to do a free month of bookkeeping for you, so you can evaluate whether you like them or not. If you’re debating between two services, I highly recommend that you just try both! If you give Bench a whirl, we’ll turn around a monthly P&L for you in 24 hours that you can use to evaluate whether we’re the best choice for you.
* Pro tip: Sign up through one of our partners (like Stripe) and you'll get 20% off for your first 6 months with Bench: https://stripe.com/works-with/bench
Best of luck, I hope you find something that works really smoothly for you!
@Pilot: welcome to the market and congratulations on your funding round! Love to see people trying to make life better for startups. I look forward to hearing about your lessons along the way!
I run a SaaS company, and I'm curious why I would need a product like this. Here's how I handle things now:
1. Stripe charges my customers' credit cards every month, and deposits money into my bank account once a week. If I want to check my revenues, I just go to Stripe's website and download the CSVs.
2. All expenses are run through a company credit card. If I want to check my expenses, I just go to my bank's site and download transaction CSVs for the card. If I want to graph them by month or whatever, I can just import that CSV into Excel.
3. For tax purposes, I need to break down #2 into categories of deductible expenses. I just wrote a perl script parse the credit card CSVs and match descriptions. So for example, it'd have a rule like "/digitalocean/ -> Cloud Hosting" and so on.
I admit I'm super-ignorant in this area, but the above solution seems to work without requiring any ongoing effort on my part. How would something like Pilot improve things for me?
Being completely ignorant of your business and noting that all businesses are different, here are some generic considerations around each of the points you raise.
1) Do customers prepay for a month or pay after services have been rendered? Cancellation terms and prorating? There are accounting nuances to each which of course can matter greatly or not very much depending on the particulars of your set up.
2) Running expenses through a company credit card certainly makes things easier but the same considerations as above can also apply...do you prepay for any services? Have long term service contracts even if you pay monthly?
3) Speaking of taxes, do you have a payroll? That's a whole rabbit hole onto itself.
As with most things, context is everything here. Having a proper set of financials that adhere to generally accepted accounting principles (GAAP) are absolutely needed for any fundraising or M&A activities you might do. At minimum, a more precise set of financials can help with understanding the nitty gritty of the business when it comes to forecasting and at-glance health of the business.
at some point of scale, it's not feasible to run all expenses through a credit card. Companies start asking for direct invoicing with bank transfers as payment to avoid the fees on credit card transactions.
I find the same thing with classifying purchases. Most shops can be classified by the shop itself. But with Amazon purchases it can be anything. I have half a mind to write a script that looks up the purchase and finds the category on the Amazon shop page and updates the qif file.
I've found that the key here is minimising the number of categories, and making sure the categories align with how you spend more than what you buy. So my "Groceries" line item includes everything I get from a supermarket, not just literal groceries.
I use https://youneedabudget.com/; it's been fantastic. The marketing focuses on the budget side, but having put together a budget you then obviously need to book-keep to use the budget you've put together :).
This seems to me like an attempt to get into the small end of SME space and also those simpler cases where there is
a) No stock involved
b) A small number of expenses (like items on a credit card)
I guess a saas startup or a freelancer tend to be relatively naive to finance so generally hire an accountant. This is not just to do their books of course, in the UK they are looking for someone to advise on structure, and this often means forming a corporation, which they will do for you. Then there is the issue of VAT (sales tax) which people don't find easy and are scared of. The accountant will do all of this for you for a fee. To save you money they might suggest that you post your expenses to some kind of portal based on xero or quickbooks. Typing in the few expenses of these kind of businesses is absolutely the most trivial of thing to any bookkeeper. They might also use this system to raise invoices from if they are a free lancer.
In a small Saas business the sales are most likely automated, the Stripe charges are a one liner, and the expenses are often all on a credit card (hosting etc). Payroll is a doddle if you are on fixed salaries and a complete PITA if variable hours no matter what the accounting system. If I was to do the monthly bookkeeping for one of these businesses I would spend more time opening the files and administering my billing to them than I would doing the books. My point here is you either a) know what you are doing and could use a spreadsheet or intuit or xero or Sage or b) need an accountant for advice beyond 'what code do I post hosting costs to'. Maybe this fills a particular niche in the US, but I don't see it troubling the accounts world in general.
