from Hacker News

Show HN: Is It Worth the Cost?

by osel on 7/14/20, 10:11 AM with 89 comments

  • by projektfu on 7/14/20, 11:43 AM

    I like the idea, but a lot of other factors are missing from the analysis.

    1. I’ve yet to find a service that saves 5 hours per week per employee. How do I estimate the actual savings in time?

    2. Work expands to fill available time. Will my employee use that time to the company’s advantage?

    3. How much does this increase or decrease my personal time required as supervisor?

    4. From a financial point of view, I’m still paying the employee and the service, so either I need enough services and time savings that I can eliminate a job, or the service has to have positive ROI of its own.

    And probably many more. I often receive proposals of this type, that I could be “saving” so much by buying something. The money goes out up front, the savings are supposed to trickle back in these hard to quantify and use ways.

  • by osel on 7/14/20, 10:15 AM

    A simple check on whether purchasing a service is worth the cost - built in an afternoon in response to a previous HN discussion [1].

    A few people expressed interest in embedding something similar in a landing page, so I've open-sourced the code for others to use if they wish.

    [1] https://news.ycombinator.com/item?id=22497093

  • by glenjamin on 7/14/20, 10:56 AM

    Might be interesting to have a line items for "cognitive cost of having to deal with the crappy in-house tool when you know a good SaaS alternative exists" and "time spent wrangling the finance department"
  • by georgyo on 7/14/20, 11:26 AM

    This seems really flawed.

    Let's plug in numbers. This says that if a service costs 125$/month/employee and I have 1000 employees ($125000/month) I will save ~5 million a year.

    However a service costing 1.5 million at such a company has other costs.

    Namely that a product that costs that much almost always needs an in house support team. I have never seen the case where an expensive product also didn't need 2-3 people in just to maintain and support it. But the employee time savings is still worth it.

    The next cost is much harder. In the tool we said this will save everyone 1 hour a day. However saving people time is not normally what a tool actually does. PagerDuty for example doesn't save people any time, it just sends alerts. Alerts that come from another tool that must be set up. At 40/person/month for 1000 people, that is 40,000 a month for zero "saved time." The value of pagerduty for very different.

    But anyone who has been in the field has seen, that list hidden cost is that a paid service isn't normally a perfect fit. It is missing something, or your particular use case doesn't map cleanly. You then have an army of people working around the tool, saving negative time. The business might have reasons to still use that tool. But saving time is not one of them.

    So seeing this tool tell me that a service which cost $2,000,000 a year is going to save me $5,000,000 feels like naive marketing nonsense. And it isn't even marketing a product.

  • by redwood on 7/14/20, 11:59 AM

    What's this type of calculator often misses is how nuanced and complex is to estimate the cost of employee time.

    It's not just the cost of an employee salary per hour... Is the opportunity cost of everything that employee could have instead done.

    After all it isn't as though with more money you can just immediately get more fully ramped sophisticated employees to add value... All of that is complex.

  • by Flimm on 7/14/20, 11:52 AM

    This looks really cool. I put in some numbers for a service that I am considering, and the answer was the opposite of what I was intuitively expecting, which proves the need for something like this.

    Some feedback:

    - s/(one person)/(per person)/g - The "cover costs" section doesn't really make sense to me. For example, in one example, the breakdown says that I will burn $10000 extra, but that I will also cover costs with 23 hours of use. How does it know how much use it will take to cover costs? How does it know how much profit is generated per hour of use? Am I misunderstanding something? - It would be nice if the formula used was displayed in small text or in a tooltip or something.

  • by drapery on 7/14/20, 3:55 PM

    Okay I really like the idea. I think the bottom line whether the SaaS is worth the cost with the premise is that the SaaS saves time. However, time saving (item 4) is pretty difficult to gauge. Therefore, I suggest having a version where time saving is the output.

    So then the question become, I have to believe that this SaaS will save XXXmin per employee for me to consider buying it.

  • by Abishek_Muthian on 7/14/20, 1:47 PM

    Good work OP. Any plans to add more variables in the future? e.g. Cost of current equivalent service being used.

    P.S. I've added your calculator to my curated list of startup tools[1].

    [1]https://startuptoolchain.com/#operations

  • by KerryJones on 7/14/20, 3:52 PM

    This is looks like a different approach to a tool I also tried to address to answer 'how much time should you invest in training' https://kerryjones.github.io/sharpen-the-saw/

    Had a whole article that went into this: https://medium.com/hackernoon/how-much-time-should-you-inves...

    I like the idea and it's certainly more detailed than mine, but I think some explanation of how this should be used and context is important.

  • by drapery on 7/14/20, 4:00 PM

    I think a lot of users are missing the point. I don't see this tool as a final decision maker and it can never be one. The final decision has a lot of nuisances to that it is difficult to capture in one tool, so might as well just open a spreadsheet and bang it out.

    I see this tool as a quick a simple way to eliminate all the tools that might look shiny and slick or save a few clicks, but doesn't actually save employee's time at all.

    So the decision becomes, I like the way this new SaaS feels or looks than our current one, but is it really worth the additional cost?

  • by _pdp_ on 7/14/20, 2:00 PM

    I like the idea.

