The subtle skill of using ChatGPT
How to best use the tool to boost productivity while keeping credibility
No doubt that by now, you will have tried out ChatGPT or other such programmes. From code reviews and ideation to investment advice and travel itineraries, ChatGPT can give you an answer to just about anything. And honestly, the results are quite impressive. But try asking the programme about something you know a lot about, and you’ll notice that the answer often isn’t necessarily wrong, but it’s not quite right either. It’s just a bit… off.
And while that’s not an issue when you know the subject and can spot it’s wrong, you have to wonder – what about when you ask Chat GPT about a topic you’re not that familiar with? How do you spot the inconsistencies or inaccuracies then?
How ChatGPT works
When using ChatGPT, it’s important to always keep in mind that the input of the model is mostly based on web pages, Wikipedia and other web texts. The way that a prompt is entered into ChatGPT is mirrored in its response, since it matches up the input given with the most likely answer based on contextual cues. For example, if the input is simply worded and minimal, then the output will be similarly so and provide little nuance or depth. It doesn’t have the creative interpretation a human would do if asked the same thing, so it provides the most mathematically correct response to the given input.
Everything ChatGPT responds with is based on an algorithm trained on a predetermined set of information. The response is built up out of mathematically ‘correct’ or ‘logical’ output but lacks the nuance and diverse interpretation of a human mind. There will be times when a literal response is fine, but plenty more where more context and human intelligence are needed to actually take the answer as truth.
Here’s an example from our own experience. Whilst in the first stages of building our data warehouse, we used ChatGPT to increase our productivity in writing SQL queries to create views in BigQuery. Views are beneficial because they refresh the table view based on the underlying data each time they’re called. Querying a view that is based on another view refreshes the entire lineage of views. Chaining the queries in this way can help in creating a clearer flow of data.
However, based on our prompts, ChatGPT generated “Create Table” queries. These queries create a new table that isn’t connected to the table it bases its result on. Also, a create table query can only be run once because you cannot create a new table if another table with the same name already exists. This makes a flow of data impossible without creating additional scheduled queries or stored procedures.
Here our human knowledge played a crucial role in spotting the inaccuracy of ChatGPT. Though the output was technically correct from a code perspective, the concept behind the code wasn’t what we intended to produce.
How should we use ChatGPT then?
ChatGPT is a productivity booster. At NSL we use it to automate mundane tasks, search for inspiration and expand our knowledge on certain topics. But we always keep in mind that ChatGPT isn’t able to fact-check with the same context and complex nuance as humans can.
Think of ChatGPT as an intern or junior, an efficient assistant that contributes to the furtherance of a need or effort. But ultimately, the person using ChatGPT is the one who needs to engage in complex thought both when giving the input and when receiving the answer.
A reminder to make ChatGPT work with you, rather than for you.
--
We’re passionate about all things data and are striving to make it more approachable, so you can expect regular updates on this topic. Get in touch if we can help guide your team to a better understanding of data, or if you’re ready to increase your revenue with the help of our experts.
By Bram Verleur
Head of Data at Not Selling Liquid
With our Inside the Industry series, we’re exploring the ever-evolving world of e-commerce uncovering trends, technologies, and insights shaping the brands of tomorrow.
Meta’s big algorithm update (Andromeda): what you should and shouldn’t do
Meta (Facebook & Instagram) rebuilt its ad “engine”: Andromeda. The system now picks from a much larger library of your ads and matches the right message to the right person. That means less micromanaging audiences, more winning with creative variety and simple account structure.
Why Retention is key to your brand longevity
Across all segments in fashion, there has been a major gap in businesses' ability to retain customers for more than a one-off purchase.
Shopify Markets: A unified home for business expansion into international markets and seamless implementation
Shopify is always evolving its services and offers. Shopify Markets is an innovation that boosts brands and businesses to wider audiences and growth. A unified home for business expansion, it encourages brands to sell to multiple markets under one store, while still creating tailor-made experiences per market.
Rethinking Marketplaces: How we can help you make the best of them
Change is the only constant in the dynamic fashion e-commerce landscape, and this time, we’re putting a spotlight on how Marketplaces will continue to shape the scene.
