14 ms·

If you want to transform IT, start with finance

293 points by feross 5y ago

endymi0n 5y ago
Oh god, no. Finance itself is the worst place to get anything done, since their view is almost always a reactive one. When accounting and controlling are done setting up their cost centers, the most logical place from their view would be to fire all developers and double the sales team, as that's where the money is made. Risky and unproven new projects with unclear (but potentially massive) returns or indirect relations like production engineers increasing the productivity of application engineers through automation do simply not exist in that world. Finance is ultimately a completely soulless place that — just like technology-centered nerds — should defer to a CEO who puts all the pieces together into a coherent vision, ruthlessly chases product-market fit and a profit-first (but NOT profit-only or self-cannibalizing) approach and thus injects business sense into all players in their company. And that includes the finance department, of which I've certainly seen some of the most stupid ideas. Like software capitalization/activation that did made sense for converting opex to capex in the GAAP sheets, but in the end drove all the best engineers away because of all the mindless paperwork. Just one of many examples. Finance isn't any more intelligent than the other players in the company.
- 908B64B197 5y ago
  > When accounting and controlling are done setting up their cost centers, the most logical place from their view would be to fire all developers and double the sales team, as that's where the money is made. That's great for long term shorts.
  - endymi0n 5y ago
    Well it _is_ a common trick for "making the bride pretty" or for short-term incentivized CEOs: You know you've got a top notch DevSecOps team for example if everything is smooth, calm and boring — because most issues are automated away. That's exactly the point when you know you can safely fire that expensive team, because it'll usually take a few years until the automation surplus has been exhausted and then a few more until the starting chaos has propagated into actual outside-visible incidents. Makes for a great bottom-line at approximately the end of shares vesting. You didn't get that pro-tip from me.
- hdvpp 5y ago
  Manage cash flow in a unpredictable ever changing environment for a month and come back and tell us what you think a finance dept does cause from the comment its quite clear you really dont know.
  - lifeisstillgood 5y ago
    Perhaps here would be a great place to let a few hundred developers know. It's worth differentiating between allocation and flow - a million dollar spend can be hard to allocate to Project X, Y or Z. But it is usually simple to see it got spent on Developers, Sales people, rent, hardware licenses etc. As the film says, if you go up high enough, there is only one man with one budget.
    - andrewlgood 5y ago
      And that man/woman is the CEO. Cash flow is the actual movement of money - paying payroll, paying vendors, receiving cash from clients (not just sending out the invoice). Lack of cash flow is what makes companies go bankrupt. Allocations are methods to attribute income and/or expenses to groups. It may be something simply such as dividing up an enterprise software license for email by the number of employees and allocating that expense to their department. It can be more difficult / squishy such as how to attribute brand marketing spend to various product lines. Generally, companies use allocations because they are trying to understand the profitability of something (a product, geography, client, etc). Your example of the million dollar spend is a good one. Very easy to see where the cash flow went (Developers, Sales people, rent, etc) but the important question to assist on future investment decisions is how much should be allocated to Project X vs Y vs Z. One would also need to understand how much cash was generated by Project X, Y, and Z so they can determine the ROI (Return On Investment) for each project. Knowing the ROI is important in making future investment decisions.
- rootsudo 5y ago
  I disagree, if you're able to understand and do a projection, forecast, risk analysis, sunk cost analysis you can easily get finance on your side. The issue is that IT/tech does not know business key words or excel spreadsheets, and finance does not know tech. Nowadays it's getting blurred with data scientsts and excel crunchers that can do a mean macro (and know more than Vlookup!) Finance people love automation. Finance people love excel. Finance people understand sunk cost fallacy. Finance people understand change. Finance people need help to embrace change, to guide decisions and make sure it's the right "tech" for the company. Many times the right tech is O365/Sharepoint/Flow automation, or nintex. Yeah, but it is not souless, it guides the entire company, it ist he company pursestrings. Get them on your side, understand how to read a P/L sheet, what FPA does and you'll not only earn respect, you'll have your name said at all the exec/leaderships meetings and you'll be loved. That's what I do in companies (consulting, self run) and w/o a good background in finance, and tech I wouldn't be where I am today (major in stats, tech was hobby, stats/math is everything etc.) So I fully disagree 100%. You had it so close with SAAS and capex vs opex, but you missed it. But yes it does move expenses around and opex is easier to navigage then a full sunk cost capex.
  - erdos4d 5y ago
    I read this comment and I feel dumber.
  - clncy 5y ago
    Spot in with the automation! Too often people rush to build a new system without thinking first about the systems and processes that already exist.
  - planet-and-halo 5y ago
    This is something I've been interested in for a while. Do you think it's worth getting an MBA or some other financial degree if you're already in the workforce, or can you learn enough on the fly?
    - andrewlgood 5y ago
      I do not recommend getting an MBA or some other financial degree. Rootsudo's response is excellent. There are actually very few financial concepts involved in these types of analyses (e.g. discounted cash flows, IRR, and NPV). IMHO, the best thing you can do is work with someone in your FP&A (Financial Planning & Analysis) group discuss your investment idea and why you think it is a good idea. If they are a decent analyst, they will help build a model to describe financially your investment opportunity. This is what they live for.
- avmich 5y ago
  > And that includes the finance department, of which I've certainly seen some of the most stupid ideas. I've heard some tales about a town in Michigan - let's call it Flint - where water was distributed using some older provider, while some newer company existed which charged less. A newly hired CFO decided, logically, to save some funds by going to the cheaper one. CFO didn't know that pipes in the town were made in the old fashion (lead metal?) and the new company, which used iron pipes, used some chemicals in water which helped to keep pipes clean. This chemical wasn't used by the old water supplier. When flown through old pipes, the chemical caused dissolving the lead into water. By the time the population of Flint switched to bottled water for everything the CFO was long gone with earned bonus for saving money to town. I wonder what such stories, if true, teach us for the future. Just general mistrust towards financiers doesn't look quite right...
  - clairity 5y ago
