Need advice on this hourly rate for a developer (RoR, AJAX, etc.)

Status
Not open for further replies.
L

Laura

New member
Jan 29, 2007
600
8
0
Any thoughts you guys have would be awesome.

I am working on two new projects that will need to be built from the ground up. Over the last few years, we have done everything from outsourcing to the cheapest of cheap places in India, to having full-time development staff, but the one thing we haven't done is hired a local development firm. Having not been overly satisfied with the results on previous occasions (though we do love our full-time programmer), I think we should go local for these two large projects.

Okay, here's where I need help. We met with a guy the other night who is PERFECT (well, it's his company). I swear that he read my mind on everything - but not in a creepy way. In a, "yes, exactly!" kind of way. This company has done consistent work over the years for a $200 million a year business that's in the niche of one of these projects, and therefore has no problems doing most of the things we need from a technical aspect (this is key - there is no learning curve on the important stuff - much of that the monetizing aspect - that would likely hang up another company) . The key is that they do RoR, which will allow us to launch far earlier and far more often than building something from scratch in PHP. They are experts in all areas that we need. (Not to mention that they LOVE the concept.)

Here's the question: They bill hourly at $125/hour per developer and they like to see two per project for various reasons (we tend to agree). That's $10k a week. Adam (hubby) thinks that's a wee bit crazy, BUT keep in mind that the development time is probably half of what it would be with PHP. I think that if he's nearly as great as I think he is (and in this case, the work they have done speaks volumes), then I think it's worth it. (Did I mention how awesome they are?)

So is this rate crazy for a local, totally in-house web development team that are clearly experts in their field? I know I could go to rent-a-coder or whatever to hire a total lametard for FAR less, but this company is amazing and innovative and experienced and they are even the core audience of one project, so building it would be almost intuitive for them. And that's not to say that we couldn't hire a team at half that rate, but then it might take them twice as long...

So before I sink more than $100k into the first project...anyone have any advice? I know WF is focused a lot on arbi and whatnot, but I am hoping someone has thoughts on this.

Laura :D
 


100k in 1 project? That really goes against everything I believe but if it's proven to make that 100k back in a years time then go for it. As far the investment it's all about an ROI in my book. I'm willing to spend 10k on a project if it means making a guaranteed 50k on the other side.

Now for Indian programmers vs. local. I just don't tihnk I could get myself to pay 250$/hour for a couple of coders. The way that seems to work best and still getting you cheap programmers is to invest in learning to code of partner with someone that does. Then give your programmers specific tasks and you be the glue that brings it all together. Doing things this way I've had some extremely elaborate systems built for 1-2k. We're talking reverse engineering ajax websites (which is a huge pain) then creating full fledge applications with the knowledge.

Here's one thing you have to consider... if you spend 100k on 1 project ,even if they do a great job, that's 100k investment.

What if for some reason that project doesn't work out?

How often do things come in as budget?

Are you really smart to hire by the hour vs. by the project? I might be tempted to charge you a few extra hours and how the hell would you know if I did.

I invite you to think about things from this prespective. If for some reason your idea doesn't work which has been known to happen in IM, as things change quickly. This about if you were to only spend 25k that would give you 4 more chances to have success. Just some food for thought. It might be more work and harder to do and manage but that's the business you're in I think. Just my 2 cents.
 
The rate is probably about right for ROR skills but the problem is the time dimension. It might take n weeks or it might take n+10 weeks and so you have no idea how much you will end up paying for the completed project and the engagement of individual developers on a time and materials basis provides them with an incentive to take longer.

The best option would be to offer a really tight specification and ask them to quote a fixed price on that. It's harder for a firm to give a big number than it is for them to give an hourly rate and leave the time frames approximate so there is an advantage in doing that. You could also suggest that any changes would be quoted on separately (fixed price as well). So if you decide, half way through, that you wanted to use a different AJAX library, they would scope the change and quote on it.

