Scope of building a site similar to a daily deals site

Cianos

Hey,
I'm looking at commissioning a project soon that will be somewhat similar in infrastructure to a daily deals site.

I'm wondering if anyone here has any experience working on one, or, what the general scope of the project would be?

Thanks

Seth Brundle

The scope of the project will be dictated in part by your available budget and presumably will involve the following areas:
* database development
* web development
* web design (look & feel)
* security (are you taking online payments?)
* testing
* search engine optimisation
* project management
* details of items not in scope
* details of dealing with any changes to scope

Will a functional spec document be required to list of more?

You will also need to detail a list of deliverables (and when these should be delivered).

I'm taking it for granted that this is a development project and the company you choose will not be involved in the marketing & promotion of the site.

However the scope of the project should be quite detailed (to protect both you and the developer) and anyone providing you with more details on it than just "similar in infrastructure to a daily deals site".

Cianos

Thanks for the reply. Yes it' just a development project and the company won't be responsible for marketing.

What is really just required is the build itself and done to a good, scalable standard. This is our main concern, as the technology behind deals sites is quite basic (and our project won't be far off in terms of spec), but we want something that can scale up with minimal hassle.

There are even Wordpress themes out there for deals sites, but I presume they scale terribly. Our database needs to run easily should the site take off and there's a quick influx of members etc.

So I suppose I'm wondering what I should be asking of developers for a build like this, so as I'm not misled by empty promises.

Seth Brundle

Ask them for the technologies they are planning on using (ASP.net/PHP/other, MySQL/SQL Server/other) along with hosting costs, domain registration & costs, etc. (all domains must be in your name for both the registrant and administrative contact).

You should decide beforehand on what functionality you want from the system (i.e. general page navigation, what users can do, what admins can do, how to add a new daily deal, how to remove daily deals, etc.).
How will someone get from the main screen to see an offer's details?
Maybe even sketch some potential screen layouts for the developer.

Ask yourself, why are you doing this? There are loads of established sites out there (pigsback.com, etc.) so yours will have to be extra special to beat these. What is your unique selling point?

Knowing what you want will allow the developer to plan something around your needs and expectations.

Also ask them about testing plans, bug fixing, post-sign off support, etc. Don't pay for a website only for the developer to walk away straight after they think its done. If they are providing a support contract, what is the SLA response time in terms of when they will react and deal with a support issue? What is covered under their support agreement? What is you don't use support - can the money spent be offset against further development work?

Also look for previous work in this area.

Its hard for me to write exactly what you should ask but you first need to know what you want. Ask them can they do it. Then ask them how long it would take and what will the overall cost be (broken down into development, hosting, etc.) and in what way will the project be paid for (don't pay more than say 30 to 50% up front (depending on the amount involved)).

Lastly, you do, to a huge extent, get what you pay for! A daily deals site that is easy to maintain will cost a decent sum.

Hope this helps.

Feathers

Just a few ramblings off the top of my head. (Sorry if it's a bit scattered, wrecked tired tonight.):

Scalability-wise, you'll need to look into how you want to host this & design the architecture of the site to allow caching were possible to deal with load.

People talk of scaling up & scaling out. Scaling up means that you increase the specs of your server to better deal with the load. Scaling out is running your site across multiple servers.

What happens if one of your servers goes down? Does this customer lose his data, but the rest are OK? Or are user sessions shared across multiple machines. How manual or automated is restarting servers if they crash — is there something backing the site to detect failures & re-route requests.

Are they going to have a 'cloud' solution to deal with load balancing? (Buzz terminology, but in a nutshell, hosting like Amazon where you use more computing power during your peak times & only pay for the bandwidth/space for that time.) — If going for a cloud solution, where will the servers be based? Will this have any implications for EU/Irish law re data storage (esp. if taking payment details).

Cianos

@Kbannon, thanks very much, a lot of helpful advice there. We're going to write out a detailed brief to present to developers and try cover as much as possible in it. We're taking care of the design work, so that'll be one big thing out of the way. Regarding the UX, we'll be figuring out a lot of that too. We've already done a few usability tests and the design has evolved from people not really getting it at all, to getting it almost straight away (a few more usability tests are needed to make this more certain, but so far so good).

When you say a 'decent sum', what kind of figures are you talking about? 5k-10k? We don't have that kind of budget, so one thing we were thinking of doing was building a small scale Beta, and if it goes well then ramping it up so that it's ready to take on a larger audience.