    In 99% of the cases, I rather buy the service than develop it from scratch and support it long-term. Most consumer-facing services I've seen are relatively cheap. I only tend to develop existing things if I find there are some functionalities which either I cannot buy directly or they require a special setup that does not go well with security and other constraints.

  • by lukasm on 7/14/20, 2:24 PM

    This is look at the cost, but what about value? Say, the total cost per hour (salary, insurance, benefits) is 50$, but how mych value does he/she create?

    I used to do this calculation with 2x assumption. Say, a new IDE plugin saves 100 engineering hours, so it's 50 000 in cost and another 50 000, because they spend time creating features.

    Is it a good assumption?

  • by saadalem on 7/14/20, 2:40 PM

    This but for E-commerce will be huge ! It would be cool to calculate a product’s overall cost through its life span.
  • by ponker on 7/14/20, 4:24 PM

    The two biggest line items are not easily measured and thus not included: The removal of an external dependency that you don't control and may not evolve with your needs, and the loss of the continuous improvement in that dependency without continuous investment from you.
  • by adibalcan on 7/15/20, 7:20 AM

    Can I save the parameters in URL in order to send at our customers?
  • by osel on 7/14/20, 7:57 PM

    Hey all, thanks for the great feedback!

    I submitted this just before retiring for the evening NZ time thinking not much of it, it was awesome to find this discussion this morning!

  • by AQXt on 7/14/20, 12:58 PM

    Feature request:

    Add "One-time payment" option to the cost of the service.

    This could be used to decide if you allocate resources to implement a new feature in an internal system.

  • by pseingatl on 7/14/20, 5:17 PM

    Saving money doesn't put cash in your pocket. It will help keep the cash you have there, but is nothing but another cost item.
  • by dzonga on 7/14/20, 7:39 PM

    but there's a hidden cost, that's never spoken of. cost of forgotten knowledge or expertise. say you use AWS RDS | Aurora, on paper it's cheap. AWS runs everything for you. what happens, when one day you need the knowledge to run your own db ? or migrate to something different altogether ?
  • by pagade on 7/14/20, 11:09 AM

    First thing that came into mind:

    Is It Worth the Time? https://xkcd.com/1205/

  • by aisofteng on 7/14/20, 3:57 PM

    What about the cost of operating an equivalent service in house?
  • by alecco on 7/14/20, 2:05 PM

    Missing: gains in time to market.
  • by ChrisMarshallNY on 7/14/20, 12:14 PM

    Good tool, but, as someone mentioned, there are a number of other factors.

    1) Brand reinforcement/damage.

    This one is really hard to quantify, but can mean the difference between the life, or the death, of the corporation. If the tool we choose causes some damage to the brand (like not giving us the ability to display the brand in an appropriate manner), or introduces a brand-damaging problem (like causing a particular kind of service to have an extra step or two, or even amplify customer pain), then there could be issues.

    On the plus side, it could also significantly fortify a brand, by doing things like having the brand appear in places that were disregarded, or considered "out of reach," or it could amplify the advantages of a brand-connected service.

    Branding is a "dark art" to many literal-minded folks like engineers, but it is unbelievably valuable. It should always be considered, when thinking about things like this.

    This morning, I had this demonstrated to me in a visceral manner. I made an order yesterday, and asked for expedited delivery. When I made the order, it said the expedited delivery would arrive today, but when I received the confirmation, it said the order would arrive tomorrow.

    Sound familiar? Amazon is notorious for this nasty little trick. Most of the time, it's Prime Delivery, which means we don't really have much room to complain.

    It wasn't Amazon. I won't name the corporation, but it is one that is synonymous with extremely high (arrogant, even) quality.

    This silly little trick, with a cheap, third-party item, caused brand damage with a loyal customer that has spent six figures with them. I won't drop them, and the world won't stop turning because I won't get my item until tomorrow, but the simple fact that they did the "shipping bait and switch" on me, means that I'll never use them again for this kind of purchase.

    That's brand damage.

    To add insult to injury, when I tried contacting them about it, I had to run the gauntlet of what I call the "AI Picador," which throws as many roadblocks as possible in your path, before allowing connection to a human.

    This workflow ensures that I will be steaming mad before I get to the human, and it's all I can do to avoid venting to them (it isn't their fault -it's their bosses').

    2) Staff morale.

    Will bringing in the service reinforce or damage staff morale? It may help folks do their jobs a lot easier (good), or it may result in a lot of folks losing their jobs (bad, or very bad, depending on many factors, like how the layoffs are done, and how the folks that remain are treated).

    3) Dependence/Addiction

    If bringing in a dependency ties us to a corporation/language/toolset, is that good or bad? It may be quite good, if it's a good service, and a good corporation, but it could also be incredibly bad.

  • by tunesmith on 7/14/20, 8:57 PM

    ha, awesome! I've had this idea so many times from the xkcd comic (which I see was inspiration). I'm glad someone actually built a tool for it. It could be a fun learning project for a simple mobile app too.
  • by thatoneuser on 7/14/20, 11:10 AM

    This is great! Walking through the numbers mentally kinda works, but it seems people are always suspicious they're being had. Seeing it live will be very reassuring.