If you go down the fixed price route, you can also say to them that it is time and materials till you get the agreed spec and then it's fixed price from then on. That mitigates against the time risk for them because the specification stage can also go for an unknown amount of time.

Hope that helps (and makes sense).
 
Btw, you mention RoR, and how it would decrease the time the project takes vs. PHP. Have you considered looking into any of the Framework PHPs offers? There's CakePHP and Symfony for example, just to name the more popular ones.
 
As far as $$/hour, that seems reasonable for good coders who know what they're doing. Not sure why you need two coders on it at once (pair programming? Pah.).

Is there any way you can break this thing down into modules? That way you're sinking substantially less that 100k into it in one go, but you've got something you can review at the end; if you can get some income out of that then you'll also be helping your cash flow. In any case this might give you a getter idea of the team you're working with and whether you've got a future together.

Am1r hits on a good point with the very tight specification, but I disgree with having a set timeframe; this is a matter of taste (and personal experience), though, so a discussion on the merits and demerits of fixed/per hour is, in the language of WF, worseless.
 
Hey guys - thank you for the feedback! I have to run to a birthday party, so I can't reply in depth, but I will when I get back. I just want to let you know that I appreciate you guys taking the time to share your thoughts.

More later - lots to say! :)

Laura
 
Back when I was working as a full time ASP developer (mind you, this was like 1997-1999 pre-dot-bust) my firm was billing anywhere from $50-110 per person per hour. That was nearly 10 years ago.

Yeah, $125 isn't cheap, but like you said - if you're getting double the production output of an outsourced firm, then you should be all good. The most important part of everything is to make sure that you have regular milestones and that you're not required to pay extra for things they mis-quoted.
 
ROR isnt gonna speed your development time up significantly compared to a similar php framework, $125 an hour is pretty standard for a consulting firm... BUT realize that they are paying the 2 developers 20 - 30 an hour MAX so you are paying a very high premium just for the convenience of not having to find your own quality programmers. Ruby on rails is simply a development framework which they have obviously sold to you as something that raises their value significantly... But in general I would stay away from development firms and try and build a solid local or outsourced development team...
 
Ditto to what gopherr and lerchmo said about RoR vs. PHP. Comparing the two is apples vs. oranges, because RoR is a framework while PHP is just a language. If you compare language to language (Ruby vs. PHP) or framework to framework (RoR vs. CakePHP, etc.) then the difference really isn't that great.

sleepylee's suggestion of splitting it up into modules is a good one. Depending on the size of the module, maybe you could even have these guys develop one module and have Rent-a-coder type guys do the same thing and see for yourself how much difference there is.
 
Back! Okay, here is some more info...

I should have clarified that I am the one who sought out a RoR firm (rather than them selling me on it). I do realize that there are PHP frameworks, but I also realize that you just have to go with something or nothing will ever get done. (Which isn't to say we aren't flexible...) Saying, "I want to try RoR," dramatically cut down on decisions and options (in a good way).

That said, this company has a great way of tracking their projects for clients, and we were able to see a few in progress and exactly how that works. Basically, the first work week is laying out the site with the team - it's a giant brainstorm session where everyone puts out ideas. Because I have a 30 pages of mockups done, ours should be easy. :) Everything gets talked about, and as we talk, things are broken down into, well, index cards (they wrote a program with a system to manage everything and track progress). So every feature is a card, and each card is assigned points based on importance and then difficulty. For example, "Users can upload profile pictures" is a card (then it's also given a very low difficulty level). Once that's done - and it's hundreds of cards, usually - they set forth a certain amount of points that they can accomplish each week (as the project grows, they can do more per week, so the first week is the slowest). What they tackle is determined by importance and difficulty. If they finish what they set out to do, they just grab more cards until the work week is over.

From that big list, we determine what's in launch one - which is probably about 3-4 weeks after we start. Then we can determine at what intervals we want to add features (ie. every two weeks, every week, etc.), and we map out what will be in those releases (which will also be determined by users). It's definitely a launch early and often approach.

