How do I find a company to make a website?
January 11, 2011 1:18 PM Subscribe
How do I find a company to make a website?
A friend of mine needs help creating a somewhat complex website and I have no idea how to begin or who to ask. If you folks know anyone who would be interested in doing this sort of thing, I will pass the info along. From my friend
"We are thinking about trying to create an online text or college study aid. We know what we want to produce, and who to sell it to. But we're going to need technical help, it terms of finding someone or a firm that can maintain a web site, do layout/design, manage servers, provide e-commerce solutions, that type of thing - and figure out how much it would all cost."
A friend of mine needs help creating a somewhat complex website and I have no idea how to begin or who to ask. If you folks know anyone who would be interested in doing this sort of thing, I will pass the info along. From my friend
"We are thinking about trying to create an online text or college study aid. We know what we want to produce, and who to sell it to. But we're going to need technical help, it terms of finding someone or a firm that can maintain a web site, do layout/design, manage servers, provide e-commerce solutions, that type of thing - and figure out how much it would all cost."
For research, have a look at freelancer.com. Start searching something like "website design".
You'll find "projects" where people ask for what they need. Start looking at the way the formulate their requests, try finding for some project that seem to have the scope/complexity of what your friend needs.
You'll see below that dozens of freelancers bid in order to get that job. You will get an idea of what would be the cost for having this done by (generally, but not necessarily) cheap workforce in india or the philippines.
This exercise will help you formulate your requests when you'll ask for somebody to do the job (whatever way you find them) and a ballpark figure of a competitive price for it.
posted by madeinitaly at 1:46 PM on January 11, 2011 [1 favorite]
You'll find "projects" where people ask for what they need. Start looking at the way the formulate their requests, try finding for some project that seem to have the scope/complexity of what your friend needs.
You'll see below that dozens of freelancers bid in order to get that job. You will get an idea of what would be the cost for having this done by (generally, but not necessarily) cheap workforce in india or the philippines.
This exercise will help you formulate your requests when you'll ask for somebody to do the job (whatever way you find them) and a ballpark figure of a competitive price for it.
posted by madeinitaly at 1:46 PM on January 11, 2011 [1 favorite]
It would be helpful to know your budget. low 5 figures, high 5 figures, 6 figures?
There are a myriad of options available here.
posted by bitdamaged at 1:46 PM on January 11, 2011
There are a myriad of options available here.
posted by bitdamaged at 1:46 PM on January 11, 2011
This is not that different from asking, "I have an idea for a house; who should I hire to build it and how much will it cost?" The answer depends on the details.
Start by searching for web developers, not just a web designer. A small to medium-sized agency is likely to be a good fit -- the right combination of breadth of skill and personal attention. You'll want to be able to give them a fairly detailed outline of what you want the site to do.
I can't say how much it'll cost without knowing more, but I'd expect it to be in the five-figure range, at least.
posted by me3dia at 1:46 PM on January 11, 2011 [1 favorite]
Start by searching for web developers, not just a web designer. A small to medium-sized agency is likely to be a good fit -- the right combination of breadth of skill and personal attention. You'll want to be able to give them a fairly detailed outline of what you want the site to do.
I can't say how much it'll cost without knowing more, but I'd expect it to be in the five-figure range, at least.
posted by me3dia at 1:46 PM on January 11, 2011 [1 favorite]
Nthing the need for a set budget. This sounds like it'd be a bit to big for a single freelancer to handle, so you're probably going to want to find a small studio.
posted by thsmchnekllsfascists at 2:05 PM on January 11, 2011
posted by thsmchnekllsfascists at 2:05 PM on January 11, 2011
I generally recommend Cotton Rendle, who have done projects of various sizes. They're good people and will give your friend decent feedback on feasibility/price. Full disclaimer: I'm personally acquaintanced with them.
posted by Zarkonnen at 2:26 PM on January 11, 2011
posted by Zarkonnen at 2:26 PM on January 11, 2011
Response by poster: Thanks! I was hoping to get some good ideas from the smart and talented MeFiers before sending him to google or craigslist. Sorry about the vague ideas of my friend. I was guessing all he needs would get somewhat expensive. Any other suggestions, let me know.
posted by freakazoid at 4:31 PM on January 11, 2011
posted by freakazoid at 4:31 PM on January 11, 2011
What others have said so far is decent, as someone who has been on both sides of large web development projects here are a few things I think are critical:
Your friend should come up with a budget and a good two-three paragraph description of what he/she wants. That is not to mean they should talk about what content management system they want to use or what sort of payment system they want set up, but general information about what the goal of the project is. The best design houses often start out with a potential client questionnaire, (check out the project planner questionnaire from Happy Cog). At the very least, your friend needs to be able to answer these questions. If they can't, they're probably not that serious about it.