    > “Just general mistrust towards financiers doesn't look quite right...” financiers, including startup investors, focus on money above all else, and that alone justifies mistrust (aka skepticism). with that kind of focus, it’s inevitable that decisions are made selfishly and greedily. that doesn’t mean it’s a tool that can’t be wielded for generally positive outcomes, but that the deck is (intentionally) stacked against it. beyond that, the flint situation sounds like a real failure of leadership all around (at the very least, basic due diligence/risk analysis). that often happens when decisions are made far removed from the expertise needed to make them.
  - thaumasiotes 5y ago
    > I wonder what such stories, if true, teach us for the future. That if you're looking to maintain [water] quality, you need to have monitoring in place. Switching to the new provider was a good idea, not a bad idea. There is no way to keep track of everything that might have happened in the past that might be affected by unspecified changes in the future, so there can't be an expectation of doing it either. But you can watch what kind of results you're getting.
    - whatshisface 5y ago
- blihp 5y ago
  Computers first came into many companies via the Finance department way back when because they recognized what computers could do as glorified calculators, spreadsheets and accounting systems.[1] IT departments were often an outgrowth of those early systems. The only 'trick' to dealing with finance people is being able to draw a direct line between a proposal and either increased revenue or decreased costs. For solutions that can do that, they are often your best champion short of having the CEO already on board. For solutions that require a leap of faith/vision/creativity, they're probably not the place to start. [1] Who do you think the numeric keypad on 'full size' keyboards popularized by the IBM PC were for? Accountants. Granted, word processing was the other early use case... that Finance signed the checks for. IBM understood their customer.
  - cratermoon 5y ago
    > Computers first came into many companies via the Finance department way back when because they recognized what computers could do as glorified calculators, spreadsheets and accounting systems.[1] IT departments were often an outgrowth of those early systems And yet, somehow, IT departments ended up getting classified as cost centers by the finance department.
    - blihp 5y ago
      If they don't directly contribute to the bottom line, that's how they should be classified from an accounting perspective. There are at least a couple of ways IT departments could avoid that fate: having internal customers pay for their services independent of the cost or directly managing projects that otherwise contribute to the bottom line. It's rare, but it does happen.
- andrewlgood 5y ago
  Full disclosure, I am a CFO and have spent much of my career in Finance. I have never worked in a place where I or anyone else there thought Finance is "any more intelligent than the other players in the company." This article is disappointing in that it gives the impression that one department makes or breaks a company. To be clear, there are poorly run Finance organizations in the world and they can make the culture at a company less than fun. There are also poorly run IT organizations (to include product development) that can ruin a company culture. Even ones run by a CTO that reports to the CEO. I have also never seen a company where the CFO does not ultimately report to the CEO. I have seen companies where the CFO reports to a Chief Administrative Officer (CAO) who reports to the CEO. I completely agree that the CEO ultimately sets the vision of the company and hires the executive team to deliver on it. That being said, a CEO is not Superman/Superwoman. They need a team surrounding them to deliver what you expect. What the article could have emphasized is how software engineers can work with the Finance team to better describe investment opportunities like hiring production engineers to increase the productivity of application engineers. Yes, investment opportunities ultimately need to generate more cash returns than the investment. This may require serious thought around the benefits from increased software quality, faster cycle times, increased cybersecurity etc. Importantly the analyses do not have to be decimal level precise. But, if you cannot describe how your investment opportunity ultimately creates more cash flow than the investment required, your opportunity will likely not be funded.
  - Fordec 5y ago
    I am at a point where I don't think there should be finance departments in tech companies. I also believe there shouldn't be siloed data departments for the record either. There should be teams, build around a cohesive product goal that solves customer needs. There should be someone who understands the money side of things embedded in that department doing analyses and making recommendations to the product manager, not direct to a CFO. Maybe half their job is regular market analysis work too if the role isn't that big. Those product managers should then have in their day to day functions a requirement to get that info to a CFO's hands (who is really just the executive teams internal finance person) to make their case for their product.
- syshum 5y ago
  CFO's makes the worst CEO's for a reason.... This has been true for all of time, if you work for a company that has a CEO that spent their entire career in Finance... RUN
- wtvanhest 5y ago
  I have worked in FP&A for the past 6 or so years after working in the investment world. For a short time (8 or so months) I supported a CIO and had to approve the NPV analysis of every tech project that the company did. The thing that most project managers / engineers don't see is that finance's real job is to simplify the story and fact check the assumptions so that execs can rely on the investment being reasonable. Execs don't care if the NPV is to the penny, they care that it makes logical sense and that business leaders (brand leaders, region leaders or whoever is SVPsh level) supports the project. More than 50% of my time was just talking to execs trying to lock down support for other people's projects so they could get them done. Often times when a project get rejected, its not because finance made the decision, its because tech didn't get the right business leaders on board before asking for a project investment. > Risky and unproven new projects with unclear (but potentially massive) returns or indirect relations like production engineers increasing the productivity of application engineers through automation do simply not exist in that world. If you do a lot of risky and unproven new projects with unclear (but potentially massive) returns, the business will go bankrupt and everyone will lose their jobs and equity. Incremental, but repeated small and correct bets usually are best for driving high returns on capital overtime. Sometimes when an intuitive exec sees an opportunity they need to take a shot on goal and risk some capital. In those cases, finance is kind of out of the picture as the exec needs to stand on their own. > And that includes the finance department, of which I've certainly seen some of the most stupid ideas. Like software capitalization/activation that did made sense for converting opex to capex in the GAAP sheets, but in the end drove all the best engineers away because of all the mindless paperwork. Just one of many examples. The opex to capex justification stuff comes from accounting, not finance. Its very painful, but the accounting teams must follow the law/gaap rules of accounting. > Finance isn't any more intelligent than the other players in the company. Absolutely correct. Just another cog in the corporate machine. If you find yourself frustrated with any cog (co-workers) its probably best to talk to them and figure out what you are missing. Most people are trying their hardest.