Full reports are given at the end of each week that show how many points were planned, how many were completed, and what those were. They like to show their clients new features daily, and we can go there to bang around ideas and interface possibilities if we want or need to.

Oh, and 98% of their clients return for a new project. That's pretty darn good.

We could always do one coder at $5k per week, but it will be slower. It's not 1 to 1 - doubling the programmers doesn't double the speed - but that's an option.

As a lot of you guys stated, we would be paying a premium for the luxury of being able to meet with the developers regularly, plus they take care of pretty much everything. Not to mention that they are extremely experienced in our niche - which means a ton. Thanks to everyone for your thoughts - it's definitely given me more to think about.

Decisions, decisions... thanks, guys. :) Any more ideas are welcome! No need to sugarcoat anything. :D

Laura
 
if they are a solid firm and you have confidence in them its definately worth the money. But I would suggest talking with previous clients about the timeliness and quality of their work..
 
lerchmo said:
if they are a solid firm and you have confidence in them its definately worth the money. But I would suggest talking with previous clients about the timeliness and quality of their work..
Click to expand...

Very, very good point. I will do that for sure.

Laura
 
Hmmmm...... Something sounds fishy. 98% return rate of clients is unheard of.

If they are as good as they say they are then they should be able to give you a firm quote and stick with it. Only modifying it when you deviate or get a little bit of that scope creep in there. It sounds like a lot of the billable time will be burnt up in meetings too. Are you relying on them for guidance on features and design, or are you pretty set with how you want things? If you want your vision to be developed then you may be paying extra for something you don't necessarily want from them. Meetings kill budgets and add billable items!

As for the framework, don't just pick a framework because you'd like to see something developed on it. Pick a platform that works for the project, is there something RoR offers that PHP or ASP doesn't? How does it effect your future scalability and hosting requirements? Will it be easy to handle higher traffic spikes or spread across multiple servers if the need arises?

I also agree that you should speak to some of their former clients. Even do some research on the net about them. Your sure to find a few disgruntled clients. When you find those, is there a pattern of the same complaint or just a few random ones?
 
This totally depends on how much it will hurt you as far as your current business income levels. If $100,000 isn't too hard on the business and you are confident that someone in India can't do the quality you need, then by all means, hire them.

Personally, I would look into someone from Britain or Russia. Those programmers can do some DAMN good work. They are usually the ones quoting you the high prices on elance and other sites because the quality is outstanding. That being said, their prices won't come close to $100,000.

All I can say, is I know this situation is coming for me soon and I personally don't want to deal with that yet.

Keep us updated.
 
If you're in the US what state are you in? $125 could either be right or high dependent on that factor alone. I don't usually see RoR or PHP for that matter getting that much.

Most professional shops are going to have significant project tracking capabilities for their clients on their extranets - it's pretty much standard and I would be leary of one that doesn't provide for that.

Most professional shops gain clients that usually return for work and I wouldn't look at that as a selling point. It's a simple fact. Once a company starts to have development done by one shop, it's more difficult to go somewhere else for enhancements and teach someone else your business model. It's just easier to stick with one shop as much as possible after you've selected them and made an investment with them. Breaking away isn't really cost-effective and can incur a certain amount of pain.

You should at the very least negotiate that hourly rate down. There should be some fluff built in for clients to be able to do that. Otherwise at that rate for RoR they're just raping you.

Since someone mentioned hourly versus flat-rate per project I will add though that most experienced development shops, that aren't some guy working out of his garage, will avoid flat-fee projects like the plague. Hourly rates allow them to be more flexible and work with the client while not getting screwed. There's always scope-creep, and when you're constantly saying "no" to the client, without an increase they tend to get a little irritated since some level of perceived control is taken away from them. Whereas with a per-hour project, they know the cost up front, have agreed to it (mentally as well), and therefore have a deeper understanding that any changes come with additional costs.
 