Shop around, do a request for proposals on sites like jobs.mefi (ahem), Authentic Jobs (which is specifically for web design), etc. It sounds like your friend is willing to go online and potentially work with a remote development house, this is great because the best design shops are likely not local. But there are likely really good ones around you, ask other people who have also had projects of similar scope and budget. Word of mouth is still one of the best ways to finding someone to work with.
I dislike places such as freelancer and elance because they tend to be filled with developers and programmers that are while not necessarily bad, do not really help you develop your project. While they may work fine when you just give them a list of deliverables, the best developers work with you to figure out not what you want, but what you need. Since your friend is likely going to be spending tens or hundreds of thousands of dollars on this project, she/he really doesn't want to hire a yes-man and have the whole project flop 18 months later because nobody uses it because it sucks. A good web developer will ask help your friend work through his/her project requirements as the project starts, keep asking questions, and help shape and reshape the project all the way through.
Do not automatically go with the lowest bidder. Seriously, look at their proposal, does it seam reasonable? Projects of any size usually take six to nine months, the larger the project, likely the longer it'll take to do well. If the developer is nearby, have a meeting or two. If they are remote, demand a phone/web conference. Actually, your friend shouldn't have to demand one. If it's a good development house, likely they will ask for one before they put together a full proposal.
A good development house does not just go over a list of deliverables, I cannot stress this enough. I've had way too many clients and have been hired at the end of too many projects where the client had a list of deliverables that were not well thought out, the developer, while perfectly competent, did not care about the viability of the project as it was and just delivered everything that was on the checklist without looking at the bigger picture. These projects almost always flounder for 6, 12, 18 months with minor fixes, and basically fail. Every time, the client was unable or unwilling to spend resources on taking care of the overall project, instead dwelling on this deliverable or that. The best way to keep from doing this is hire a good developer, and listen to what they're saying. When something is untenable or unviable, the best developers will tell you no.
posted by thebestsophist at 9:05 PM on January 11, 2011 [2 favorites]
Your friend should come up with a budget and a good two-three paragraph description of what he/she wants. That is not to mean they should talk about what content management system they want to use or what sort of payment system they want set up, but general information about what the goal of the project is. The best design houses often start out with a potential client questionnaire, (check out the project planner questionnaire from Happy Cog). At the very least, your friend needs to be able to answer these questions. If they can't, they're probably not that serious about it.
Shop around, do a request for proposals on sites like jobs.mefi (ahem), Authentic Jobs (which is specifically for web design), etc. It sounds like your friend is willing to go online and potentially work with a remote development house, this is great because the best design shops are likely not local. But there are likely really good ones around you, ask other people who have also had projects of similar scope and budget. Word of mouth is still one of the best ways to finding someone to work with.
I dislike places such as freelancer and elance because they tend to be filled with developers and programmers that are while not necessarily bad, do not really help you develop your project. While they may work fine when you just give them a list of deliverables, the best developers work with you to figure out not what you want, but what you need. Since your friend is likely going to be spending tens or hundreds of thousands of dollars on this project, she/he really doesn't want to hire a yes-man and have the whole project flop 18 months later because nobody uses it because it sucks. A good web developer will ask help your friend work through his/her project requirements as the project starts, keep asking questions, and help shape and reshape the project all the way through.
Do not automatically go with the lowest bidder. Seriously, look at their proposal, does it seam reasonable? Projects of any size usually take six to nine months, the larger the project, likely the longer it'll take to do well. If the developer is nearby, have a meeting or two. If they are remote, demand a phone/web conference. Actually, your friend shouldn't have to demand one. If it's a good development house, likely they will ask for one before they put together a full proposal.