I work in a different space. We have stock, and an ERP keeping count of sales, cos etc. This is the space your Oracle, Sage, SAP are operating in with big money implementations, and i don't see anything to trouble that here.
Due Disclosure: Worked at Intuit, once upon a time
QuickBooks has network effects (most accountants use it and are familiar with it) and lockin effects (once you have many years of your data in it, there is no way to get it out and into another software. at least that is the way it was when I worked there).
Other products with these characteristics have been disrupted and so can QB. But it will take a lot of work, or a powerful change in perspective. That said, from the article on TC, I don't think Pilot is actually competing with QB.Seems to be more like a layer on top.
I also agree with most of what curun1r says. Intuit is a weird company in many ways, but they do know what they are doing businesswise.
As a former Big 4 accountant in SF, I see Pilot addressing a huge need for startups with a fresh technology layer. I'm sure customers will be happy to get back to building their companies!
Good. We need a substantive competitor for Quickbooks and some real innovation in the space.
You can’t underestimate the complexity and specific requirements of accounting software. Startups in the space begin their journey in a nuclear minefield of customer requirements, and have to navigate not only the professions inane nuances, but also continually evolving bureaucracies and the accounting and tax realities they create. Not. Easy. Stuff.
I wish them luck. Seriously. It’s nice to see entrepreneurs attacking the boring companies. :)
As someone who’s been in this field for quite a while now, my experience in working with most freelance bookkeepers is that...they’re terrible. It’s a trade that has its fair share of thorniness, and most freelancers I’ve worked with simply don’t have a sufficient grasp of it (or are simply unmotivated to care enough to follow through).
With sufficient (very deep) automation it can be done properly, which is what Pilot seems to be aiming for.
Somewhat off-topic, but are there products that do this for home accounting? Like it hooks into my bank account(s), grabs all the transactions, categorizes them, and lets me run reports on them? I have a hodge-podge setup going right now with ledger/plaid.com/custom web UI, but missing things like notifications and budgeting.
I've looked around a bit but not found anything that lets me just set it up once and leave it (and learns from categorization data).
I've been working, unsuccessfully, on accounting software for small businesses for like 8+ years now. I am surprised they got so much funding to come into the space. Good luck to them. It's a great space to build software for because people rely on it and they are willing to pay for it. I wish I had that kind of funding to continue doing it.
Had some customers. But never got that hockey-stick growth all VCs want. And you're always competing with the big boys, Quickbooks, Netsuite, etc... It's a constant arms race to add more features, more integrations, etc... A lot of customer support because your customers rely on you to run their business.
Here's the thing that's a bit different about Pilot: it's not a software product — it's "You have a bookkeeper who does your books (in QuickBooks Online), that you can call and email when you have questions, and that sends you a report every month."
Under the hood, we're building the Iron Man suit for the bookkeepers—letting our team do all the heavy lifting to make sure the work is super-accurate in a way that your typical bookkeeper definitely can't.
Basically, the last thing I wanted as a startup founder was "Yet another software product"—I wanted someone to just take care of the problem for me —which is why we built it this way.
I also find it annoying when people gate software behind sales forms — "Just let me try it!" — but in this case, there really is nothing for you to try, by design. We do all the work for you.
(The next steps today are: we quickly chat with you to better understand your business, and if it's a fit, we pair you with an account manager here, and then send you through a web flow to get read-only access to all the systems we need — your bank, payroll processor, etc.)
Brilliant model. PayPal's fraud detection is usually the most cited human intervention model for building "smart systems", but the idea goes back further: Amazon's backend in the 90s, CD baby's recommendation engine [1] and one can probably find examples that are decades old. I have built large operations teams around this central idea. Happy to chat if you want to compare notes and experience doing this.
1. Our books are stored in QuickBooks Online, and you have access, so you're never locked in. (The "Leave inDinero" migration is a pretty rough one.)
(Strategically this is also important for us because it lets our engineering team focus on building software to assist with the most thorny/painful parts of the process, as opposed to making a QuickBooks clone.)
2. We do accrual by default (inDinero will do it but only at the most premium tiers)
3. Most importantly: our customer service is way better—if you're seriously considering inDinero, I'd encourage you to talk to a few current users and see how they feel about it. (Not fearmongering, but the stuff we've heard has not been great.)