- xiezhifeng 5y ago
  Agree, so my best choice is start from value stream. Culture is not specific.
- draw_down 5y ago
  Grow up
clipradiowallet 5y ago
Finance is an excellent place to start. It's very common for IT directors to report to the CFO - not the CIO or CTO, but the chief financial officer. CFO's are (ideally) great at saving money, and it's likely they will begin to do so by pinching pennies within IT. Think about that - someone who has an entire career around a specific discipline(finance) directly governing another discipline(IT). They cannot hope to make informed decisions based on the current state of the industry unless they are neglecting their primary responsibility of finance. tldr; yes, "starting with finance" is a good way to transform IT.
- jjeaff 5y ago
  That doesn't make any sense at all. What is the thought process behind putting IT under finance? Seems like if you don't have a cto or cio then reporting to a coo or even ceo makes more sense than cfo.
  - austincheney 5y ago
    IT is a cost center and very expensive.
    - xxpor 5y ago
      This is fundamentally what Europe especially and some non-tech large US firms don't get. Repeat after me: software development *is not* IT. IT is the department that hands out laptops and buys office licenses from MS. Software Engineering is a product development group. They're two entirely different functions. IT can make sense to have under the CFO, but product development almost never does.
      - walshemj 5y ago
        Not everyone makes that hard distinction
        xxpor 5y ago
kingrazor 5y ago
The stages mentioned in this article are exactly what the company I last worked for went through. Inevitably they lost customers to other products with faster release cycles. The company is still limping along but at a fraction of their former revenue. It's sad, really.
yardie 5y ago
From an operational view finance and IT are more aligned than they are not. Finance is all about cash flow; where it's coming from and where it's going. And IT is all about data flow; where is good data coming from, what is it doing, where is it going. In my experience in IT most of the infosec applies broadly to the organization it's especially the case when working with sensitive financial information. Also, because of recent ransomware attacks I've been given (almost) carte blanche to secure our network and operational infrastructure with a special focus on finance, because the spice must flow.
- tmitchel2 5y ago
  Last part made me chuckle
landryraccoon 5y ago
I’m seriously stumbling over the description of phases I and II. Wait, this is a startup that doesn’t have a strong product vision from day 1 and isn’t aggressively iterating to find product market fit? Here’s a tell tale sign that you have product market fit - your product is buggy and sucks and your customers constantly complain and demand more improvement but they can’t give you up. They need to pay you because you serve a need they can’t or wont find somewhere else. The customer needs a bespoke feature that really only they will use and isn’t general to anyone else? Is the founder team disciplined enough to say no and focus on what really adds value? Sure, they would love for you to be a cheap outsourced dev shop, but that’s why your CEO needs to push back and sell the vision of the company. Phases 1, 2 and 3 are backwards. Why is the startup trying to be a consulting company? This doesn’t sound like the story of any visionary startup I know of. It sounds like a group of developers who can make a profitable software consulting company trying to pivot to making a product. IMHO a startup can be really dysfunctional in its dev process and management and still succeed, if they find product market fit first.
- hyperpape 5y ago
  He says this is common to pre-SaaS companies. So he's mostly describing companies started at least a decade ago, perhaps two or three. However, the problems he describes can easily affect enterprise SaaS companies in a niche. If your market literally has a few thousand potential customers, you could end up in a situation where onboarding a big new customer is a double-digit percentage of your current revenue. At that point, it doesn't matter if you're a SaaS or not, you're gonna feel the pressure to do things for the sake of them.
- throwaaskjdfh 5y ago
  > Why is the startup trying to be a consulting company? This doesn’t sound like the story of any visionary startup I know of. Why does a successful software business need to be a "visionary startup"? Taking on consulting work and discovering the product by listening to customers seems like a reasonable approach.
  - landryraccoon 5y ago
nhoughto 5y ago
Seems very familiar. Its not that finance get things done, its that they can prevent things being done entirely, or reduce funding/investment to the point they naturally peter out.
slt2021 5y ago
IT and Finance report to CEO. If you have clueless boomer CEO who doesnt understand what is digital transformation, then nothing will help
- acdha 5y ago
  If the CEO doesn't listen to anyone who works for them, you're going to have many problems. If not, you might want to work on your communications strategy: in particular, ask whether the problem is that you failed to communicate the business benefits from a proposed modernization plan or are just repeating consultant babble.
  - slt2021 5y ago
    IT and Finance are just executing CEO's strategy and vision. A lot depends on hiring - whether CEO/Board is capable of hiring competent CFO and CIO. WIth CFO it is easier, because finance has been around since money was invented and a lot of hard problems have been solved in finance and there is vast theoretical body of knowledge backing it up. a lot of it is straightforward. with IT it is hard, because the field is changing rapidly and hiring a competent CIO/CTO is harder. this is why CEO is critical, and he is ultimately responsible for the overall outcome
underwater 5y ago
I worked for a company where capex was cheap and easy and opex was hard to get. So engineering managers would pitch a project, fund a bunch of consultants, get a massive piece of tech built and then fail to support it. They knew the downsides, and wanted to build long lived teams that built and support and evolve products, but there was too much push back from finance and IT. I realised that the company would prefer to spend two or three dollars in capex every year than a dollar of opex. This helped me understand how and entire industry of overpaid an underqualified consultants existed. So my question is, where is that accountability for the finance function? Their controls and constraints caused the company to flush huge amounts of money down the drain, just so their books could have numbers appear in the columns they wanted.
- catmanjan 5y ago
  Was capex preferred due to preferential tax treatment?
  - andrewlgood 5y ago
    No. You actually get better tax treatment for opex as you can record the full amount of the expense immediately as a deduction against your income. Capex spreads the deductibility of the expense over years. As I mention above, opex vs capex is usually a compensation issue.
- andrewlgood 5y ago
