Skip to the content

Startup: Building a digital marketplace


It seems like every other day, we get a request asking about how much to build the next Uber, or the next Trade Me, Upwork, AirBNB, or Seek. Although the product/service offerings are wildly different, the underlying concept for each of these technical juggernauts is the same. They are all marketplace solutions i.e. platforms that provide an intermediary service for buyers and sellers.

Some are seller led e.g. Trade Me and AirBNB where the party with the goods places an advertisement and responds to potential buyers. Others are buyer led e.g. Upwork and Seek where the buyer places an advertisement searching for someone to provide a product or service. There are obviously differences in the workflows and how the platforms are all monetised, but as marketplaces, they also have a great deal in common.


Types of marketplaces

GOODS

SERVICES

MEDIA

INVESTMENT

eBay/Trade Me

Uber

Twitch

Kickstarter

Amazon

AirBNB

Soundcloud

Indiegogo

Etsy

Booking.com

YouTube

Patreon

Google Play

Upwork

Yelp

GoFundMe

Apple App Store

Couchsurfing

Spotify

Fundly

 


Community fuelled growth


The opportunity to create a rapidly scaling business, fuelled by the efforts of marketplace participants themselves, rather than your own complex logistics is certainly an attractive feature. Most marketplaces are at least partially self-policing i.e. they rely on the community to make the space safe for all users. However before we get too far ahead ourselves, understanding the concept and purpose of a minimum viable product (MVP), is a critical first step.


Developing a Minimum Viable Product (MVP)


Before you clear out all your savings and hit up friends and family for seed capital, you need to validate your idea, and one of the best ways of doing that is with an MVP. An MVP is the very first version of your product which only has the features necessary to attract early adopters. It will be enough to prove that the idea has legs, but will definitely be leaving your users wanting more.

Building an MVP will help to:

  • Check if your product resonates with your target audience
  • Confirm that you have understood the core functionality your users want
  • Verify which features are most useful and which are not
  • Create an initial base of potential clients and product users
  • Expose any problems that need to be solved before further work is undertaken



Key benefits


Idea validation


Whether it’s passion, ambition or some other motivational force that’s driving your effort to create a marketplace, chances are you will have blind-spots, and to give yourself the best chance of success you need to identify them and mitigate risk as early as you can.

Perhaps you have overestimated the importance of your key unique feature, or have underestimated the importance of something else. Maybe your audience isn’t ready to pay for a particular piece of functionality, or perhaps there’s a cultural issue that prevents users in your target market from engaging fully.

An MVP will help you test your ideas and find the answers to these questions in the early stages of your company’s development. It will allow you to redirect time and effort towards features, services and products that your customers really want.


Focus to protect cash


Most of the larger marketplaces you are familiar with, will have already spent millions of dollars on their technology platforms, and so to attempt to directly match them feature for feature, you are going to need very deep pockets.

Instead, developing an MVP and specifically focusing on a very small target market helps you conserve cash by developing only those features that are necessary for your target audience to fall in love with your product.


Attracting early adopters


Your early adopters are the first people who jump in and use your product. Keep them happy and continue to scale out the solution to meet their ongoing needs and you will build not only a loyal pool of users, but hopefully a bunch of active product evangelists.


Data


After launching your marketplace MVP and getting enough traction, you’re now in a position to approach prospective investors with real data.


Steps to creating a marketplace MVP

An MVP should be designed as an experiment. It's key purpose is to test your ideas, including whether you are able to attract a probably small but happy group of initial users. The steps to building an MVP are no different to any other software product.

  • Discovery
  • Defining the core features
  • UX concept & wireframes
  • MVP Development
  • Capture user feedback
  • Iterate


Step 1: Discovery


Whether you’re entering an existing space, or carving out something entirely new, documenting what you’re doing with some basic market research will help define your strategy, and inform your first set of tactical decisions - including what features need to go into your MVP.


Step 2: Defining the core features


With any marketplace, there will be a bunch of “table-stake” features that are absolutely necessary to build. These include registration & authentication, profile management, product or service listings, two-way communication, and payment methods. On top of these, there may also be features that make your product unique, or tightly tailored to your target market. These are normally written as a set of user stories i.e. requirements expressed from the user’s point of view.

User stories follow a common pattern e.g. As a {User}, I want to {feature}, so that {benefit}. For example, the high-level user story for a user feedback feature might be.

“As a Customer, I want to provide a star rating and give feedback on my experience with a Supplier, so that the community can make more informed decisions when selecting Suppliers.”


Step 3: User Experience (UX) design


Before the coding starts, the UX designer will map out how the application works using a combination of wireframes and user flows. This process helps us make sure the product is easy to use and doesn’t include any usability issues that will frustrate people and potentially stop them from returning.


Step 4: Development


It’s likely that from this point on, the project will be developed using the agile methodology. This method is favoured by most software developers due to its flexibility, and focus on building deliverable software in short “sprints”. A sprint is generally a two-week period where the development team builds, tests, and releases a subset of functionality. Feedback from user testing for each iteration is considered, and if appropriate, will be incorporated into future sprints. Change is embraced rather than feared.

Another important consideration for the development phase is the technology stack chosen. A tech stack consists of programming languages, frameworks, libraries, various development tools and programming approaches themselves. Choosing the right stack is important, although there is not necessarily “one” correct solution.


Step 5: Getting market feedback


An MVP is by nature an experiment. It is designed to test the viability of an idea, the uptake of custom features, as well as general usability of your product. Building feedback loops into the product itself will help you keep track of key user sentiment metrics, and help guide decision making for future iterations.

 

Step 5: Iterate

Rinse and repeat. Take what you've learned from the first iteration and feed that back into the next to start the process all over again, hopefully gaining new users and revenue along the way.

 

What does it cost?

There are a few ways of getting an MVP built. In some cases, you may be able to purchase a ready-made marketplace solution “off the shelf”. This is going to be the cheapest approach, but it won’t be unique, customisation will likely be hard, and maintaining the solution over time may prove a challenge. Importantly, if you want to own the IP to your solution, then this approach will probably be a false start.

Building the solution from scratch has its own challenges, but ultimately it will give you a product that is yours, tailored to your customers, with features that are unique to your value proposition.

The base functionality for a simple goods or services marketplace might look something like this:

 

HOURS

Functionality

BA

UX/UI

Dev

QA

PM

TOTAL

Registration & Authentication, Roles, Permissions, Payment methods

14

21

48

24

11

118

Listings

6

12

32

16

6

72

Proposals/Applications

12

23

52

26

10

123

Search, filter, and organise

16

24

60

30

12

142

Messaging

4

8

48

24

6

90

Payments

8

13

44

22

8

95

System Admin, Reporting, Dashboards

12

23

94

47

15

191

 

  • BA = Business Analysis
  • UX/UI = User Experience/User Interface design
  • Dev = Frontend and Backend development
  • QA = Testing
  • PM = Project Management

Software development contractors in New Zealand earn between $80-100 per hour, so to engage a professional services firm, expect to pay somewhere between $150-220 per hour. At the lower end, the functionality in the above example could therefore be delivered for approximately $125K.


Get in touch

If you’d like to set up an initial consultation to discuss your ideas for a digital marketplace, drop us a line at [email protected] to organise a time.

By Rowan Schaaf
Development