A good development house does not just go over a list of deliverables, I cannot stress this enough. I've had way too many clients and have been hired at the end of too many projects where the client had a list of deliverables that were not well thought out, the developer, while perfectly competent, did not care about the viability of the project as it was and just delivered everything that was on the checklist without looking at the bigger picture. These projects almost always flounder for 6, 12, 18 months with minor fixes, and basically fail. Every time, the client was unable or unwilling to spend resources on taking care of the overall project, instead dwelling on this deliverable or that. The best way to keep from doing this is hire a good developer, and listen to what they're saying. When something is untenable or unviable, the best developers will tell you no.
posted by thebestsophist at 9:05 PM on January 11, 2011 [2 favorites]
I run a web development company that builds custom sites for startups and businesses. (Link in profile, if you're so inclined.) thebestsophist is spot-on with a lot of his advice:
You're looking (or your friend, such as the case may be) for a relationship, not just an individual or organization looking to get you in and out of the door as quickly as possible. This isn't a euphemism for someone to spend more money with—on the contrary, we frequently talk clients out of spending money on features or components we think will struggle to offer real value.
Talk to a lot of shops. Good shops shouldn't charge you for a consultation, but you're going to need to have a reasonable budget in mind to work with them. Custom software projects of any complexity are typically not four-figure projects, but much closer to five- or six-figures, and rightfully so. They are in depth, soul-searching endeavors that make use of great developers, a solid project manager, and proper QA and testing.
More importantly, the good shops will push back. Right out of the gate, any one you speak with should be asking questions, developing the conversation, and you should feel that they're trying to get a solid grasp of what you're trying to accomplish. They should be suggesting additional features and approaches, bringing up potential stumbling blocks and cost considerations, and discussing competition and monetization strategies for you. Within the first two conversations.
You can always try the elance/rentacoder-type approach, but my experience has been that at the very best, you'll end up with an outsourced solution that is done exactly to whatever specifications you're able to put together. And that's not what you need. Since you're not web experts, putting together a coherent spec document will be extremely hard. And outsourced solutions or freelancers rarely apply critical thinking to the overall concept and execution of the site. You get what you ask for, if you're lucky.
On the flip side, using a freelancer or outsource house can be *dramatically* cheaper. At least in dollar terms. In real terms, it doesn't matter how cheap it is if you have to throw it out. If you're going to go with a freelancer or outsource solution, you need to have everything down to a T. You should try to spec things out as comprehensively as possible, but you shouldn't expect the freelancer to ask the right questions or help you with this process. Some might, but experience can differ wildly, and a single person can easily become overwhelmed with a very large project. For outsourcing, communication issues can be a very real concern, along with a general lack of innovation. My biased opinion, that.
You should definitely be able to answer most of the questions put forth in the Happy Cog questionnaire. You should have the initial makings of a business plan in hand and have some thought put forth in how you'd like the site to function, what's special about it, and how you plan on attracting users and making money. A good shop will help you flesh these themes out into more specific strategies, and shops like mine will push you towards what we call the Minimum Viable Product: essentially the cross-section of total number of features and complexity necessary to determine how viable your concept will actually be.
If you spend tons of time and money on a feature that no one will ever use, it's money wasted. The MVP approach attempts to identify what you can get out the door and still use to ascertain a realistic gauge on how viable that feature or entire concept is. A good shop will lean back on their experience and work with you to identify the real priorities. They'll suggest things you never thought of, help you make the hard decisions, and work to build something both you and they can be proud of, and they'll take a very real interest in your success. My favorite clients are the successful ones who keep coming back to me for more meaningful features.
Not all shops will bid things out the same. Some use the traditional, "waterfall" development process. Waterfall calls for a linear approach to a project, beginning with a specification, moving to design, construction, implementation, testing, and finally maintenance. These are typically bid out as fixed-bid projects, but there can be serious problems when working on a startup under the waterfall model. Startups change tack, alter their strategy, or will suddenly consider a new feature to be of the utmost importance. Waterfall, by its nature, is stubborn to these sorts of changes, and they tend to increase stress on both sides of the table.