  This is not a Finance function issue. Finance has absolutely no preference between opex and capex. There are accounting rules that determine the appropriate classification and any reasonable CFO simply wants it done correctly. The opex / capex argument is about compensation. Opex reduces income immediately while capex gets spread out over the projected life of the asset (software, power plant, computer, etc). By shifting items from opex to capex, a company temporarily increases its net income. Increasing net income often leads to higher bonuses. This is what drives the silly behavior. To make matters worse, once a company goes down this path, it cannot get off as it is usually evaluated on a year-over-year basis. As for accountability, CFOs of public companies can be found criminally guilty (i.e. go to jail) if they not only commit accounting fraud, but simply do not establish a robust enough control framework to ensure the financial statements are correct. CFOs are always accountable ultimately to the CEO and sometimes to the Board of Directors. If you think the CEO was unaware the company was spending "two or three dollars in capex every year than a dollar of opex," your either mistaken or you had a clueless CEO.
rdxm 5y ago
Tried to plant this seed ~3 years ago at a large defense contractor. They were frustrated by the fact they weren't getting the velocity out of gov cloud they expected. my response was: "you literally applied the same procurement / process model to GC that you had for on-prem, wtf did you expect?!?" I called my pitch: "AdminOps", basically the message was the finance/biz side of the house needed the same type of transformative event engineering got with devops. They are still slow....go figure. Enterprises gonna Enterprise....
cratermoon 5y ago
Wasn't if finance that decided that IT should be cost center for every company?
krishvs 5y ago
Wow - the 4 stages of product development seems quite spot on from my experience working on b2b enterprise software products. No doubt that the Finance team is generally the most powerful entity in shaping a companies direction both internally as well as in your clients organisation. Finance teams love control, are (mostly) completely risk-averse and change averse. The same can also be said about badly run IT teams. I have seen product/automation implementations being deployed seamlessly in all departments from operations to procurement but completely fail due to the finance teams. In extreme cases I have seen finance teams have so much power that they even override the CEOs decisions to make implementations fail if they feel that they are loosing control or do not like the new process workflow that the implementation brings in. For digital transformation projects the implementation team needs to speak the language of finance, operations as well as IT. Otherwise the chances of the implementation really succeeding is close to nil.
throwaway984393 5y ago
Conway's Law Revisited: If finance dictates your organizational structure, then finance is developing your software.
lincpa 5y ago
Development ideas from A Certified Public Accountant: Computer science is essentially a management science, and vice versa. Follower's success stories: Apple M1 chip + MacOS come soon: Qualcomm, MS, Intel, AMD, Samsung, Cray, etc. The Grand Unified Programming Theory: The Pure Function Pipeline Data Flow with Principle-based Warehouse/Workshop Model https://github.com/linpengcheng/PurefunctionPipelineDataflow https://github.com/linpengcheng/PurefunctionPipelineDataflow
goforbg 5y ago
I think this is mostly true for B2B SAAS startups but not for the rest.
goforbg 5y ago
I think this holds true mostly for B2B SAAS platforms but not for the rest.
shkurski_ 5y ago
Looking at Marx's quote on relations of production, could the money flow INSIDE a company be another topic that is worth exploration? From my simple understanding, in Marx's theory it was social aspect that mattered most. Do we still face such problems as it was in 19th century? I'm curious whether it would be a viable company where the surplus-value is distributed among workers and I have a fixed transparent salary as CEO and how it will survive the stage I and scale to stage II, IIIa, IIIb.
Vaslo 5y ago
Funny to read this because I happen to work in Finance and many requests come through my team. I’m going to add some advice here for what it’s worth. A manager once told me that Legal and IT will always tell you no before telling you yes. Basically you need to argue and escalate to get things done, and it’s frustrating. After 20 years I’ve found that pretty true, though my current company is a bit better than others I’ve worked for where IT pretty much said no to everything. Everytime you are going to stop me from writing queries to your server, or add more security that makes it harder to do my job without sufficient support to rightfully get through that security, or deny me the ability to download legit software from the likes of Microsoft and other reputable vendors because you have some ridiculous amount of red tape, I’m absolutely going to question every dollar and be skeptical of how the funding is going to help me or my partners in HR, Sales, Supply Chain etc. If your projects help me, and you help me, I’m definitely willing to help you more and fight for you and your budget.
- NullPrefix 5y ago
  Did you just admit that you fight legal and IT on purpose, just to have a fight?
  - ativzzz 5y ago
    He chooses to not help people who make his job harder, sounds like a normal human being to me.
    - 542458 5y ago
      Lots of people make other people’s jobs “harder” with good reason. Things like code review, legal compliance, IT policy, etc are just “overhead” if you assume that your knowledge is infallible - but people are fallible and we need systems like this that, yes, slow us down in order to keep from getting burned.
      - Vaslo 5y ago
        The overhead needs support. If you put up additional security where only IT staff can do x, then you need to ask for or transfer headcount to ensure it’s just as nimble as before. If you demonstrate that ability to manage change, then you also demonstrate that the limited funds are best in your hands.
oneplane 5y ago
It's obviously going to strongly depend on the context, but I've found that there is a shared amount of risk in organisations where IT and Finance both have their choices to make together. Say you have some on-prem datacenter stuff for a manufacturing plant, you might have the request from factory operations to be as available and consistent as possible. The cost for such a thing might be astronomical, but even a single day of downtime would cover it. On the other hand, the finance department might already have made a risk assessment and set aside some money or had it insured in some way and now suddenly the impact of that risky thing that might happen is a whole lot lower in terms of money. For a lot of people that can be news, or unexpected. If you silo yourself you might end up engineering something that is completely unnecessary. At the same time, someone might be buying some expensive insurance, but forgets that unrecoverable data doesn't magically yeet itself into existence just because the insurance contract says so. So while on paper it's "covered and we just need to recover our systems and get on with it", in reality that might not be possible. Starting a company from scratch (IT-wise) leaves you with empty systems. Those are about as useful as a brick. A functional brick, but perhaps not the spreadsheet, email or CRM database you had in mind. This is a lot of words just to say: in most companies (as long as they don't get too big), most people are there for a reason, and most departments are there for a reason. Go up the chain of reasoning and you'll find your common reason ("because the company wants to make money" for example), and now you have all you need to 'justify' to yourself that 'those other people' are there for the same reason and with the same goal as you.
steve76 5y ago
More like end with finance. It's inevitable if you are successful. Why build it if you can buy it? Save. Just don't loose money. Finance is not what most people think. Frugal. Charitable.