Creative is the new Targeting
The world of e-commerce is constantly evolving, always influencing or adapting to user behaviour. With recent innovations and technological advancements such as Privacy updates (like iOS14), rising CPMs, and platform algorithms, targeting is no longer what it used to be.
Why migrating to Shopify should be every fashion brand’s strategy for sustained growth and success
The e-commerce landscape is anything but static, and the ever-changing environment requires brands and businesses to stay sharp for long-term success. Many established fashion brands have switched to Shopify as their preferred online platform due to its many advantageous features that directly address industry challenges – and others are sure to follow.
The importance of securing your Shopify and ad platform log-ins with two-factor authentication: don’t overlook this crucial security step
With cybersecurity threats on the rise and becoming increasingly harder to spot, it’s even more important for e-commerce businesses to protect themselves from hackers. Two-factor authentication (2FA) adds an extra layer of security by requiring users to provide two forms of identification before accessing their accounts. This significantly reduces the risk of unauthorised account access, especially in cases of compromised passwords or phishing attempts.
2024 Peak Season: Our BFCM Results & Learnings
Another Black Friday/Cyber Monday week(end) has been and gone, and it’s time to face the numbers. Looking at the data from all of our clients in north-western Europe, what have we learnt from this year’s most busy weekend in the e-commerce calendar, and how can we help you create a strategy for the upcoming year?
Not selling liquid becomes the exclusive digital partner for CIFF, offering expertise to retail brands in the fashion world
E-commerce agency Not selling liquid (NSL) has formed a unique partnership with the Copenhagen International Fashion Fair (CIFF). Not selling liquid builds, manages, and grows the online presence of fashion brands in the digital space and their expert knowledge will provide CIFF and its partners with the tools they need to become future-proof. This partnership with CIFF is one-of-a-kind, as there has never been an agency/tradeshow relationship like this in the past.
Important updates to our unique Data Dashboard Services, providing you with even better insights for growth
Since launching our Data Dashboards earlier this year, we’ve continued to refine and develop them to ensure we provide you with the most up-to-date input possible. Our latest release has added several features and has significantly upgraded the services. These Data Dashboards are a game-changing development in fashion e-commerce, combining insights from all ad managers, Shopify stores, and Google Analytics.
Three tips to boost your sales by escaping the standard sales calendar and creating a tailor-made solution for your brand
In the e-commerce fashion world, many brands follow the same calendar: they launch their seasonal (AW/SS) collection and then have a sale before the next season launches. In between you’ll (usually) find an archive sale and then the expected BFCM weekend boost.
Black Friday 2024: Our curated list of resources that can help your strategy
You know that summertime is over when that first flurry of Black Friday newsletters starts streaming in. A mass of information offering you advice, help, and tools to make the most of the upcoming sale weekend. From our fashion perspective, here are the top five resources from our esteemed partners worth diving into.
Black Friday 2024: Leveraging Peak Season for Long-Term Growth
As we enter the month of October, you’ve probably already started receiving tips, tricks, tools, and tactics on how to approach the whirlwind that is Black Friday/Cyber Monday. We’re here to help keep you grounded and share some real advice on how to go about this significant moment in the e-commerce calendar.
E-Commerce Design: How Creativity and Strategy Shape Success
In the competitive world of e-commerce, design plays a pivotal role in driving sales and fostering customer loyalty. There’s a skill in designing a beautiful website that translates the brand identity clearly, as well as being functional and easy to use.
5 things to A/B test in your Shopify store before peak season
A/B testing provides data-driven conclusions that you can transform into actionable insights and drive your business growth. A cornerstone of Conversion Rate Optimisation (CRO), it systematically compares and tests two versions of a webpage and determines what works better and for who.
How product performance analysis helps you boost profits
Throughout all types of industries, businesses have products that perform brilliantly and generate income and others that don’t do quite as well. The key is knowing which products are the star performers and boosting them as much as possible, and also evaluating the ones that don’t perform well, taking in a number of factors. This is where analysing data correctly is hugely beneficial.