CopyCraft
ChatGPT Dashboard for Copywriters
2024
Brief
GenAI is changing the way copywriters work., freeing them up to take on a broader range of responsibilities.
This dashboard was designed to help copywriters keep track of these responsibilities and their ChatGPT-supported work processes.
A solo project completed as part of a UX design module.
My role
User experience design, prototyping, user interviews, user testing, branding
Contents
DISCOVER
THE PROBLEM: CHATGPT CAN WRITE COPY. IT CAN’T BUILD MARKETING CAMPAIGNS.
If you’re a copywriter, you’re probably no stranger to using ChatGPT to churn out a few drafts for you to refine. You might have even written something similar to this:
Write copy for a marketing Email Direct Messaging (EDM) for a Black Friday sale. The product is an OLED 32” Monitor. Its main selling points are colour-accurate display, 4K HD resolution, an ergonomic monitor stand.
Your target audience is graphic designers working from home.
But when chat logs just keep piling up and a lot of them look the same, it’s hard to find that one session where you curated the perfect copy– the one that’s really doing numbers for your latest marketing campaign.
So what?
Even though it’s useful for curating prompts, the chat format makes it difficult to track, reuse and improve on past sessions.
Design opportunity: a dashboard that supports the copywriter in tracking ad analytics and adapting past campaigns.
Target users
Copywriters (in-house, agency and freelance), who are managing copy for multiple products, platforms and target audiences.
User research
I used convenience sampling to recruit 3 participants for a semi-structured interview on their experiences using ChatGPT as part of their work process.
Junior copywriter
Freelance
Branding, social media, website UX copy
Senior copywriter
SME e-commerce company
EDMs, desktop, mobile copy
Copywriting intern
Local F&B brand
Social media, websites, delivery apps
Key metrics
From my user interviews, I found that copywriters kept a close eye on 2 main metrics: click-through rates and unsubscribe rates.
For a GenAI-assisted workflow, this didn’t change, but users were interested in tracking the number of prompts it took to curate a satisfactory draft.
1
Click-through rates for ChatGPT-assisted campaign copy
2
Unsubscribe rates for ChatGPT-assisted campaign copy
3
Number of prompts for satisfactory draft
Insights
Users felt that the chat format of ChatGPT became difficult to use with too many sessions.
Users felt that the time taken to curate a satisfactory response from ChatGPT was slowing down their workflow.
Users found a disconnect between their writing workflow and their other analytics tracking responsibilities.
Affinity mapping
DEFINE
Problem statement
Copywriters use ChatGPT to write compelling copy. However, it can be difficult to keep track of which prompts are the most and least successful, in terms of click-through rates, unsubscribe/unfollow rates, and time taken to curate the ChatGPT responses.
Persona
DESIGN
How might we...
Provide a variety of options for users to view ad analytics data?
Support users comparing data for different ChatGPT-assisted campaigns?
Allow users to easily customise campaign prompts for ChatGPT writing sessions?
The solution
A dashboard that integrates ChatGPT and ad analytics trackers like Klaviyo, Mailchimp and Google Analytics.
It should help copywriters to easily track key metrics: click-through rates, unsubscribe rates and number of prompts it takes to produce a satisfactory draft for each campaign.
A prompt creator form that allows copywriters to initialise ChatGPT sessions quickly and easily, supporting efficiency as a user goal.
Initial design exploration
First, I looked to common ad analytics trackers for inspiration on how they displayed information, as I would be designing for similar users. This included Klaviyo, Mailchimp and Google Analytics, which also offered dashboards.
These dashboards first provide an overview of information, and then narrow down, allowing the user to compare performance of specific campaigns across platforms.
Then, I explored dashboard design as a whole.
Power BI, Microsoft’s business analytics service, prioritised consistency and clarity over sleek or aesthetic-first designs.
They also recommended putting the highest level of information at the top, before providing more specific or granular data views below.
DELIVER
Wireframes
Based on my findings, I chose a card-based approach: modular, compact and concise.
Wireframe 1: Desktop, Dashboard view
The first tabs “above the fold” would be the Overview card and Best performing card, providing the most crucial information at a glance.
Further down, the user is able to compare campaigns and view campaigns that ChatGPT had found issues with.
I also designed an initial prompt creator for ChatGPT, for the user goal of producing copy quickly and efficiently.
The user is able to input their desired tone, product and campaign details, and are taken to a ChatGPT session initialised with this prompt.
This saves time and effort in writing the initial prompt.
Wireframe 2: Desktop, Prompt Creator view
Final prototype (mid-fi)
In the transition to mid-fidelity, I focused on providing the user with a variety of methods to interpret the data from the ad analytics tracker.
Frame 2: Desktop, Prompt Creator page
Prompt Creator page
The Prompt Creator page streamlines the process of writing an initial set-up prompt for each session.
Users can quickly fill in the campaign’s voice, target audience, product and other sale details.
Helpful prompts are provided to guide new users.
That takes them straight to a ChatGPT session initialised with those details.
If editing an existing campaign, the user can also see the statistics of that campaign.
Recent campaigns:
Allows the user to compare the success of their 3 most recent campaigns.
Best performing:
Shows the user the highest-performing campaigns, based on the 3 key metrics found from the user research.
A convenient “Continue session” button allows users to build off the success of this campaign.
By campaign:
Provides the user with cross-platform data for each campaign, across the last six months.
Compare campaign:
Supports the user in comparing the success of two specific campaigns.
Useful for deciding between different approaches, similar to A/B Testing.
Campaigns with issues:
Alerts the user to any campaigns with specific issues.
“View recommendations” provides the user with AI-assisted recommendations on how to improve the campaign.
Frame 1: Desktop, Home page
Tablet view
With a tablet view, campaign analytics are within easy reach, without disrupting the writer’s workflow.
The dashboard is split into different tabs in the side navigation, according to user intention.
Figma prototype
View the desktop prototype
View the tablet prototype