293 points by feross 5y ago

Oh god, no. Finance itself is the worst place to get anything done, since their view is almost always a reactive one. When accounting and controlling are done setting up their cost centers, the most logical place from their view would be to fire all developers and double the sales team, as that's where the money is made. Risky and unproven new projects with unclear (but potentially massive) returns or indirect relations like production engineers increasing the productivity of application engineers through automation do simply not exist in that world. Finance is ultimately a completely soulless place that — just like technology-centered nerds — should defer to a CEO who puts all the pieces together into a coherent vision, ruthlessly chases product-market fit and a profit-first (but NOT profit-only or self-cannibalizing) approach and thus injects business sense into all players in their company. And that includes the finance department, of which I've certainly seen some of the most stupid ideas. Like software capitalization/activation that did made sense for converting opex to capex in the GAAP sheets, but in the end drove all the best engineers away because of all the mindless paperwork. Just one of many examples. Finance isn't any more intelligent than the other players in the company.

908B64B197 5y ago
> When accounting and controlling are done setting up their cost centers, the most logical place from their view would be to fire all developers and double the sales team, as that's where the money is made. That's great for long term shorts.
- endymi0n 5y ago
  Well it _is_ a common trick for "making the bride pretty" or for short-term incentivized CEOs: You know you've got a top notch DevSecOps team for example if everything is smooth, calm and boring — because most issues are automated away. That's exactly the point when you know you can safely fire that expensive team, because it'll usually take a few years until the automation surplus has been exhausted and then a few more until the starting chaos has propagated into actual outside-visible incidents. Makes for a great bottom-line at approximately the end of shares vesting. You didn't get that pro-tip from me.
hdvpp 5y ago
Manage cash flow in a unpredictable ever changing environment for a month and come back and tell us what you think a finance dept does cause from the comment its quite clear you really dont know.
- lifeisstillgood 5y ago
  Perhaps here would be a great place to let a few hundred developers know. It's worth differentiating between allocation and flow - a million dollar spend can be hard to allocate to Project X, Y or Z. But it is usually simple to see it got spent on Developers, Sales people, rent, hardware licenses etc. As the film says, if you go up high enough, there is only one man with one budget.
  - andrewlgood 5y ago
    And that man/woman is the CEO. Cash flow is the actual movement of money - paying payroll, paying vendors, receiving cash from clients (not just sending out the invoice). Lack of cash flow is what makes companies go bankrupt. Allocations are methods to attribute income and/or expenses to groups. It may be something simply such as dividing up an enterprise software license for email by the number of employees and allocating that expense to their department. It can be more difficult / squishy such as how to attribute brand marketing spend to various product lines. Generally, companies use allocations because they are trying to understand the profitability of something (a product, geography, client, etc). Your example of the million dollar spend is a good one. Very easy to see where the cash flow went (Developers, Sales people, rent, etc) but the important question to assist on future investment decisions is how much should be allocated to Project X vs Y vs Z. One would also need to understand how much cash was generated by Project X, Y, and Z so they can determine the ROI (Return On Investment) for each project. Knowing the ROI is important in making future investment decisions.
rootsudo 5y ago
I disagree, if you're able to understand and do a projection, forecast, risk analysis, sunk cost analysis you can easily get finance on your side. The issue is that IT/tech does not know business key words or excel spreadsheets, and finance does not know tech. Nowadays it's getting blurred with data scientsts and excel crunchers that can do a mean macro (and know more than Vlookup!) Finance people love automation. Finance people love excel. Finance people understand sunk cost fallacy. Finance people understand change. Finance people need help to embrace change, to guide decisions and make sure it's the right "tech" for the company. Many times the right tech is O365/Sharepoint/Flow automation, or nintex. Yeah, but it is not souless, it guides the entire company, it ist he company pursestrings. Get them on your side, understand how to read a P/L sheet, what FPA does and you'll not only earn respect, you'll have your name said at all the exec/leaderships meetings and you'll be loved. That's what I do in companies (consulting, self run) and w/o a good background in finance, and tech I wouldn't be where I am today (major in stats, tech was hobby, stats/math is everything etc.) So I fully disagree 100%. You had it so close with SAAS and capex vs opex, but you missed it. But yes it does move expenses around and opex is easier to navigage then a full sunk cost capex.
- erdos4d 5y ago
  I read this comment and I feel dumber.
- clncy 5y ago
  Spot in with the automation! Too often people rush to build a new system without thinking first about the systems and processes that already exist.
- planet-and-halo 5y ago
  This is something I've been interested in for a while. Do you think it's worth getting an MBA or some other financial degree if you're already in the workforce, or can you learn enough on the fly?
  - andrewlgood 5y ago
    I do not recommend getting an MBA or some other financial degree. Rootsudo's response is excellent. There are actually very few financial concepts involved in these types of analyses (e.g. discounted cash flows, IRR, and NPV). IMHO, the best thing you can do is work with someone in your FP&A (Financial Planning & Analysis) group discuss your investment idea and why you think it is a good idea. If they are a decent analyst, they will help build a model to describe financially your investment opportunity. This is what they live for.
avmich 5y ago
> And that includes the finance department, of which I've certainly seen some of the most stupid ideas. I've heard some tales about a town in Michigan - let's call it Flint - where water was distributed using some older provider, while some newer company existed which charged less. A newly hired CFO decided, logically, to save some funds by going to the cheaper one. CFO didn't know that pipes in the town were made in the old fashion (lead metal?) and the new company, which used iron pipes, used some chemicals in water which helped to keep pipes clean. This chemical wasn't used by the old water supplier. When flown through old pipes, the chemical caused dissolving the lead into water. By the time the population of Flint switched to bottled water for everything the CFO was long gone with earned bonus for saving money to town. I wonder what such stories, if true, teach us for the future. Just general mistrust towards financiers doesn't look quite right...
- clairity 5y ago
  > “Just general mistrust towards financiers doesn't look quite right...” financiers, including startup investors, focus on money above all else, and that alone justifies mistrust (aka skepticism). with that kind of focus, it’s inevitable that decisions are made selfishly and greedily. that doesn’t mean it’s a tool that can’t be wielded for generally positive outcomes, but that the deck is (intentionally) stacked against it. beyond that, the flint situation sounds like a real failure of leadership all around (at the very least, basic due diligence/risk analysis). that often happens when decisions are made far removed from the expertise needed to make them.