Some shops, like mine, lean more towards the agile development process. Agile is far more incremental: you typically break up a project into goals or "user stories" which define all that you want each user type to be able to accomplish. These stories are typically prioritized into sprints, which are development sessions of anywhere from two weeks to a month, sometimes more, depending on the shop and the team. During that time, your team will work tirelessly on the items slated for that sprint. Any changes are typically held until the sprint is complete, at which point the next sprint is reprioritized based on the results of the last sprint.
This sort of iteration allows companies to inject critical changes without seriously impacting the productivity of the development team. If you allow a team to get 75% into a feature and then insist that it is no longer necessary before moving them to another feature, it doesn't take long for you to have 3/4s of a broken project. Agile insists on completing the work in front of you, but reevaluating often enough to insert the changes that are necessary at a time that works to reduce impact and additional costs.
This makes bidding a project accurately from the onset a bit difficult, but waterfall bids are typically either horribly inaccurate or so incredibly padded that they come nothing close to resembling reality in any event. Agile is typically priced out on a per-sprint basis, with a rough estimate provided to completing the bulk of what's being asked for, in terms of sprints. This number is subject to, and WILL, change, sometimes up and sometimes down, depending on the direction the project takes. This sort of feedback brings the client much closer to the process and lets them evaluate the nature of their requests and see where their money is going.
Overall, there are a lot of shops out there that can help you. They definitely don't need to be local: with Skype, phones, and screen sharing, we manage several projects for clients all over the country just fine. Some, we've never even met in person. We also have local clients in to our office, because sometimes, it's nice to see that a company physically exists, has an office, and to meet the people behind your project.
Talk a lot, get a good feel for the company that's within your budget, and focus on building a rapport. You should walk away feeling like that shop has great ideas they want to contribute and know how to get it done. You should feel that they're in it for the long haul with you, and that they're committed to delivering a great product. And you should have a list of questions and points to think on from your discussions with them, because they should be asking you challenging questions that you hadn't even considered, right out of the gate. That's the mark of a good shop.
posted by disillusioned at 9:53 PM on January 11, 2011
You're looking (or your friend, such as the case may be) for a relationship, not just an individual or organization looking to get you in and out of the door as quickly as possible. This isn't a euphemism for someone to spend more money with—on the contrary, we frequently talk clients out of spending money on features or components we think will struggle to offer real value.
Talk to a lot of shops. Good shops shouldn't charge you for a consultation, but you're going to need to have a reasonable budget in mind to work with them. Custom software projects of any complexity are typically not four-figure projects, but much closer to five- or six-figures, and rightfully so. They are in depth, soul-searching endeavors that make use of great developers, a solid project manager, and proper QA and testing.
More importantly, the good shops will push back. Right out of the gate, any one you speak with should be asking questions, developing the conversation, and you should feel that they're trying to get a solid grasp of what you're trying to accomplish. They should be suggesting additional features and approaches, bringing up potential stumbling blocks and cost considerations, and discussing competition and monetization strategies for you. Within the first two conversations.
You can always try the elance/rentacoder-type approach, but my experience has been that at the very best, you'll end up with an outsourced solution that is done exactly to whatever specifications you're able to put together. And that's not what you need. Since you're not web experts, putting together a coherent spec document will be extremely hard. And outsourced solutions or freelancers rarely apply critical thinking to the overall concept and execution of the site. You get what you ask for, if you're lucky.
On the flip side, using a freelancer or outsource house can be *dramatically* cheaper. At least in dollar terms. In real terms, it doesn't matter how cheap it is if you have to throw it out. If you're going to go with a freelancer or outsource solution, you need to have everything down to a T. You should try to spec things out as comprehensively as possible, but you shouldn't expect the freelancer to ask the right questions or help you with this process. Some might, but experience can differ wildly, and a single person can easily become overwhelmed with a very large project. For outsourcing, communication issues can be a very real concern, along with a general lack of innovation. My biased opinion, that.
You should definitely be able to answer most of the questions put forth in the Happy Cog questionnaire. You should have the initial makings of a business plan in hand and have some thought put forth in how you'd like the site to function, what's special about it, and how you plan on attracting users and making money. A good shop will help you flesh these themes out into more specific strategies, and shops like mine will push you towards what we call the Minimum Viable Product: essentially the cross-section of total number of features and complexity necessary to determine how viable your concept will actually be.
