My attempt to build a MVP without knowing how to code

The Design

For our last design challenge, my teammates and I were approached with a question that many organizations face: how can insights and resources from each employee’s experience be passed on to others in order to increase efficiency and avoid making the same mistakes?

Our answer was Project Post Mortem, an internal search and aggregation platform The basic idea is that project managers would create a project pages to document useful information (files, links, structures, personnel, retrospectives, action plans), and distribute post mortem surveys at the conclusion of the project to evaluate success. Each project is then turned into a tagged report that is searchable by other project managers internally for starting new projects.

Project Post Mortem

I wanted to further test the design by building out the core functionalities into a self-sufficient system. My first instinct was to try my hand at creating a minimum viable product (MVP).

Minimum Viable Product

A Minimum Viable Product (MVP) is a product that has the minimum set of features to prove the most essential hypothesis in your business.

If you’re starting with an idea and nothing built, your first goal is to prove that people want what you’re planning to build. A Minimum Viable Product would be what you could build at a minimum to prove that.

— Mikael Cho, Founder/CEO of Unsplash

A MVP is a very important part of the product process, especially if resources are limited. Development can be very time consuming and expensive so being able to build a MVP means you can test and validate the main hypothesis of your product as quickly as possible before spending time and resources to build something that might be completely off the mark.

Process

Setting Goals

With a design in hand, it is tempting to try to include as many features as possible. Thinking back to the overarching goal can help narrow down the list down to just the necessary ones.

For our project, Project Post Mortem, the guiding goal was this: to pass on knowledge, insight, and resources gained from one project to the rest of the organization.

Keeping that in mind, I began to narrow down the features

Deconstructing and prioritizing

Our PM platform had many different features so the first step was to prioritize which ones are absolutely necessary.

First, I listed out all the features in the design. Luckily for me, our design did not have an overwhelming number of features. I then took this list and ranked each feature on a scale of 1–5 according to these categories:

1. User goal:

Who is the ideal user and what is the main function this user wants to use this product for?

For Project Post Mortem, the main user and use case is a Project Manager going on the platform to research trends, resources, and insights from related project. This means that the Project Manager needs to be able to search, view past resources, send survey, and view survey results.

2. Business goal:

What is the business’s objective in creating this product? How does this project allow the business to reach it’s goals?

In my case, the business is the organization in which the Project Managers work in. For the organization, it’s important to be able to easily view trends, and view post mortem findings.

3. Organizational readiness:

How ready is the organization to implement the changes? Culturally, structurally, etc.

For project PM, it’s important that the proposed features would not create too much interference with the current flow of things. The features proposed would have to be low in resistance.

4. Technical implementation:

How difficult or time consuming will the implementation of this feature be?

Because I don’t know how to code, I made sure to look at this category based on existing technology that is available to me.

Balancing the ranking afterwards by taking the average of each, I was able to create a ranking of priorities of each feature:

I narrowed down the highest priority features to: Post mortem survey, post mortem results, project report, search results, and trends.

Finding existing tools

Since I don’t know how to code, I needed to find as many existing resources and tools out in the wild as possible to make my vision happen.

I’ve been lucky enough to have the opportunity to work at a couple start-ups where I’ve seen many instances of “hacking” together of a product. Based off of those experiences (plus a ton of Googling), I distilled down the tools I needed to use:

1. Typeform — A survey with an easy to use interface and the ability to integrate with other platforms.
2. Zapier — A web-based service that allows users to integrate different web applications.
3. Google Sheets — A web-based spreadsheet by Google that has the ability to integrate with other platforms.

Planning and strategizing

Based off of the priority list and the platforms I chose, I proceeded to plan how to build the features.

1. Post mortem survey — This was pretty simple with a direct solve using Typeform’s interface.
2. Post mortem results — Results needed to reflect individual employee’s feedback for each project. I knew that Typeform has the ability to track answers pretty easily. However, the answers needed to be gathered into some sort of Database. I used Zapier to connect Typeform results to a Google Sheet.
3. Project reports — Typeform has the ability to allow users to upload files but will be difficult to compile all the information into groups by project.
4. Search results — This was the tricky part, although Typeform is able to collect answers, the ability to search based on certain specification was more difficult. I dove deep into Google to find functions that could set up an easy way to find information. Due to limitations, I settled on the ability to filter.
5. Trends — Using Google Sheets, I knew I could manipulate the data to get some general trends, especially if I could find a way to separate out each category of industry, client, etc.

With this plan in hand and a lot of Googling, I came up with my version of a MVP.

Result

Post Mortem Survey (Typeform):

The survey includes questions grouped into Industry, Client, Management, Project Type, Discipline, Team, and Resources. Questions populate with conditional answers to keep the consistency.

Future version would include auto-detection of user, project, client, discipline so that the user would not have to fill out those information. In addition, future version would have a better way to organize uploaded files and add descriptions to each.

Post mortem results/Database

This Google Spreadsheet is set up into the different categories corresponding to the categories in the survey. Using functions (=AVERAGEIF) I created input boxes in each section for searching data on a specific word and the average for each response.

Future vision would see this on it’s own platform.

Project reports

This is the result of linking Typeform up to Google Sheets with Zapier! Answers from the survey are automatically populated onto the next line on the Google Sheets I’ve set up. This was a little difficult to do as there are many steps and every step needed to be exact.

Search results and Trends

Here is the resulting MVP. Results are collected into the Google Sheets. As a PM, you can come into this sheet, search based on industry, client, and PM name to see the average score for satisfaction, interest, etc.

The sorting mechanism on Google Sheets then allows users to sort alphabetically so that resources (files and links) can be grouped and viewed together by industry, client, PM name, and project type.

Evaluation and Learnings

1. The less features that are built out, the more work for the user.
2. According to some definitions, this might not qualify as a MVP, but rather, a prototype.
3. This process can only work for certain projects. Most MVPs, I feel, would require some kind of code to deliver actual value. Since Project PM is an internal tool for people who are already familiar with technology, there is more leeway.
4. The more “hack-y” the MVP, the better labeled and laid out it must be. I originally wanted to create separate sheets for each category, however, I realized it was very difficult to do and that I could solve the issue at hand with better organization of the cells.