- thaumasiotes 5y ago
  > I wonder what such stories, if true, teach us for the future. That if you're looking to maintain [water] quality, you need to have monitoring in place. Switching to the new provider was a good idea, not a bad idea. There is no way to keep track of everything that might have happened in the past that might be affected by unspecified changes in the future, so there can't be an expectation of doing it either. But you can watch what kind of results you're getting.
  - whatshisface 5y ago
blihp 5y ago
Computers first came into many companies via the Finance department way back when because they recognized what computers could do as glorified calculators, spreadsheets and accounting systems.[1] IT departments were often an outgrowth of those early systems. The only 'trick' to dealing with finance people is being able to draw a direct line between a proposal and either increased revenue or decreased costs. For solutions that can do that, they are often your best champion short of having the CEO already on board. For solutions that require a leap of faith/vision/creativity, they're probably not the place to start. [1] Who do you think the numeric keypad on 'full size' keyboards popularized by the IBM PC were for? Accountants. Granted, word processing was the other early use case... that Finance signed the checks for. IBM understood their customer.
- cratermoon 5y ago
  > Computers first came into many companies via the Finance department way back when because they recognized what computers could do as glorified calculators, spreadsheets and accounting systems.[1] IT departments were often an outgrowth of those early systems And yet, somehow, IT departments ended up getting classified as cost centers by the finance department.
  - blihp 5y ago
    If they don't directly contribute to the bottom line, that's how they should be classified from an accounting perspective. There are at least a couple of ways IT departments could avoid that fate: having internal customers pay for their services independent of the cost or directly managing projects that otherwise contribute to the bottom line. It's rare, but it does happen.
andrewlgood 5y ago
Full disclosure, I am a CFO and have spent much of my career in Finance. I have never worked in a place where I or anyone else there thought Finance is "any more intelligent than the other players in the company." This article is disappointing in that it gives the impression that one department makes or breaks a company. To be clear, there are poorly run Finance organizations in the world and they can make the culture at a company less than fun. There are also poorly run IT organizations (to include product development) that can ruin a company culture. Even ones run by a CTO that reports to the CEO. I have also never seen a company where the CFO does not ultimately report to the CEO. I have seen companies where the CFO reports to a Chief Administrative Officer (CAO) who reports to the CEO. I completely agree that the CEO ultimately sets the vision of the company and hires the executive team to deliver on it. That being said, a CEO is not Superman/Superwoman. They need a team surrounding them to deliver what you expect. What the article could have emphasized is how software engineers can work with the Finance team to better describe investment opportunities like hiring production engineers to increase the productivity of application engineers. Yes, investment opportunities ultimately need to generate more cash returns than the investment. This may require serious thought around the benefits from increased software quality, faster cycle times, increased cybersecurity etc. Importantly the analyses do not have to be decimal level precise. But, if you cannot describe how your investment opportunity ultimately creates more cash flow than the investment required, your opportunity will likely not be funded.
- Fordec 5y ago
  I am at a point where I don't think there should be finance departments in tech companies. I also believe there shouldn't be siloed data departments for the record either. There should be teams, build around a cohesive product goal that solves customer needs. There should be someone who understands the money side of things embedded in that department doing analyses and making recommendations to the product manager, not direct to a CFO. Maybe half their job is regular market analysis work too if the role isn't that big. Those product managers should then have in their day to day functions a requirement to get that info to a CFO's hands (who is really just the executive teams internal finance person) to make their case for their product.
syshum 5y ago
CFO's makes the worst CEO's for a reason.... This has been true for all of time, if you work for a company that has a CEO that spent their entire career in Finance... RUN
wtvanhest 5y ago
I have worked in FP&A for the past 6 or so years after working in the investment world. For a short time (8 or so months) I supported a CIO and had to approve the NPV analysis of every tech project that the company did. The thing that most project managers / engineers don't see is that finance's real job is to simplify the story and fact check the assumptions so that execs can rely on the investment being reasonable. Execs don't care if the NPV is to the penny, they care that it makes logical sense and that business leaders (brand leaders, region leaders or whoever is SVPsh level) supports the project. More than 50% of my time was just talking to execs trying to lock down support for other people's projects so they could get them done. Often times when a project get rejected, its not because finance made the decision, its because tech didn't get the right business leaders on board before asking for a project investment. > Risky and unproven new projects with unclear (but potentially massive) returns or indirect relations like production engineers increasing the productivity of application engineers through automation do simply not exist in that world. If you do a lot of risky and unproven new projects with unclear (but potentially massive) returns, the business will go bankrupt and everyone will lose their jobs and equity. Incremental, but repeated small and correct bets usually are best for driving high returns on capital overtime. Sometimes when an intuitive exec sees an opportunity they need to take a shot on goal and risk some capital. In those cases, finance is kind of out of the picture as the exec needs to stand on their own. > And that includes the finance department, of which I've certainly seen some of the most stupid ideas. Like software capitalization/activation that did made sense for converting opex to capex in the GAAP sheets, but in the end drove all the best engineers away because of all the mindless paperwork. Just one of many examples. The opex to capex justification stuff comes from accounting, not finance. Its very painful, but the accounting teams must follow the law/gaap rules of accounting. > Finance isn't any more intelligent than the other players in the company. Absolutely correct. Just another cog in the corporate machine. If you find yourself frustrated with any cog (co-workers) its probably best to talk to them and figure out what you are missing. Most people are trying their hardest.
xiezhifeng 5y ago
Agree, so my best choice is start from value stream. Culture is not specific.
draw_down 5y ago
Grow up

clipradiowallet 5y ago

Finance is an excellent place to start. It's very common for IT directors to report to the CFO - not the CIO or CTO, but the chief financial officer. CFO's are (ideally) great at saving money, and it's likely they will begin to do so by pinching pennies within IT. Think about that - someone who has an entire career around a specific discipline(finance) directly governing another discipline(IT). They cannot hope to make informed decisions based on the current state of the industry unless they are neglecting their primary responsibility of finance. tldr; yes, "starting with finance" is a good way to transform IT.

