What does an app cost to build, really?
When I wrote about validating your big app idea one of the questions asked was ‘what does an app cost to build’? This is the next logical step and a hard one to answer. The goal of this is to ‘demystify’ the app building process and to give you all an insider look at how we quote apps. There’s always two sides of this conversation — “That’s too expensive” and “We can’t do it cheap”. There’s a huge rift in the middle that this article hopefully educates you on.
“As long as a piece of string”
This is the answer you will get from most agencies, designers and developers when you ask them what it’ll cost to build. It’s a frustrating thing to hear (we are even guilty of it occasionally) so we need to start somewhere: what does this app look like? What are the key features required to launch an MVP (minimum viable product)? What do we need to scope to figure out the app cost?
For the sake of this article, we’re going to look at building a loyalty/rewards program (CC Christina). Christina asked;
Break down a loyalty/ Rewards apps! Don’t forget the maintenance/upgrade side too which many people don’t account for
Initial Scope
We do a couple of things to figure this out when you come to us. We’ll take a look at what others in the space do and take a list of features that you specifically want in your app (we’ll have this from the lean canvas we’ve filled out during the validation phase) in order to figure out the app cost.
This list could go on — we’ll cap it at five common features for now.
We then use an effort/impact matrix to figure out which features make the cut. I’ll write up blog post dedicated to this specific method and why it works. But the tl;dr is that anything that makes the top-left quadrant makes the cut.
Snake, for example, made it in because it has a relatively high impact (entertain your customers while they wait) and low effort as the script is essentially written for us (play snake online). Business self-serve on face value seems like it’s an essential feature. It certainly has a high impact, however, is also very high effort. In the early stages of your app, you will be going café to café signing businesses up personally to further validate your app. So whilst it is an essential feature, we always look to keep your app cost low early on by omitting unnecessary features.
Creating the user stories
So here’s our final feature set in user stories to estimate the app cost;
- As a business, I can set up a profile in order to manage my loyalty program
- As a business, I can change the rewards in order to suit my business model
- As a business, I can see which customers are the most loyal in order to treat them better
- As a customer, I can anonymously collect points in order to get free rewards
- As a customer, I can scan a QR code in order to get points
- As a customer, I can play snake in order to be entertained while I wait for my coffee
There are two types of users here — businesses and customers. These types of apps are called ‘marketplace’ apps and are one of the hardest to pull off successfully as you’re essentially catering to two very different, very needy user groups. You’ve been warned!
Breaking that down
I wish it was as easy as plucking a number out of the air. Alas, it is not. There’s even further work that goes into your figuring out an app cost before we send you the quote.
Internally we do a quick user flow map of what this looks like on a whiteboard. This allows us to get a very raw, birds-eye view of all the components involved in creating your app. It also allows us to see any potential problems that might occur during design, build and development. The first version of this user flow might look something like this;
User flows are always evolving and this provides us with a starting point. From that we can deduct that we need;
* API = Application Programming Interface [?]
Timeline Estimation
Now that we’ve got a bird’s eye view of all the screens we break these tasks down with a tool like Asana. We like Asana because it can integrate with a tool called InstaGantt which we share with our clients. I’ve used Trello in this example so I can link the board publically.
The purpose of breaking these tasks down is to estimate the time each team member needs to spend working on the task. You can check out a live example of this.
In total, we estimate we’d spend the following time designing, building and testing the app;
- Setup & Project Management — 8 hours
- Database & API Setup — 20 hours
- UX Design & Review — 18 hours
- Visual Design & Review — 22 hours
- Build & Integration — 90 hours
- Testing & Launch — 48 hours
- Total: 206 hours
Believe it or not, these are conservative estimates. One of the biggest struggles we have as an agency is balancing a customer’s expectations with what we need to do to keep the lights on. Just the same as customers need to mind their budgets and get the best bang for their buck.
From this estimate, we can expect that it’ll take around 5–6 weeks based on 40 hour weeks. Our designers and developers usually work during the same time, so it could be even shorter. We break all these tasks into digestible, two-week sprints a la agile.
Strict, traditional agile is amazing for product teams but falls apart when clients are involved. We have come up with our own flavour of agile that works in an agency environment to bridge this gap. Internally, we see a board. Externally, clients see a Gantt chart. It works for us!
What about on-going maintenance?
I’m glad you asked — apps are never a set-and-forget thing. There’s always support, bug fixes, maintenance and on-going features. We typically have clients on a retainer post build. These are once again dependant on your needs. We typically start at $1000 per month for general support, consulting and figuring out what comes next as products are always an iterative process.
TELL ME THE TOTAL APP COST ALREADY!
Alright, alright! This part is simple. Based on 206 hours of build, multiplied by our rate of $150 incl. GST the most you’d pay to get a basic, loyalty MVP built (with this feature set) is $30,900AUD and $1000 monthly if you want on-going support.
The obvious question here is — what goes into $150 per hour? Every agency and freelancer has their own hourly rate. Our rate is a calculation of a huge number of factors that include but are not limited to;
- Our wages / paying our maytes (staff)
- Time spent talking to clients
- Software subscriptions we need to complete the work
- Any contractor expenses we have to pay
- Office rent
- Oh, and most importantly, a little bit of profit sprinkled on the top
So apps cost $30k to build?
Yeah, once again — as long as a piece of string. I have estimated this based on the feature set we ‘decided’ with the client through our processes. A real client might want blockchain integration, etc. This will change the price dramatically! If it’s a much simpler app that doesn’t have two sides to it (i.e. non-marketplace app) then it could be dramatically cheaper.
I hope, that after this article, the veil has been lifted a little on what goes into planning, designing and developing an app and what sort of app cost you might be looking at. So next time you get a big quote from an agency — don’t scoff — be sure that plenty of consideration and planning has gone into the price they’ve given you. We, after all, have to keep the lights on too!
What sort of app would you like to see broken down next? Let me know!
Got a great app idea? We’d love to hear about it. Chat with our digital assistant Lexi or get in touch.
Originally published at Mayte.