Fair enough. I actually used to use them and the above resonate with me. From my perspective, your solution looks like something that would be worth looking into if I were starting fresh, given that we're a SaaS startup with the added nuances that entails.
I'm having a hard time understanding how this is worth it compared to Xero and their rule-based functions. It's really easy for Xero to learn just the same as a person (if not better).
Pilot isn't "more bookkeeping software", it's "hire this company so that you don't need to worry about getting into the weeds in QuickBooks Online every month, because you have a person who will do all your work in QuickBooks Online for you (assisted by our software)".
It's not that bookkeeping is rocket science or totally unlearnable (you can definitely learn it), the question is really whether it's actually the best use of your time to master it (when you could instead be focusing on running your business).
Thanks for all your replies in this thread, @wdaher!
Can you share how you are approaching growing / scaling the human side of your bookkeeping army? E.g. the dedicated account managers — when I shared your site with a CPA friend, he asked if you have outsourced teams in India doing the categorizing.
Great question, and definitely one of the operational complexities of the business.
Our thesis is: by relying heavily on software from day one, we can actually do it with a small, super-highly-skilled team (rather than a giant army of folks.)
If the task can really be done by an arbitrary person somewhere without context on the business itself, I suspect we can execute that same task ourselves in software (only with much higher consistency and much lower cost). Transaction categorization is a great example of this, actually.
I think the alternative—scaling by just hiring a ton of people—is compelling because it's short-term easy but long-term hard. By analogy to a software system: if you scale with an army of people, you're building a giant distributed system of super-unreliable hardware (the error rate of people is pretty high!), and you can't reprogram it by running "git push".
I'll venture to make a stronger statement: if you care at all about consistency of quality, I think you have to do this mostly in software -- because otherwise the quality of your service is going to be entirely the quality of the individual bookkeeper you use, which is fundamentally going to be pretty variable.
on a somewhat tangential note, how on earth did they secure that domain? Surely buying it would cost a fairly large fraction of that $15M investment if they were to have done it at market rates.
These are good points. I've worked in this space and it's a huge time sink. And you can't afford to get it wrong because you're dealing with people's livelihood.
I'm currently working on a time attendance solution with an emphasis on usability for non technical users. You'd be amazed at how much paper / friction still exists in this domain.
Funny I was working on the same thing (time/attendance solution). Square has timekeeping using mobile app...punch in your ID, clock in. It had more features than I needed...I also have volunteers so they don't get paid but need to keep track of their time. I don't think it was the best solution for me.
I ended up just having a sign in sheet....It works.
My brief advice is: it's easier to work backwards.
Said another way, rather than falling in love with one domain and working hard to get it, try to come up with a shortlist of ~10 domains that are acceptable where they aren't obviously in use, and then work that list until you get something that's acceptable for you.
(Also, depending on your budget, also happy to connect you with the domain broker we used.)
Yeah that's the angle I'm trying to work - I'm bootstrapping so I have to try to be creative with a small domain budget. Given the product name I've gone through a fair number of domain variants. A domain broker might work if I had the budget to drop $10k+ on a domain as one I've been eyeing up is currently owned by the infamously-hard-to-contact twitter. I think I have a viable option on sedo though so I'm going to see if I can snag that :)
First, Quickbooks is frequently not chosen directly. Many, many small businesses choose their bookkeeper first and use whatever they use. One trick Intuit uses both in Quickbooks (bookkeepers) and TurboTax (CPAs) is to put a lot of effort/money into focusing on those relationships knowing that they have a huge indirect impact on acquiring and keeping customers.
Second, the fact that Quickbooks is so awkward and inconsistent to use isn't a UI problem, it's a feature. This is an important lesson in designing software that people use for their jobs. When you make elegant, intuitive software that almost anyone can use in short order, it stops being an impressive item to put on a resume. And it stops being a barrier to entry for competing professionals. Software like Quickbooks that's hard to use correctly becomes a selling point for bookkeepers and almost accomplishes the same purpose (though to a lesser degree) than professional licensing organizations...it limits the competition and keeps the rates they can charge high.
So many Quickbooks competitors have failed to knock them off their perch by not realizing that their customers aren't really their target audience and what should be their target audience doesn't want easy.