jjeaff 5y ago
That doesn't make any sense at all. What is the thought process behind putting IT under finance? Seems like if you don't have a cto or cio then reporting to a coo or even ceo makes more sense than cfo.
- austincheney 5y ago
  IT is a cost center and very expensive.
  - xxpor 5y ago
    This is fundamentally what Europe especially and some non-tech large US firms don't get. Repeat after me: software development *is not* IT. IT is the department that hands out laptops and buys office licenses from MS. Software Engineering is a product development group. They're two entirely different functions. IT can make sense to have under the CFO, but product development almost never does.
    - walshemj 5y ago
      Not everyone makes that hard distinction
      - xxpor 5y ago

kingrazor 5y ago

The stages mentioned in this article are exactly what the company I last worked for went through. Inevitably they lost customers to other products with faster release cycles. The company is still limping along but at a fraction of their former revenue. It's sad, really.

yardie 5y ago

From an operational view finance and IT are more aligned than they are not. Finance is all about cash flow; where it's coming from and where it's going. And IT is all about data flow; where is good data coming from, what is it doing, where is it going. In my experience in IT most of the infosec applies broadly to the organization it's especially the case when working with sensitive financial information. Also, because of recent ransomware attacks I've been given (almost) carte blanche to secure our network and operational infrastructure with a special focus on finance, because the spice must flow.

tmitchel2 5y ago
Last part made me chuckle

landryraccoon 5y ago

I’m seriously stumbling over the description of phases I and II. Wait, this is a startup that doesn’t have a strong product vision from day 1 and isn’t aggressively iterating to find product market fit? Here’s a tell tale sign that you have product market fit - your product is buggy and sucks and your customers constantly complain and demand more improvement but they can’t give you up. They need to pay you because you serve a need they can’t or wont find somewhere else. The customer needs a bespoke feature that really only they will use and isn’t general to anyone else? Is the founder team disciplined enough to say no and focus on what really adds value? Sure, they would love for you to be a cheap outsourced dev shop, but that’s why your CEO needs to push back and sell the vision of the company. Phases 1, 2 and 3 are backwards. Why is the startup trying to be a consulting company? This doesn’t sound like the story of any visionary startup I know of. It sounds like a group of developers who can make a profitable software consulting company trying to pivot to making a product. IMHO a startup can be really dysfunctional in its dev process and management and still succeed, if they find product market fit first.

hyperpape 5y ago
He says this is common to pre-SaaS companies. So he's mostly describing companies started at least a decade ago, perhaps two or three. However, the problems he describes can easily affect enterprise SaaS companies in a niche. If your market literally has a few thousand potential customers, you could end up in a situation where onboarding a big new customer is a double-digit percentage of your current revenue. At that point, it doesn't matter if you're a SaaS or not, you're gonna feel the pressure to do things for the sake of them.
throwaaskjdfh 5y ago
> Why is the startup trying to be a consulting company? This doesn’t sound like the story of any visionary startup I know of. Why does a successful software business need to be a "visionary startup"? Taking on consulting work and discovering the product by listening to customers seems like a reasonable approach.
- landryraccoon 5y ago

nhoughto 5y ago

Seems very familiar. Its not that finance get things done, its that they can prevent things being done entirely, or reduce funding/investment to the point they naturally peter out.

slt2021 5y ago

IT and Finance report to CEO. If you have clueless boomer CEO who doesnt understand what is digital transformation, then nothing will help

acdha 5y ago
If the CEO doesn't listen to anyone who works for them, you're going to have many problems. If not, you might want to work on your communications strategy: in particular, ask whether the problem is that you failed to communicate the business benefits from a proposed modernization plan or are just repeating consultant babble.
- slt2021 5y ago
  IT and Finance are just executing CEO's strategy and vision. A lot depends on hiring - whether CEO/Board is capable of hiring competent CFO and CIO. WIth CFO it is easier, because finance has been around since money was invented and a lot of hard problems have been solved in finance and there is vast theoretical body of knowledge backing it up. a lot of it is straightforward. with IT it is hard, because the field is changing rapidly and hiring a competent CIO/CTO is harder. this is why CEO is critical, and he is ultimately responsible for the overall outcome

underwater 5y ago

I worked for a company where capex was cheap and easy and opex was hard to get. So engineering managers would pitch a project, fund a bunch of consultants, get a massive piece of tech built and then fail to support it. They knew the downsides, and wanted to build long lived teams that built and support and evolve products, but there was too much push back from finance and IT. I realised that the company would prefer to spend two or three dollars in capex every year than a dollar of opex. This helped me understand how and entire industry of overpaid an underqualified consultants existed. So my question is, where is that accountability for the finance function? Their controls and constraints caused the company to flush huge amounts of money down the drain, just so their books could have numbers appear in the columns they wanted.

catmanjan 5y ago
Was capex preferred due to preferential tax treatment?
- andrewlgood 5y ago
  No. You actually get better tax treatment for opex as you can record the full amount of the expense immediately as a deduction against your income. Capex spreads the deductibility of the expense over years. As I mention above, opex vs capex is usually a compensation issue.
andrewlgood 5y ago
This is not a Finance function issue. Finance has absolutely no preference between opex and capex. There are accounting rules that determine the appropriate classification and any reasonable CFO simply wants it done correctly. The opex / capex argument is about compensation. Opex reduces income immediately while capex gets spread out over the projected life of the asset (software, power plant, computer, etc). By shifting items from opex to capex, a company temporarily increases its net income. Increasing net income often leads to higher bonuses. This is what drives the silly behavior. To make matters worse, once a company goes down this path, it cannot get off as it is usually evaluated on a year-over-year basis. As for accountability, CFOs of public companies can be found criminally guilty (i.e. go to jail) if they not only commit accounting fraud, but simply do not establish a robust enough control framework to ensure the financial statements are correct. CFOs are always accountable ultimately to the CEO and sometimes to the Board of Directors. If you think the CEO was unaware the company was spending "two or three dollars in capex every year than a dollar of opex," your either mistaken or you had a clueless CEO.

