It is the question on nearly every call and it is a fair one. Someone in the Philadelphia area has an idea that’s worth building or a process they know could run a lot better and before going any further, they want a sense of what it takes to get there.
Most agency websites answer with “every project’s different.” That is true enough. It gives you nothing to plan around. So here are the numbers we actually quote for custom software development in Philadelphia and the surrounding area, and what moves them.
What we usually quote
A business website generally starts in the five figures. An eCommerce store starts in the five figures. Custom software is the one to pin down because the term covers everything from a small scheduling tool to a system that runs your whole operation.
Take two jobs we did recently. One was a scheduling tool for a contractor with a dozen crew. The other was a quoting system that had to pull pricing out of the clientโs ERP. Both are custom software. One took a week. The other took months. Same category, nothing like the bill.
So the ranges are a starting point. What follows is the stuff that decides where in the range you land.
Where the money actually goes
People assume they are paying for development time. They are. It is less than half the bill.
A tenth to a seventh of a project goes on working out what you need in the first place. That means sitting with the people who will use the custom software and watching what they actually do, which is never quite what the process document says they do. Half to sixty percent is design and build. Then a fifth or so goes on testing and launch and the remainder on training your team and handing it over.
That testing chunk is the one that gets cut when a quote comes in suspiciously low. It is also the one you notice missing about a week after goโlive.
What pushes the price up
How other systems it has to talk to. This is the one. Software that sits on its own is simple enough. Software that pulls from QuickBooks, writes to your ERP and syncs a supplier feed is three things that can break so all three need building and testing properly. If you only pay attention to one item on this list make it this one.
How many different people use it. One screen for one kind of user is cheap. Once you need a view a staff view, a customer login and something readโonly for your accountant that is four sets of screens and four sets of permissions to get right.
Regulated data. Patient records, card details, anything with a rulebook attached. The features are not harder to build. Proving they are handled correctly is what costs. Around here that catches a lot of people because the region is full of practices and pharma suppliers.
Replacing something that already exists. This one catches people out. Replacing a system usually costs more than building from scratch because the data has to come across cleanly you have to understand all the odd behaviour the old thing had and you can rarely switch one off on Friday and the other on come Monday. You run both for a while. That overlap costs money.
What about apps?
App development in Philadelphia gets quoted over the place and part of the reason is that people mean different things by “app.”
A web app, something your staff or customers open in a browser, is usually the route. There is one version to build. You can fix a problem in the afternoon. Web app development in Philadelphia tends to sit in a bracket to a decent eCommerce build, though it climbs fast once there are several user types or an integration involved.
A phone app is an animal. If it has to run on iPhone and Android, you are paying for two builds unless we use something crossโplatform and either way there is App Store review, device testing and the ongoing business of keeping up with OS updates. It is worth it when you genuinely need the camera, GPS or offline use. It is expensive if you just want something on a phone screen, in which case a web app usually does the job for less.
How long it takes
Websites, one to two weeks. E-commerce: two to four . A first working version of a custom software: three to five.
Many people think we can give answers. In truth, projects are rarely delayed by code. Usually a project waits two weeks for an internal person to confirm how a process should work. If a project can point to a person who can make decisions without a meeting, that will save more time than any technical decision we make.
Fixed price or hourly
We offer both fixed price and hourly options. Which option we choose depends on how clear the brief’s
Fixed price works best when everyone already knows what the project will build. The project knows the cost up front. The downside is that if the project changes its mind partway a change order is needed, because the fixed price was set for a list.
Hourly or monthly billing suits projects that are still shaping or that will keep evolving after launch. The project trades the certainty of a number for the freedom to change direction.
One thing to be wary of: anyone who offers a fixed price on a brief. That person may have padded the price to protect themselves or may return for later. Neither outcome is good.
Things that should make you suspicious
If a quote arrives and nobody asked any questions first the quote is a guess. It may be educated,. It is still a guess.
If the quote omits any mention of testing see the point
If the quote is vague about who owns the code ask for a written answer. The project should own the code the designs and the hosting accounts. The project hands all of that over as standard. Expects any decent provider to do the same.
If the quote stops at “go with no mention of what comes after that is only half a project. The project needs support.
If one quote is dramatically cheaper than the others something has been left out.
Sometimes the lower price comes from a team with lower overhead, which is fine. Often the price is set to win the job and the real cost appears later as extras.
Worth noting: if the project compares a firm with an offshore quote the project is not comparing the same thing. Many good developers work overseas. The project adds a time difference. When something breaks at nine in the morning the project will know how much that matters. Most people who call the project after being let down cite distance, not skill as the cause.
Spending less
Sometimes the project can spend less and any good provider will tell the project.
Occasionally the honest answer is that an offโtheโshelf product already does what the project needs so the project should just buy it. The project has given that advice before. Lost the work, which is fine.
Times the project does not need the whole system pictured, only the part that consumes the projectโs week. Building a version that does that important part well then adding more after a month of use, usually costs less than trying to specify everything before anyone sees it.
Getting a real number
Ranges only provide information. To give the project a figure the project must know what the software must do what the software must connect to and who will use the software.
That is a halfโhour conversation and the project does not charge for it. The project will leave with a scope, a timeline and a ballpark figure, whether the project hires the provider or not. If the project is not a fit the provider will tell the project and point to someone better suited.
The project is a custom software development agency in Marlton, twenty minutes from Center City. Empircal Edge INC. has been building software for businesses around Philadelphia and South Jersey for over twenty years. Empircal Edge INC. is happy to come to the projectโs office or discuss the project over the phone.
Written by: Empirical Edge
