Ballpark pricing for PHP site development

[DSG]

I've been talking with 4 vendors in the last few months on a PHP-based site with the following features:

1. Social networking
2. Search (robust) with results dependent on relationship in #1
3. Multi-lingual, probably on two separate platforms (English + Chinese)
4. Chat/IM (Ajax implementation)
5. Admin (could be Joomla-like)
6. Field-separated batch upload (through CSV file, for example)
7. Obviously requires XML database to track members, etc.
8. Credit card payment module
9. Optimized tagging/design for SEO purposes

Should initially support 5K to 10K concurrent users.

Being located in China, I was expecting fairly low estimates. The range is US $1.5K for the entire project to US$ 29K. Seems quite high, particularly at the top end. Yes, there will be some custom coding that will need to be done, but most of the modules can be found either free or for very little $$$.

I'd like to know what the community feels in terms of "fair pricing" for such a project.

Thanks for your help. You can PM me if you / your group would like to be considered.

[jmeijers]

I think that 1.5K is way too low to be any good. I would expect 5K if you use standard parts and have some really good programmers, maybe a bit more. Although I have to admit we work with companies in Eastern Europe, not China.

May I warn you that it sounds like you haven't made a very detailed plan and that may cause problems. I have had some bad experiences where I had high expectations and the final implementation wasn't good. The difference is pricing may be a result from your lack of very detailed documentation (i.e. stuff like 'use <insert php module here> for IM').

[PsiPro]

4. Chat/IM (Ajax implementation)
5. Admin (could be Joomla-like)
6. Field-separated batch upload (through CSV file, for example)
8. Credit card payment module
9. Optimized tagging/design for SEO purposes
Should initially support 5K to 10K concurrent users.

1. Social networking
2. Search (robust) with results dependent on relationship in #1
3. Multi-lingual, probably on two separate platforms (English + Chinese)
7. Obviously requires XML database to track members, etc.

I made a site for a client with almost these exact requirements (minus the SEO and batch upload). I took their requirements. snipped out everything in the top box and told them this would take 6 months to do properly and priced the work accordingly. For a US company I say $30k is cheap. Keep in mind decent programmers working for a firm make $40-60k a year conservatively. Top that with a 25% overhead costs for the firm and your only looking at being able to hire a single programmer for less then 6 months.

Again these are US standards for work. $1.5k is an EXTREMELY low price (I won't consider a larger project for less then $1.2k) and I would doubt their abilities.

Yes, there will be some custom coding that will need to be done, but most of the modules can be found either free or for very little $$$.

This thought process has pervaded a few of my clients. Yes, a programmer can mash-up what free code he can find (license permitting) to give you a site. But then you have to deal with security issues related to the software your using (now you need to worry about upgrading 10 different packages). Additionally only the coder who wrote it will really be able to upgrade it because he will likely have hacked the core files and you won't know what he changed. This will end up costing you more money every time you need to push a new security patch or major version upgrade.

In the end the mash-up of code will cost you more money, and with the mash-up you won't succeed because it will quickly become a massive headache.

Sorry you get what you pay for and you want a LOT for not much apparently.

When considering the bids ask them how long you think it will take, press them on security issues, ask them if they stand by their code and if they will track and fix security issues for you (they should do it free, I have never had a problem with a paid project).

[DSG]

I made a site for a client with almost these exact requirements (minus the SEO and batch upload). I took their requirements. snipped out everything in the top box and told them this would take 6 months to do properly and priced the work accordingly. For a US company I say $30k is cheap. Keep in mind decent programmers working for a firm make $40-60k a year conservatively. Top that with a 25% overhead costs for the firm and your only looking at being able to hire a single programmer for less then 6 months.

Again these are US standards for work. $1.5k is an EXTREMELY low price (I won't consider a larger project for less then $1.2k) and I would doubt their abilities.