If you spend tons of time and money on a feature that no one will ever use, it's money wasted. The MVP approach attempts to identify what you can get out the door and still use to ascertain a realistic gauge on how viable that feature or entire concept is. A good shop will lean back on their experience and work with you to identify the real priorities. They'll suggest things you never thought of, help you make the hard decisions, and work to build something both you and they can be proud of, and they'll take a very real interest in your success. My favorite clients are the successful ones who keep coming back to me for more meaningful features.
Not all shops will bid things out the same. Some use the traditional, "waterfall" development process. Waterfall calls for a linear approach to a project, beginning with a specification, moving to design, construction, implementation, testing, and finally maintenance. These are typically bid out as fixed-bid projects, but there can be serious problems when working on a startup under the waterfall model. Startups change tack, alter their strategy, or will suddenly consider a new feature to be of the utmost importance. Waterfall, by its nature, is stubborn to these sorts of changes, and they tend to increase stress on both sides of the table.
Some shops, like mine, lean more towards the agile development process. Agile is far more incremental: you typically break up a project into goals or "user stories" which define all that you want each user type to be able to accomplish. These stories are typically prioritized into sprints, which are development sessions of anywhere from two weeks to a month, sometimes more, depending on the shop and the team. During that time, your team will work tirelessly on the items slated for that sprint. Any changes are typically held until the sprint is complete, at which point the next sprint is reprioritized based on the results of the last sprint.
This sort of iteration allows companies to inject critical changes without seriously impacting the productivity of the development team. If you allow a team to get 75% into a feature and then insist that it is no longer necessary before moving them to another feature, it doesn't take long for you to have 3/4s of a broken project. Agile insists on completing the work in front of you, but reevaluating often enough to insert the changes that are necessary at a time that works to reduce impact and additional costs.
This makes bidding a project accurately from the onset a bit difficult, but waterfall bids are typically either horribly inaccurate or so incredibly padded that they come nothing close to resembling reality in any event. Agile is typically priced out on a per-sprint basis, with a rough estimate provided to completing the bulk of what's being asked for, in terms of sprints. This number is subject to, and WILL, change, sometimes up and sometimes down, depending on the direction the project takes. This sort of feedback brings the client much closer to the process and lets them evaluate the nature of their requests and see where their money is going.
Overall, there are a lot of shops out there that can help you. They definitely don't need to be local: with Skype, phones, and screen sharing, we manage several projects for clients all over the country just fine. Some, we've never even met in person. We also have local clients in to our office, because sometimes, it's nice to see that a company physically exists, has an office, and to meet the people behind your project.
Talk a lot, get a good feel for the company that's within your budget, and focus on building a rapport. You should walk away feeling like that shop has great ideas they want to contribute and know how to get it done. You should feel that they're in it for the long haul with you, and that they're committed to delivering a great product. And you should have a list of questions and points to think on from your discussions with them, because they should be asking you challenging questions that you hadn't even considered, right out of the gate. That's the mark of a good shop.
posted by disillusioned at 9:53 PM on January 11, 2011
You can always try the elance/rentacoder-type approach, but my experience has been that at the very best, you'll end up with an outsourced solution that is done exactly to whatever specifications you're able to put together.
As a part-time rentacoder, I promise you that it is a terrible idea to go this route for a project of this scale. Seriously, half the gigs I pick up on guru involve patching up cut-rate work done by the lowest bidder from these sites.
Find a few small/medium sized houses, preferably local, and shop your project around.
posted by thsmchnekllsfascists at 7:53 AM on January 12, 2011
As a part-time rentacoder, I promise you that it is a terrible idea to go this route for a project of this scale. Seriously, half the gigs I pick up on guru involve patching up cut-rate work done by the lowest bidder from these sites.
Find a few small/medium sized houses, preferably local, and shop your project around.
posted by thsmchnekllsfascists at 7:53 AM on January 12, 2011
« Older Help make my wife's birthday in Paris awesome? | Is there a cheap and easy fix for our Honda door... Newer »
This thread is closed to new comments.
Good luck!
posted by pakoothefakoo at 1:43 PM on January 11, 2011