Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Ask HN: How much is reasonable to charge for rescuing a web app?
20 points by AlexITC on Feb 7, 2023 | hide | past | favorite | 28 comments
Hi fellow freelancers at hn,

Over the last years, I have been commonly found with potential customers in situations where their production app is either down or its main feature is not working (my last work involved a project where previous dev deleted all prod infrastructure).

I have helped them to get their production app back to a working state, which usually takes less than 10h, I just did that again and it took me only 2 hours (causing a tiny bill)! I'm feeling that these small bills aren't worth it for the emergency mindset.

I'm aware that billing by value instead of by time is a common advice but it isn't clear to me on how much I could actually charge for this kind of task. Is this something we can put a price tag before having enough context?

P.S. The reason I usually charge by time is because I have got many estimations wrong, there was a time where I worked on a so-complex project where getting it back on track took many exhausting days.

I'll appreciate any advice.



Let your customers tell you: just keep raising your rate until potential customers stop converting.

I recall a HN comment that said they kept doubling their rate way beyond what they believed they could charge but the (new) customers kept happily paying.

Marketing skills will help increase the maximum rate customers are willing to pay.

Some ways to figure out the value to your customers:

- Estimate the cost of not fixing the problem. How many customers will they lose; what is the LTV of a single customer? How will they have to compensate current customers? etc...

- Estimate how much time it would take them to fix the problem themselves, and multiply by $50-$100/hr. I assume you can fix the problem much faster than your customers, and people generally think their own time is worth at least $50/hour.

---

Related anecdote: apparently plumbers' favorite contracts are fixing hot water boilers. It's a relatively easy procedure, but they are able to charge relatively higher compared to other types of contracts. When people don't have hot water, they want to fix the problem fast! So they don't haggle or even shop around much.

Compare to a leaky faucet: it doesn't need to be fixed right away, so customers tend to shop around and complain about the cost.


Everyone wants to charge based on the value you can bring your client.

But, based on my experience, it depends a lot on your client and business relationship with them. If you have a good relationship with a client, you can generally get a sense where their finances and business trajectory is. When the big estimate comes, they're okay with it because they know they'll get the results, and finding someone else has huge opportunity cost for them (the devil you know). And, the opportunity cost of finding someone else would be astronomical if it's an emergency.

So when your next agreement expires, just up the price to whatever you think is fair for you, you can even just come down on price if it looks like they won't close. Plus, emergency response is a huge inconvenience to you, so that should cost big time. Personally, I would create a separate fee structure for it, and of course a different line item on my invoices.

Whatever you do, just remember that it's a business relationship, and relationships go both ways.


1. Don't expect you'll be cost effective for all potential clients that come to you with this sort of situation.

2. Charge a fixed fee - or even hourly - to figure out wtf is going on and assess things. This is effectively a discovery/consult. For some reason many independents can't wrap their heads around charging for what seems - to them - to be "nothing" but it's literally what consulting is. Get into the issue and help the client map a way out. Then...

3. If you want to help the client implement the map, you are also now in an informed position to give them a fixed fee arrangement to do this - the implementation bit. Or if you stick with hourly you can more realistically set expectations and estimate workloads.

Most independents are great at #3 (though most seem far more comfortable charging hourly for this implementation work rather than fixed fee, but that's fine too - at least they'll be able to estimate it better after doing #2 and end up with a better reputation), but horrible at realizing their real value is in #2 (as well as bridging the gap between #2 and #3).

P.S. If you invest in getting better (more comfortable and effective) at charging fixed fee you can rightfully charge more because you are literally being paid by the client to offload uncertainty (risk) onto you.


From what I can tell no one answering his pointed out that you have the core of a very very good business. And since you can roughly predict how long you will take, you can now bill out at whatever amount makes your life most comfortable.

I am fascinated that you have done enough of this work to know that you can estimate (to yourself) about 10 hours per job. I can also tell that you are a kind person, because you are clearly undercharging and you even felt guilty for solving a problem fast.

As the owner until recently of a web based business for 20 years, I would have loved access to a resource like you. I assure you that when a website goes down, the owner will normally be thrilled at an estimate of, say, $2000 for 10 hours’ labor (not that you would disclose this number; I mentioned because it means the turnaround would be fairly fast).

You should be loving life right now. The only reason you aren’t is because you are undercharging by a lot.


Add a minimum fixed fee + hourly after that. For example, you can say "I usually charge a minimum of 5K to solve these problems but it could bhe more depending on additional time spent"

Up to 10 hours: fixed fee (.e.g $5K)

11 hour or beyond: $5K + $x/Hour

Charge the min. amount upfront and then bill later for additional hours.

Something like that can ensure that you are getting paid some minimum but are not going to lose money on large amount of work.


Something like this is a good fit for a retainer basis -- e.g., $250/hr for a minimum of 10 hours/month, paid regardless of whether they consume those hours or not, with additional hours at fee. Just make sure that your SOW is explicit on when and why they can pull you in.


