Creating a faster order decisions with SlackBot

Creating a faster order decisions with SlackBot

2021 • eFishery • Supply Chain Ops • 7 min read

2021 • eFishery • Supply Chain Ops • 7 min read

Overview

Overview

Live-stock (fish commodity) management in the supply chain process is not an easy thing to do. It's heavy on operational. Since eFishery Fresh business began, eFishery has consistently iterated on the optimization and efficiency of the process.

One of the spotlighted challenges is on managing incoming orders to matchmaking with available supply & logistic. In this project initiative, we want to cut SLA of order confirmation & making operational activity more efficient.

Live-stock (fish commodity) management in the supply chain process is not an easy thing to do. It's heavy on operational. Since eFishery Fresh business began, eFishery has consistently iterated on the optimization and efficiency of the process.

One of the spotlighted challenges is on managing incoming orders to matchmaking with available supply & logistic. In this project initiative, we want to cut SLA of order confirmation & making operational activity more efficient.

Due to some of the data & work is confidential, the following designs, numbers, text,

have been modified, blurred, or omitted on purpose.

Due to some of the data & work is confidential, the following designs, numbers, text,

have been modified, blurred, or omitted on purpose.

Digging Insights on Existing Condition

Digging Insights on Existing Condition

To deeply understand about existing condition, PM and I started to conduct an evaluative research on current user activity, ops flow, & related data.

To deeply understand about existing condition, PM and I started to conduct an evaluative research on current user activity, ops flow, & related data.

  1. Talk with Internal Users

  1. Talk with Internal Users

We conducted interview with users who involved in order confirmation flow, from Sales to Admin. We tried to find out what is their main activity when an order submitted until it is confirmed. Including the pain points & their expectation of existing flow.

We conducted interview with users who involved in order confirmation flow, from Sales to Admin. We tried to find out what is their main activity when an order submitted until it is confirmed. Including the pain points & their expectation of existing flow.

  1. Talk with Buyers

  1. Talk with Buyers

We also conducted interviews with buyers to find out what is their goals expectations when placing an order to us, to see how our current service affected them.

We learn & analyze the related documents & data of order confirmation. How our internal users managed that document & how they managed the data. Also we look into the existing tools and how they interacted with it. Even, we deep dive about relation on this operational flow to company metrics.

After doing several research & analysis, we could map the activity along with pain points & the expected goals.

We also conducted interviews with buyers to find out what is their goals expectations when placing an order to us, to see how our current service affected them.

We learn & analyze the related documents & data of order confirmation. How our internal users managed that document & how they managed the data. Also we look into the existing tools and how they interacted with it. Even, we deep dive about relation on this operational flow to company metrics.

After doing several research & analysis, we could map the activity along with pain points & the expected goals.

Simplified User Journey of Order Confirmation.

Simplified User Journey of Order Confirmation.

Simplified User Journey of Order Confirmation.

Common Order Confirmation Method

Common Order Confirmation Method

Before the ideation, I took the time to casually look at the order confirmation process in e-commerce in general. Even is not apple to apple, because we are B2B.

But, as we know, commonly e-commerce has a same processing time when the seller thinks to choose processes or reject an order. And sometimes when the seller suddenly can’t process the order, it can make us frustrated.

Before the ideation, I took the time to casually look at the order confirmation process in e-commerce in general. Even is not apple to apple, because we are B2B.

But, as we know, commonly e-commerce has a same processing time when the seller thinks to choose processes or reject an order. And sometimes when the seller suddenly can’t process the order, it can make us frustrated.

Potentially E-commerce User Journey on Order Confirmation part

Potentially E-commerce User Journey on Order Confirmation part

On the other side, in my opinion OTA service design has a better order confirmation, though the business is slightly different. They provided options and “stock” as seats/rooms, hold for us when we fill the form to reserve, paying, & we got the ticket/voucher!

On the other side, in my opinion OTA service design has a better order confirmation, though the business is slightly different. They provided options and “stock” as seats/rooms, hold for us when we fill the form to reserve, paying, & we got the ticket/voucher!

Exploring Idea

Exploring Idea

Following all the findings and insights above, I tried to make a question with HMW style:

How might we could speed up the order confirmation process so that we can immediately inform the buyer about their order?