Thanks again

@Feathers, Yes these are the more technical things we've no knowledge of at this point. We're hoping to work with a developer who has built large databases before and can advise us on the best approach for how they should be structured, hosted and so on.

Again, we're hoping for an approach that will allow for a low cost set up, but the ability to scale up the operation if the site builds traction. I don't know how realistic this is, maybe it's a case of it having to be built a particular way whether there are 10 users or 10,000.

Smoggy

The scalability will come from 2 main areas :

How well is the site written (i.e can it span multiple servers)
How much money do you have to pay for hosting (hardware)

I'm afarid the costs of creating a website, whether for a small amount of users or a large amount of users is pretty much the same, as the gains of scalability comes down to ramping up the hardware.

Savings could be made on hosting, such as using MySQL / PHP etc. If you want to start small and at a low cost, maybe look into wordpress and if it takes off pay the money to get the customised solution.

As for development costs, I can't see any software / web company taking this on for less than 10k. However you might find a developer who will work on it in their sparetime for < 5k.

Seth Brundle

@Cianos - have a look online for some kind of free or cheap reliable/reputable website framework that does this kind if thing and that your developer can modify. It could save a fair bit.

With regard to cost, its hard for us to gauge a rough price as we don't know the full requirements. However when getting prices try and opt for a fixed cost from the developer. Also make sure your contract covers you properly in the event of difficulties during the project

jmcc

Smoggy wrote: »

The scalability will come from 2 main areas :

How well is the site written (i.e can it span multiple servers)
How much money do you have to pay for hosting (hardware)

There's a bit more to scalability than that. Much of it has to do with the design of the site and the backend. There are two basic (and other more complex ones) models in scalability:

Scale Up: Bigger servers, (more RAM, more RAIDed harddrives).
Scale Out: More servers with the site being distributed across these servers.

I'm afarid the costs of creating a website, whether for a small amount of users or a large amount of users is pretty much the same, as the gains of scalability comes down to ramping up the hardware.

Some scalability in the website design has to be done at the design stage as elements that could be cached or delivered by a content delivery network would be designed at that stage.

Savings could be made on hosting, such as using MySQL / PHP etc.

Getting people who know about database architectures and with the ability to design high traffic databases and systems is going to be expensive. It doesn't matter if it is MySQL or Oracle or Microsoft Server.

As for development costs, I can't see any software / web company taking this on for less than 10k. However you might find a developer who will work on it in their sparetime for < 5k.

If you are dealing with a high traffic/heavy database backed site, you don't need a web developer who thinks that he or she knows about database work. You need people who really know how to put such a site together and that expertise alone may be over 10K. Of course there will always be developers who will have a go for 5K or so. Though without the full spec, this is all just tyre kicking.

Regards...jmcc

jmcc

Just a quick book recommendation:
Building Scalable Websites by Cal Henderson (O'REILLY)

Regards...jmcc

Cianos

Thanks again for all the input on this, a lot of helpful info. Although it wasn't what we were hoping, it's good to know what the estimated cost of building something like this is.

Probably unsurprisingly, we don't have 20k to put in to it, so we're looking at rolling the site out with a stripped down Beta and working from there.

(apparently Groupon used PDFs to send out deals in the early days!)

If the concept proves itself, we might be able to pick up some funding to build it as it should be.

I thought I'd ask if anyone had a few ideas on what we could use to set up the service, maybe there are 3rd party sites/apps we could utilise to make dealing with a low volume of users very easy.

For a basic release we don't necessarily need user registration, people can opt in with their email addresses, be contacted via Mail Chimp for example, and payments can happen directly between the customer and the business using a voucher code.

Thanks again

[CrimsonGhost]

I've heard (not read it myself) that this book is quite a good read for someone in your position looking to engage with a developer.

“Avoiding a Goat Rodeo” will save you time and money the next time you hire a developer to build you a website. More importantly though, it will teach you how to describe what you want to your developer/designer in a way that they can understand so that you get the web site you want, not what they thought you wanted.

http://www.amazon.com/gp/product/B005PTWS4M/ref=as_li_ss_tl?ie=UTF8&tag=postcarfrommy-20&linkCode=as2&camp=217145&creative=399373&creativeASIN=B005PTWS4M