Flat fees will work if the specification is highly specific, agreed by both parties and there are mechanisms for managing any changes or 'scope creep'. For example, we usually do pro rata (time and materals) up until we have the formal agreement and specification, then they build the spec for the agreed price and any additional functionality or changes get paid for pro rata or by agreement. It seems to work well but, as I said, the proviso is that you have a water-tight and detailed specification that the developers can work from.

People also forget that there is a layer between business knowledge and development knowledge and that is the analysis/design. You can engage developers in the third world who are cheap, technically brilliant BUT you run a major analysis/design risk because good analysis/design involves an iterative and interactive process that can't be easily done when the analyst is sitting in a different time zone and on a different continent. It's fine if you have that development document down already because then they can just work from that but in the case of someone who isn't' at that stage, then I recommend they look for a local company. Often some of the problems won't appear until later on when you discover, for example, that the database design doesn't accomodate your complete business model (a common problem with poorly designed/analysed projects).
 
I'd say that's way too much, but this is from a RoRs programmer's perspective. I know it's much different in the business world. I guess that's where I need to be.
 
Last edited:
By the way, for a true professional programming company, $125 is not a bad price. I charge anywhere from $100 - $150 an hour. One company I work with bills me out to their clients at $250 an hour.
 
smash said:
Hmmmm...... Something sounds fishy. 98% return rate of clients is unheard of.

If they are as good as they say they are then they should be able to give you a firm quote and stick with it.

As for the framework, don't just pick a framework because you'd like to see something developed on it.
Click to expand...

I agree, it seems a bit high, but we did look at numerous different projects for several large clients (each client had multiple different projects that they hired them for), so if those clients (which most of us would recognize - very large brands) returned time and again, IMO that says something about their ability to stay on target. That doesn't say everything, of course, but it's worth something... (Plus, what they built was seriously hot. The backends on these custom sites were amazing.)

I chose RoR because it made sense for what we want to do. I spent a couple of months reading and trying to understand different platform options, and RoR always came up on top of my short list given our needs and desired features. Once I thought I had a good grasp on the options, I talked to my sister at length, who is a developer at Microsoft, for her opinion. I sent her a 30 page write up on one of the sites and she suggested RoR as one option that would work well. That pretty much sealed it for me as a good direction to go in because it backed up what I learned myself. Hopefully, that makes the logic more clear. I don't want it to seem like I got sucked into RoR without much thought. And who knows, we may still go with a PHP platform after all...

am1r said:
People also forget that there is a layer between business knowledge and development knowledge and that is the analysis/design. You can engage developers in the third world who are cheap, technically brilliant BUT you run a major analysis/design risk because good analysis/design involves an iterative and interactive process that can't be easily done when the analyst is sitting in a different time zone and on a different continent. It's fine if you have that development document down already because then they can just work from that but in the case of someone who isn't' at that stage, then I recommend they look for a local company. Often some of the problems won't appear until later on when you discover, for example, that the database design doesn't accomodate your complete business model (a common problem with poorly designed/analysed projects).
Click to expand...

I think you totally nailed what I was trying to express (poorly!). While I have exact specs of what I want (features, interface, etc.), just the one meeting with the owner of the company gave me ways to improve what I had - and I hadn't even shown everything to him. It was amazing to be able to develop it further with him in just an hour and a half. Also, it's a complicated application that requires major forethought and planning for it to run properly once active. They understand this and can easily plan for this.


Thanks again to everyone for sharing your thoughts. I truly appreciate the insight. I am going to meet with them one more time to talk more about the project, timelines, etc., and then see where that leaves us. My gut tells me this is the right company for this job, but I can also see each of the points that you guys make about why it might not be the right plan.

Either way, I will let you guys know. This scares the hell outta me, but that's all the reason I need to push forward. :)

Laura
 
Status
Not open for further replies.
Top Bottom
Back