This thought process has pervaded a few of my clients. Yes, a programmer can mash-up what free code he can find (license permitting) to give you a site. But then you have to deal with security issues related to the software your using (now you need to worry about upgrading 10 different packages). Additionally only the coder who wrote it will really be able to upgrade it because he will likely have hacked the core files and you won't know what he changed. This will end up costing you more money every time you need to push a new security patch or major version upgrade.

In the end the mash-up of code will cost you more money, and with the mash-up you won't succeed because it will quickly become a massive headache.

Sorry you get what you pay for and you want a LOT for not much apparently.

When considering the bids ask them how long you think it will take, press them on security issues, ask them if they stand by their code and if they will track and fix security issues for you (they should do it free, I have never had a problem with a paid project).

As for both replies, this is very helpful. We will be scoping out page-by-page wireframes for the chosen partner to follow. Also, I suspect we'll be spending a lot of time at their office to help with the day-to-day decisions.

Regarding the comments above, this is the very thing we are wrestling with, the quality/cost equation. I've had experiences where vendors have provided high-cost garbage, low-cost garbage, low-cost quality and high-cost quality. In some partnerships it's easier to control because the service is easy to understand. In a situation like this, where we have little knowledge of guts of the product, it makes it more difficult.

In your opinion, is it better to bring in a partner with a coding background to assist? Also, when is the best time to do this, and can it be done after the vendor has finished his initial work? How valuable is high-quality documentation in maintaining the platform after delivery? How often would you expect code maintainence on a project like this -- every day? Three times a month?

In China, a typical salary for a developer would be RMB 3500 to 8000 per month, or US$ 6200 to $14K per year, so $7K for the project would not be unusual, maybe you throw on another grand or two for overhead. 6 man months sounds about right, and then you have testing and project managment costs. Although you say US$1.5K is super low, it's more possible in China, given the low cost of labor.

Thanks again for your comments.

[NickReffitt]

PsiPro hits it on the head, and too many companies expect to pay ridiculously low prices for highly complex systems developed.

You also mention bringing in a partner, but you will struggle to find a quality developer who will partner up with you, simply because everyone asks for a partnership as opposed to paying for the project. Large-scale projects usually require a team of developers too.

[PsiPro]

In your opinion, is it better to bring in a partner with a coding background to assist? Also, when is the best time to do this, and can it be done after the vendor has finished his initial work? How valuable is high-quality documentation in maintaining the platform after delivery? How often would you expect code maintainence on a project like this -- every day? Three times a month?

Maintenance on a completely in-house system will be less then grabbing parts from here and there. The reason for this is instead of malicious users being able to download and view the code on some other website, they have to guess and check where your programmer missed an escape or some other flaw.

Any in house system won't necessary have fewer security flaws, but they will be much harder for a malicious users to find an exploit.

That being said, I would expect maintenance on a 'mash-up' system to be dependent on patches to your component system. At least once a month, and these patches will likely be complicated to install to insure full compatibility.

In a fully in-house system the maintenance will be less, unless there is a problem your are only going to need maintenance when you want to add a feature or find a bug. (And your programmer should be willing to fix any real bugs you find for free, after all they screwed up).

It may be hard to find a coder who will want to partner up with you, keep in mind we get project proposals all the time and people try to sell us a 'stake' in the final product. We typically are not terribly interested in the business so much as we are in making the program (I know I enjoy it). It would be good to bring someone on if you can find a programmer, mostly because he will be qualified to evaluate the quality of the code.

Another option is hiring a coder on 'retainer', I started doing this when people came to me month after month for changes. I am paid a larger monthly amount ($10k, I also work hourly for smaller clients) and do (with limitations) all of the maintenance coding and bug squashing they need. Most of my job is code review though... I require all of the code written for my client to be published to svn (or equivalent version control system) and I check every change for basic security stuff. This is mostly because it takes longer to fix it later then it dose to catch it early. For my more active clients it takes me hours every week to review all of their website's code changes.