Interesting, so, would you be billing X/month + any hours they consume? Sounds like a good idea.

I have got a few retainers worth ~5h/month where I just reserve that time.

Thanks.


Yep! Basically, you'd point out that it's not economically-viable for you to provide a small number of high-value, low-revenue on-call hours on demand, so they'll need to reserve some time each month to make sure you're available when needed. (It's a bit like the old joke: "Ten cents for tapping with the hammer, ten thousand for knowing where to tap with the hammer.") If you've done BigCo consulting, it's essentially an extended hypercare scenario.


You should look into Jonathan Stark and “hourly billing is nuts”

https://jonathanstark.com/hbin

He also talks about “value pricing”, where you charge clients a percentage of the value you bring them.


I'll check it, thanks.

> He also talks about “value pricing”, where you charge clients a percentage of the value you bring them.

Sometimes I wonder on how many freelancers actually do this, I have met many people like me who are not brave enough for this.


> I have helped them to get their production app back to a working state, which usually takes less than 10h, I just did that again and it took me only 2 hours (causing a tiny bill)! I'm feeling that these small bills aren't worth it for the emergency mindset.

As others on HN will tell you (I'm pretty bad at following these advice, but I try) you should charge based on value. That means, the value you are providing to your customer. Max Value = prod * days of lost revenue. (could be zero if the App is not monetizing). The minimum amount you can charge is your hourly rate, that's your base, you shouldn't go lower (example of an app that is making less per hour)


> As others on HN will tell you (I'm pretty bad at following these advice, but I try) you should charge based on value

I wonder how many people actually do this, I have failed to follow this advice many times.


How you maintain your business relationship with your client is up to you of course and involves more variables than we have access to, but I'd suggest rather than thinking about charging them more per hour (unless it's a weekend/holiday/after-hours) make sure you never charge so few hours because it's super annoying for any developer to drop everything and work on a new project for just 2 hours of income.

And this isn't a dishonest recommendation. If they make some error, try and figure out a way to prevent it in the future: whether it's a better deployment script or something else that works to prevent this class of error.


> make sure you never charge so few hours because it's super annoying for any developer to drop everything and work on a new project for just 2 hours of income.

This is definitely what I should have done differently, like another hn'r said, setting up a minimum amount of hours would have been better.

Thanks.


You are on call presumably and called on demand so work is patchy? I would charge 10x normal rates but fix the price (the quicker you fix their issue the more you are worth). Maybe $20k per incident? Then that frees you up to keep up to date on techniques, tools and so on to make your next hero feat even better. Remember: they called you because their guys couldn’t fix it, so all these will be perceived as hard problems.

I remember someone on HN charged a monthly fee, spent all the time making the prod environment solid then collected a big salary for being on call but never being called :-)


One way:

Contract work (e.g. fixed allocation of hours every month) gets charged at $ X / hour.

Ad-hoc (e.g. emergency work, one-off tweaks etc.) work gets charged at $ (X * 8) / hour.

I negotiate my hourly rate depending on how much work is on offer.

When a client ask I explain that a repeating allocation of fixed hours means I don't have to spend time on sales & negotiation to fill those time slots every month. That's why it's so much cheaper.


Hourly is the way to go, just set a minimum charge or an added charge because they want it now. I would suggest a minimum of a workday.


> I would suggest a minimum of a workday.

This is key. An emergency disrupts your scheduling and easily justifies a workday as a minimum.


Agree, another friend suggested setting an emergency rate which is 2x the normal rate.


I like this idea, it would have been specially useful in my last situation where I figured out how to deploy the app in just 2h, now, my customer thanked me and moved on to give the missing work to someone else.

Thanks.


Ask them what their budget is.

I was in similar situation to you, and because the work is easy for us it's easy to undervalue ourselves, but ask them what budget they're working with and go from there!


Ask them what it is worth, not what they think they can pay. What is the dollar cost of their site being down?


This sounds like "what's dollar-cost to you not getting the urgent surgery today?"

While some customers can agree, it just does not seem correct, at least, not something I'd do.


Agreed with this. I have never bought anything from a person who asked us what our budget is.

We had a situation where we were evaluating several crappy providers and a really good one. The really good one's price was "call us for pricing". We just mentally translated to say "basically all the money we have" and went with a crappy provider.


That's basically extortion. If I applied your logic to a car, my next repair bill would equal a new car.


> That's basically extortion. If I applied your logic to a car, my next repair bill would equal a new car.

No it won't. Just based on this HN comment, it's obvious you don't value your car like a new one, and you will take it to another car repair place even if you've gone to your 'extorter' car repair for 20 years.


It would be extortion if you caused the downtime.


I will certainly do this next time, like another hn'r suggested, billing for at least 1 day seems worth it too.

Thanks.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: