10 things to consider when budgeting bespoke software

In my dark past, I’ve served as a Business Development Manager and looking back at those 8 years of selling digital services and IT provisioning, I’ve decided to write about real-life situations rather than clutter you with yet another SEO-based lead generation blog. Please take these lines with a grain of salt; since all of those future posts are based on my own “meandering” experience.

Since money “makes the world go around” let’s I’ll focus on them — specifically on how much you’d need to kick-off your first app . Please note that I’ll express the cost here in engineering hours, since time-vs-material rates vary (from 35 to 50 EUR depending on the applications complexity, teams’ size and experience). For a fair assumption, multiply those to a rough medium of 35 EUR/hour and you can easily deduct the cost assumption of your app.

1. Processing your idea through a legal checklist can really save you a buck

(!) No one is above the law — so read-up on your obligations

Let’s assume you’ve narrowed down your idea and: you know what to expect from your digital service, why you’re doing it and who is your target audience. Before you do anything else, I’d strongly advise checking if you’re compliant to legal, financial and distribution rules of your market.

While this might seem a bit overkill — you’d be amazed to find out how many startups fail only because this check hasn’t been processed. At the end of the day, fines become money you risk sending down the drain, especially when regulations take you by surprise.

For instance, let’s consider you’re delivering a retail-oriented service that picks up within the EU market. Even If a few of those GDPR requirements are missed, you still risk an effective, proportionate and dissuasive administrative fine of up to 4% of annual global turnover or €20 million (whichever is greater) for non-compliance — money that can be well spent on marketing your product/service or RnD. Rules variate from market to market, so before you even get to prototyping, make sure all those legal checkboxes are ticked.

2. Prototype before you build

(!)A prototype will save you a lot of money

The clearer your idea is — the better your budget estimate. No one can guarantee you a fixed budget. To get a realistic cost edge, you need a well-described idea. At this stage you don’t need advanced tools: that brain of yours knows better, so don’t be afraid to use it. A simple yellow-legal pad and a sharpened pencil should suffice.

Start with the service/product you want to deliver, envision it and build a step-by-step instruction for your end-user. Once you’re done, think of how your operations team would use it and how could your app ease their day-to-day process: build those instructions too. Specify what third-party services you’d like to integrate (CRM, ERP, Social integrations or syncs) and check if those provide an API — it usually takes a google search.

When you’re done with these, draw that interface. Simple boxes and “this button does this, that button does that” is a good start and can sometimes be enough to design a Minimum Viable Product without additional costs. It can take one or two legal pads to complete all of the above, but in the end, you’ll get a neat set of requirements. In addition to reducing your budget estimates, specific requirements will also keep you away from snake-oil UX unicorns that can upsell you their vision instead of implementing yours.

3. Decide whether you want a full-blown app or a Minimum Viable Product (MVP)

(!) Don’t throw away the donut!

If this is your first time (building an app) you might not want to exchange your kidneys for a full-blown app. These usually cost around 80 to 100 thousand Euro (depending on complexity) and if you’re not sure 2000% that app will make actual money in the next 12 months, a Minimal Viable Product might actually be a better fit — “think big, start small and scale fast”.

A well-built, though generic MVP might cost you up to 30 000 EUR per platform (per iOS app or Web app or Android App), it’s easy on your wallet and you still get to test those waters — without drowning.

In addition to the above, an MVP would deliver you the most constructive feedback you’ll ever get, a chance to attract investors and early adopters. Not to mention that your full-feature product will benefit of User Experience inputs that come from genuine customers — not focus groups.

This is a proven matter: remember Facebook? It actually started as “FaceMash” — a simple website that used photos compiled from the online facebooks of nine Houses, placing two next to each other at a time and asking users to choose the “hotter” person. That MVP was enough to validate the idea and led to developing a 35,409 billion Euro company.

4. Your User Experience can cost you more than you thought

(!) Budgeting your first app: UX will not be as cheap as you expect

Remember those sketches drawn on that yellow legal pad? If you have them ready to go, any respectable software house can design that interface without the requirement of an expensive UX research in about 100 hours (depending on the number of screens you have). At this stage, you can decide together with your service provider on nice-to-haves and core functionalities. For an MVP, the UX and the application’s design would cost your around 3 000 Euro (about 20% of your total budget per platform).

If you’ve decided to go with a full-featured app, you might as well invest in a separate UX milestone, where you’ll get a clear map for your user’s actions (on mobile or web) and a feedback-acquiring prototype, designed to be tested across potential audiences. These efforts alone can take up to 400 hours, depending on how many platforms you’re building across, how wide is your trial audience and how complex your usability is. After receiving that UX feedback, applying changes and designing the final User Interface would take another 6 to 8 hundred hours. However, if you’re building on top of a previous MVP — you can save up to 600 hours on building your UX since your active users will definitely let you know what’s wrong with your app and which improvements you can perform (no need for focus groups here).

5. The application’s front-end can reach up to twice of your UX budget

(!) UI/UX won’t be cheap

Set your expectations straight: implementing the User Interface design could take twice as much. Here, you should account for animations, charts, UX elements and other eye-candy thingies designed for your end-user. Almost no one can deliver a fixed figure here.

The best you’ll get is an estimate with a 20–30% variation. Assuming we’re talking about an application that relies on 20 mobile views and 40 web views, you’re looking at a gap of 700 to 900 hours (depending on those front-end effects you’re looking to embed).

6. The most finance-hungry part of your app is a reliable API and a well-designed backend

(!) Keep an EYE on your API!

If you have a mobile platform, you’ll most definitely need an Application Programming Interface. The API is that part of the system that acts as a waiter in displaying the right info/options and/or operations to your end-user and controls some of your business logic.

Nowadays any mobile app needs at least a dashboard powered by an API, that enables you to control the inner workings of your product, see usage statistics or enable services for your target audience. Here, we’re stepping into quite a grey area since developing an API is strictly related to your product’s definition.

The same can be said about designing the database schema and your application’s infrastructure. To get a ballpark on how much your backend would cost, make use of this rule of thumb: take the number of hours allocated to mark-up and multiply it by 3. In our hypothetical case, for that 20 / 40 view multi-platform app, you’re looking at 1 400–1 800 engineering hours.

7. Understand how estimates are made — this will help you filter contractors based on pricing and quality

Even if your requirements are nailed down to the tiniest detail, the UI final and deemed as complete and all those business logic integrations are clear, there’s still a 20% error margin for any project estimates.

Considering the input delivered to a developer, System Architect or Project Manager, there are a set of system-related variations that contribute to budget fluctuations. To pin those down, industry professionals make use of “the cone of uncertainty” — an estimation tool designed to determine error margins across product development stages.

(!)Know THY Cone!

For instance, an API-related task might be more time consuming due to conflicts with your business logic, or because you’ve missed specifying a filter, or there are some logic-related bottlenecks that have not been foreseen prior to development. All of these generate uncertainty during development phases and an accurate estimate must include that variability phenomenon. At least that’s how professional developers asses their tasks.

Once those sources of uncertainty are further investigated and pinned down, budget fluctuations can diminish or increase, depending on decisions and needs deemed as important by your system’s business logic.

Cost-wise, an estimation based on the above model places almost any bespoke build between 40 to 160 thousand Euro. If you don’t have a dedicated team and you’re looking to hire one, pick quotes that place their budgets within that range and beware of low prices.

Mature, highly efficient digital products cannot be built within 700 hours — you need at least 1 400 to 1 800 hours to build an OK digital product. Vendors that quote you low are in most cases beginners, will use unsustainable technology, or resort to re-usable components aka third-party code-bases that (in most of the cases) will cost you more in maintenance hours and limit your service scalability.

Providers that deliver full-feature product quotations between 1 400–2 000 hours are less likely to fail since they’ve already planned for contingencies and know what they are up to. These guys will usually deliver fixed prices between 100 to 120 thousand Euro or estimates between 90 to 140 thousand Euros. Again — keep in mind: I’m referring to that 20 / 40 view multi-platform app (prices might vary depending on your application’s scope).

(!) Ask about maintenance upfront!

Make sure to ask each vendor about their maintenance plan: if your development estimates are lower than their 1 year of maintenance plan or if you get a “pay as you go” proposal — run forest, run!

A well-designed product will not cost you a development budget during its first year in production (unless we’re talking evolutive maintenance, with implies releasing new features).

Usually, a 4 000 hours project cannot exceed 1 200 hours of annual maintenance (which is 30% of the cost). Reasonable maintenance plans reach about 20% of your application’s development cost per annum.

8. Managing uncertainties is not a gimmick — it’s a life-saving procedure to make your budget risk-proof

By now, your budget could look like this:

While it’s quite easy to estimate an MVP (due to a minimum set of features and low complexity) estimating a full-featured product is a totally different story.

In our case, I’ve considered a scenario where the mobile version would take about 30% of the app’s cost and the other 70% are allocated to API, front-end and backend (by rule of thumb). All those figures might seem high, but keep in mind, we’re talking about a vaguely specified project. The most accurate figures here will still include a +/-50% error margin.

When you build a digital product it’s better to plan for more rather than end-up with no money and a half-designed app. Plan for the highest estimative figure, agree for a medium fixed-amount (for our sample app, that would be around 120 000 EUR) and if you’re short on money, acquire some investments before you build. Chances are you end-up spending less, pushing those remaining resources to a rainy-day fund or allocating them for eventual maintenance and/or development.

9. It will never be fast, cheap and good (you must pick two)

When building a full-blown product you can’t avoid that “Iron Triangle” everyone is talking about. This one states the following:

• Developing something quickly and of high quality is quite expensive. It simply implies more expert resources (not necessarily more development hours). Too many cooks in the kitchen will definitely spoil the broth — so we’re talking expertise here not manpower.
• Developing something quickly and cheaply will result in a five-wheeled vehicle — with square wheels. Quality needs time and expertise — cut them both and you end up with a broken piece of software that will most definitely have to be rewritten.
• Developing something of high quality and low cost will take a ridiculous amount of time. A junior developer will deliver, but it will take him thrice the time compared to even a middle developer. His/Her trial and error process might cost you more than you’re expecting.

Remember, this applies to mature products. MVPs are easier to manage and since they are focused on main functionality, you might end up getting something cheap, in a timely manner that resembles quality for its purpose.

10. Launching your product will definitely cost you some money

Once you’re done building your application development budget estimate, don’t forget to add an additional line before you close totals. Placing an app in production does also imply some costs, especially when we’re talking multi-platform digital products.

Deploying your iOS product will cost you about 100 EUR per annum, your Android app will take a 25 EUR one-time fee and hosting your API and web product in the cloud can vary. Depending on your application’s scope, architecture and minimum hosting requirements you’re looking at a bill that can cost you from 3 000 EUR to 10 000 EUR per annum.

Bottom line — How much will an app cost you?

As stated, it really depends on the path you are taking. If you’re building an MVP, expect a bill around 15 000 to 30 000 EUR per app (plus hosting and publishing costs). If you’re looking to build a full-featured digital service, in most of the cases, you’ll need at least 80 000 EUR to get started and cover your contingency costs.

Again, these are industry standards validated by those 8 years I’ve spent as a Business Development Executive. Of course, I’d love to hear your thoughts on this (who knows — maybe you have a different take on those figures) so feel free to share your own experience.

We’re also here to listen to any software idea you might have. If you’re looking for a reasonable ballpark figure or simply our opinion on it, reach our experts here for an exploratory call and a personalized quote. To keep up with our weekly posts follow us on Twitter, Linkedin or Facebook.