Anyway in the long run it will be cheaper to run an in-house solution, but in the short term it will take longer to code and be more expensive. However you will be much more satisfied with a custom solution. And yes, if you can get somebody in at the ground floor to at least follow the coding team you will be in a better position then bringing in somebody later to do maintenance.

In China, a typical salary for a developer would be RMB 3500 to 8000 per month, or US$ 6200 to $14K per year, so $7K for the project would not be unusual, maybe you throw on another grand or two for overhead. 6 man months sounds about right, and then you have testing and project managment costs. Although you say US$1.5K is super low, it's more possible in China, given the low cost of labor.

Good to know I have only outsourced over there a handful of times and its always be exceedingly small jobs I was just to tired to do. Always small enough I could do a code review.

Sorry I tend to ramble.

[DSG]

Great advice, thanks. We have talked about a retainer situation. I'm leaning towards initially outsourcing and then hiring a developer(s) after launch. If the platform features lots of out-of-the-box solutions, support will be much more robust. If we go a purely custom route, the documentation needs to be top-shelf so that the developer can easily get a handle on the system, yes?

Regarding the partner issue, I kind of agree. Your partner(s) need to buy into the business idea and it must be more attractive that their current situation. This is not impossible, but if they already have a thriving business, there is no point in taking a flyer on something that may or may not work.

Like your profile pic, I'm a diver myself. Cheers.

[PsiPro]

Well it sounds like you have thought things through well, keeping that long term goal in mind is important.

I love diving, that picture of me in ginnie springs, gainsville florida. Did my first double stage dive last week... 120min at 100ft with 30min of deco.. was sooo tired. Man I love diving.

[DSG]

Well it sounds like you have thought things through well, keeping that long term goal in mind is important.

I love diving, that picture of me in ginnie springs, gainsville florida. Did my first double stage dive last week... 120min at 100ft with 30min of deco.. was sooo tired. Man I love diving.

Yeah, thanks for your help. I may have other questions that I'll post in the forum after our 2nd round meetings with the vendors in a couple of weeks. Appreciate everyone's help.

Nice dive, have always wanted to do Florida and the manatees. This is quickly degenerating into a diving post.

I did Philippines and Sulawesi, Indonesia last year -- un-freakin-believable. The stuff you see in Sulawesi is from a different planet. Highly recommend.

[PsiPro]

Since I live in Florida I really need to visit the Bahamas, I want to do some deep wall trimix diving...

I really wish I could dive the Philippines....

[clmpwnz]

You could get away with $15k-$20K with this in the US. Tell me in six months that I'm right about what's gonna happen with your project:

1. You're gonna outsource this to someone in china who is hugely under experienced, has poor English, and sells you on a low price point
2. When the project is supposed to be done, and after you've sent the first payment three months ago, all you'll have is a half-assed design and MAYBE and html/css framework
3. The project continues for another three months, and they get a couple of those "pre-built apps" you mentioned integrated within the html / css framework, but none of them work together and you have 5 different databases installed
4. Frustrated, you finally give up and discount all web designers and developers as scam artists

But go for it, you obviously knew how you were going to proceed before you asked for advice = )

[Woodcs82]

Hey try Naples,Bonita,Fort Myers,Florida Web Site Design/Development & SEO - FREE QUOTE

Tell are very professional,great work and honest.

I have worked with them in the past.

[rpeterson]

Hello,

Why don't you go with joomla ?? Joomla has capabilities to manage these sort of big websites and no one knows joomla better than us as we have been working with joomla from its existence.

We are able to development community website for you in the requested manner while covering all the functionality you have mentioned and budget will be near or lower to $1500.

Please go through our Joomla portfolio(through my signature) where you will be able to find good number of social community websites developed by our professional joomla/php programmers.

Thanks and looking forward to have your thoughts.

[ChristopherCloudNine]

I would agree that 20K - 40K is a reasonable price for the project. Too low equates to low quality. Most people do not understand the backend development in web design and expect low prices.