rdxm 5y ago

Tried to plant this seed ~3 years ago at a large defense contractor. They were frustrated by the fact they weren't getting the velocity out of gov cloud they expected. my response was: "you literally applied the same procurement / process model to GC that you had for on-prem, wtf did you expect?!?" I called my pitch: "AdminOps", basically the message was the finance/biz side of the house needed the same type of transformative event engineering got with devops. They are still slow....go figure. Enterprises gonna Enterprise....

cratermoon 5y ago

Wasn't if finance that decided that IT should be cost center for every company?

krishvs 5y ago

Wow - the 4 stages of product development seems quite spot on from my experience working on b2b enterprise software products. No doubt that the Finance team is generally the most powerful entity in shaping a companies direction both internally as well as in your clients organisation. Finance teams love control, are (mostly) completely risk-averse and change averse. The same can also be said about badly run IT teams. I have seen product/automation implementations being deployed seamlessly in all departments from operations to procurement but completely fail due to the finance teams. In extreme cases I have seen finance teams have so much power that they even override the CEOs decisions to make implementations fail if they feel that they are loosing control or do not like the new process workflow that the implementation brings in. For digital transformation projects the implementation team needs to speak the language of finance, operations as well as IT. Otherwise the chances of the implementation really succeeding is close to nil.

throwaway984393 5y ago

Conway's Law Revisited: If finance dictates your organizational structure, then finance is developing your software.

lincpa 5y ago

Development ideas from A Certified Public Accountant: Computer science is essentially a management science, and vice versa. Follower's success stories: Apple M1 chip + MacOS come soon: Qualcomm, MS, Intel, AMD, Samsung, Cray, etc. The Grand Unified Programming Theory: The Pure Function Pipeline Data Flow with Principle-based Warehouse/Workshop Model https://github.com/linpengcheng/PurefunctionPipelineDataflow https://github.com/linpengcheng/PurefunctionPipelineDataflow

goforbg 5y ago

I think this is mostly true for B2B SAAS startups but not for the rest.

goforbg 5y ago

I think this holds true mostly for B2B SAAS platforms but not for the rest.

shkurski_ 5y ago

Looking at Marx's quote on relations of production, could the money flow INSIDE a company be another topic that is worth exploration? From my simple understanding, in Marx's theory it was social aspect that mattered most. Do we still face such problems as it was in 19th century? I'm curious whether it would be a viable company where the surplus-value is distributed among workers and I have a fixed transparent salary as CEO and how it will survive the stage I and scale to stage II, IIIa, IIIb.

Vaslo 5y ago

Funny to read this because I happen to work in Finance and many requests come through my team. I’m going to add some advice here for what it’s worth. A manager once told me that Legal and IT will always tell you no before telling you yes. Basically you need to argue and escalate to get things done, and it’s frustrating. After 20 years I’ve found that pretty true, though my current company is a bit better than others I’ve worked for where IT pretty much said no to everything. Everytime you are going to stop me from writing queries to your server, or add more security that makes it harder to do my job without sufficient support to rightfully get through that security, or deny me the ability to download legit software from the likes of Microsoft and other reputable vendors because you have some ridiculous amount of red tape, I’m absolutely going to question every dollar and be skeptical of how the funding is going to help me or my partners in HR, Sales, Supply Chain etc. If your projects help me, and you help me, I’m definitely willing to help you more and fight for you and your budget.

NullPrefix 5y ago
Did you just admit that you fight legal and IT on purpose, just to have a fight?
- ativzzz 5y ago
  He chooses to not help people who make his job harder, sounds like a normal human being to me.
  - 542458 5y ago
    Lots of people make other people’s jobs “harder” with good reason. Things like code review, legal compliance, IT policy, etc are just “overhead” if you assume that your knowledge is infallible - but people are fallible and we need systems like this that, yes, slow us down in order to keep from getting burned.
    - Vaslo 5y ago
      The overhead needs support. If you put up additional security where only IT staff can do x, then you need to ask for or transfer headcount to ensure it’s just as nimble as before. If you demonstrate that ability to manage change, then you also demonstrate that the limited funds are best in your hands.

oneplane 5y ago

It's obviously going to strongly depend on the context, but I've found that there is a shared amount of risk in organisations where IT and Finance both have their choices to make together. Say you have some on-prem datacenter stuff for a manufacturing plant, you might have the request from factory operations to be as available and consistent as possible. The cost for such a thing might be astronomical, but even a single day of downtime would cover it. On the other hand, the finance department might already have made a risk assessment and set aside some money or had it insured in some way and now suddenly the impact of that risky thing that might happen is a whole lot lower in terms of money. For a lot of people that can be news, or unexpected. If you silo yourself you might end up engineering something that is completely unnecessary. At the same time, someone might be buying some expensive insurance, but forgets that unrecoverable data doesn't magically yeet itself into existence just because the insurance contract says so. So while on paper it's "covered and we just need to recover our systems and get on with it", in reality that might not be possible. Starting a company from scratch (IT-wise) leaves you with empty systems. Those are about as useful as a brick. A functional brick, but perhaps not the spreadsheet, email or CRM database you had in mind. This is a lot of words just to say: in most companies (as long as they don't get too big), most people are there for a reason, and most departments are there for a reason. Go up the chain of reasoning and you'll find your common reason ("because the company wants to make money" for example), and now you have all you need to 'justify' to yourself that 'those other people' are there for the same reason and with the same goal as you.

steve76 5y ago

More like end with finance. It's inevitable if you are successful. Why build it if you can buy it? Save. Just don't loose money. Finance is not what most people think. Frugal. Charitable.