Long story short, after did some brainstorming rite, we are converging on one idea to createn an Auto-confirm Order. But, when we challenge this idea and assess with existing conditions, we can’t go on with the idea to fully automatisation the confirmation. So, we rework the idea and come up with an initiative to provide order confirmation suggestion using Slack Bot.

Following all the findings and insights above, I tried to make a question with HMW style:

How might we could speed up the order confirmation process so that we can immediately inform the buyer about their order?

Long story short, after did some brainstorming rite, we are converging on one idea to createn an Auto-confirm Order. But, when we challenge this idea and assess with existing conditions, we can’t go on with the idea to fully automatisation the confirmation. So, we rework the idea and come up with an initiative to provide order confirmation suggestion using Slack Bot.

User Journey Proposal for Order Confirmation.

User Journey Proposal for Order Confirmation.

SlackBot Design

SlackBot Design

Due to the system limitation of Slack, I can’t custom the bot interface/chat. So, I just beat the layout exploration, set the flow, & make the copywriting following the appropriate design that can be used in Slack.

Due to the system limitation of Slack, I can’t custom the bot interface/chat. So, I just beat the layout exploration, set the flow, & make the copywriting following the appropriate design that can be used in Slack.

Below is the view when the order is just approved/rejected. The Bot will be shrinking the details of the order and will change them into order summary. It will help our admin to distinct unapproved orders & approved/rejected orders.

Below is the view when the order is just approved/rejected. The Bot will be shrinking the details of the order and will change them into order summary. It will help our admin to distinct unapproved orders & approved/rejected orders.

Following the two gestalt laws, proximity & similarity, I break the order details into sections. And to make it fastest to skim, I sort the sections to be a pattern like telling a story. So, our admin could be easier to get details of the order.

Following the two gestalt laws, proximity & similarity, I break the order details into sections. And to make it fastest to skim, I sort the sections to be a pattern like telling a story. So, our admin could be easier to get details of the order.

In addition, we put the details of the assessment on the thread of the bot, to provide the reason why our system assessment gave that status, and make our admin decision more accurate.

In addition, we put the details of the assessment on the thread of the bot, to provide the reason why our system assessment gave that status, and make our admin decision more accurate.

Internal Demo & Testing

Internal Demo & Testing

When development is almost done & the staging was ready, PM & I bring this bot to our internal users. We show how this bot work and let them try. And the responses are positive, even our internal users were enthusiastic and participate to give us advice to improve some parts of this bot.

When development is almost done & the staging was ready, PM & I bring this bot to our internal users. We show how this bot work and let them try. And the responses are positive, even our internal users were enthusiastic and participate to give us advice to improve some parts of this bot.

Result & Lessons

Result & Lessons

After the bot was released, this bot cut the lead time of order confirmation from >4 hours to <4 hours. It also affected to increase buyer satisfaction. And this project could be the foundational system to actualizing the auto-confirmation order in the future.

For the lesson learned, one of the highlighted things to know is the close collaboration between Designer x Back-end Engineer.

At the discovery phase collaboration began with PM & Stakeholders, and when it was defined, I closely collaborate with BE Engineer. When commonly designers are closer to front-end engineers in collaboration to make an interface, we prove that designer could be also involved & collaborate with back-end engineers in making service design better. Yes, we prove that; experience design beyond the interface.

After the bot was released, this bot cut the lead time of order confirmation from >4 hours to <4 hours. It also affected to increase buyer satisfaction. And this project could be the foundational system to actualizing the auto-confirmation order in the future.

For the lesson learned, one of the highlighted things to know is the close collaboration between Designer x Back-end Engineer.

At the discovery phase collaboration began with PM & Stakeholders, and when it was defined, I closely collaborate with BE Engineer. When commonly designers are closer to front-end engineers in collaboration to make an interface, we prove that designer could be also involved & collaborate with back-end engineers in making service design better. Yes, we prove that; experience design beyond the interface.

Back to home

Back to home

Let’s talk &

Get in touch.

thalhah.crdv@gmail.com

Thalhah S. Robbani © 2024

Let’s talk & Get in touch.

thalhah.crdv@gmail.com

Thalhah S. Robbani © 2024