UX: Restaurant Insights & Delivery Planning Dashboard

Tanzir Rahman
Dec 28, 2018 · 7 min read

TL;DR: I designed an improved goal-oriented overview page for restaurant dashboards compared to what’s already available in the market. I also designed a page for delivery dispatch planning.


From 2016 to 2017, I worked with a few restaurant owners (located in South Asia) to help promote their business using Facebook Ads. I helped the owners bring in about $1000/day in revenue with just a $10/day budget.

In this article, I explain how I utilise my experiences, observations and research to empathise and synthesise action points, which I then used to ideate an innovative restaurant insights and delivery planning dashboard.


Key challenges in operating a restaurant:

Out of these challenges, the ability of a restaurant to sell more is absolutely critical because it gives them much needed cash to pay suppliers on time and operate optimally. I know this because I’ve witnessed first-hand, the inability to pay suppliers due to low sales.

Therefore, helping restaurants sell more helps them run more efficiently and eventually solve other managerial issues.

One other problem that is a blocker to more sales that I have witnessed over the years (as someone who orders food delivery multiple times a week) is the difficulty restaurants face maintaining in-house delivery personnel. Their dispatch planning is ad-hoc which results in inefficiencies, the personnel heads out for delivery and new orders come in just moments after s/he has left and the customer experiences long wait times.

Restaurants typically want to maintain their own delivery fleet (outside delivery services by mobile app based companies) to cater to orders they receive directly from the customer via various channels (phone, website, social media, messaging apps, etc).

Mobile app based delivery services charge restaurant owners a commission which eats into a big portion of their profit margins.

Possible Solution

In order to sell more, restaurants need insights of consumer habits and demands. In the past, when I would market restaurants, we would always try to come up with fresh and relevant offers and promotions that would perfectly fit consumer tastes. We would make educated guesses and whenever the offer met consumers demands, it would significantly increase sales — for example, lunch discount on certain days of the week to draw in office/corporate crowd.

Food delivery services such as Foodpanda often provide printed sales reports to justify their value proposition and motivate the restaurants to keep performing on their platform. One of my past clients (restaurant owner) showed me a printed report highlighting overall sales metrics, that Foodpanda had given him.

However, there are multiple digital channels of placing orders now, some people order through Facebook inbox, others might do it through phone calls or the several food delivery apps.

To make sense of this immensely segregated consumer data, a dashboard would really help to reign in on a restaurant’s optimisation of service delivery.

For instance, many restaurants now actively hire people with the job descriptions that demand them to be able to manage multiple food delivery apps to fulfil orders and operate at scale. A unified dashboard would make the lives of these individuals easier and be critical to the restaurant’s success.

The screenshots provided are just samples of data from social media. One of the restaurant owners I worked for, owned three different brands of chain restaurants that he developed from scratch and he is well connected with other successful restaurateurs and industry professionals.

In addition, I knew other restaurant owners and had keenly observed their operations. Hence, I felt fairly confident with my derivations.


“Elaborate usability tests are a waste of resources. The best results come from testing no more than 5 users and running as many small tests as you can afford.”

Source: https://www.nngroup.com/articles/why-you-only-need-to-test-with-5-users/

Following is the user persona of a restaurant manager, Shafi — who accurately depicts the typical end-user of our dashboard design.

Research & Ideation

Having clearly defined the goals, I started sketching using a pencil and paper. I kept in mind that most restaurants typically use some form of Point of Sale (which is mostly focused on billing and inventory management rather than providing detailed, contextual and actionable insights). Therefore, what I am designing would either be an improvement or additional feature to their existing system.

To further improve the quality of my ideations, I conducted competitive research. Sources included this Dribbble bucket I created for inspiration, this article — which makes it clear that the market is quiet segmented (I knew this but just for proof), also I had previously used WordPress restaurant order management plugins and other free online scripts such as GloriaFood.

The common theme among these solution is that they all provide order and sales trend data. However, what they lack is a clear sense of what this data is aiming to solve which makes it less user-friendly to interpret. For instance, simply telling me popular orders and sales trend does not tell me what needs to done to improve.

To further help the end user, we need to provide more controls to the data so that it can be easily manipulated and filtered to extract contextual meaning. Example of other variables we can bring into the dashboard are time of order, number of orders for each menu item at a granular level and make this all filterable and easily digestible.

a. Overview Page:

(i.e. Restaurant Insights — an Export CSV button may or may not be useful here.)

In the above wireframe, the graph would ideally be filterable by hour, day, week, month, etc. The day filters (Sunday, Monday, Tuesday, etc.) would potentially disappear when anything but hour is selected.

The avg. serving time statistic would help identify if restaurant performance is being hampered due to being understaffed, employee inefficiency or other operational issues such as the supply-chain. Employee related performance statistics could have its own page that could be navigated to from this main overview page.

Delivery and visiting customers could have a unified profile, which could then be developed into a sophisticated yet simple CRM — tailored for the restaurant management experience. Currently, some restaurants have a form of this, where a telemarketing software pulls up a customers profile, as soon as the customer calls the restaurant.

b. Delivery Planning Page:

The primary purpose of the above ideated design is to plan orders for delivery and manage the in-house delivery fleet for optimal dispatches. Anticipating orders ensures that the restaurant managers only dispatch when either the delivery personnel is at capacity or if there are no more incoming orders for their route.

The traffic source data is good to have to identify where orders are coming from — the wireframe displays mobile app delivery services but it could also have multiple messenger apps where the restaurant receives orders directly from the customers. It’s flexible. Benefits of having this:

In addition to planning for delivery, the historical trends of customer location and traffic source data points will aid in planning future expansions and promotions for the restaurant.

Knowing statistics such as Added to Cart, Cart Abandonment (Sales Funnel Drop-off) and Conversion Rates are commonplace in the eCommerce industry. The restaurant business is growing increasingly competitive where multiple restaurants are competing in the same neighbourhood with similar offerings.

In order for these restaurants to survive, they need to adopt more advanced digital technologies to find insights and tailor their offerings to better suit niche customer demands (for example, late night delivery in certain neighbourhoods, better offers/promotions for segmented audiences, lower wait times, etc).

Further Testing & Improvements

The best way to test would be to create a UI-based prototype because it enables test users to envision the service without getting sidetracked by the missing details.

Further improvements might include, clicking on the data trends and getting grouped data on customer demographics such as age, gender, location and so on — to better segment the restaurant customers and deliver the service experience to their unique preferences.

Tanzir Rahman

Written by

Data-driven product designer; passionate about coding and working with big data. Slightly geeky, slightly goofy. • tanziro.com | gadgetmama.com