A practical guide to content analysis with ChatGPT for marketing and product.

Like many, we're excited about the possibilities of Gen AI - and how it can help time-pressed startups to turbocharge their operations. There's a lot of advice out there, most of it covering GPT's party tricks, little on how to get valuable, high-quality results ("act like a marketer and write me a social media plan" anyone?)

Here's my theory. Humans matter more than ever to drive quality work. Functional expertise, a product and a customer understanding are key. Back it with a well-structured process, and you can squeeze proper, lasting value from ChatGPT and peers.

In the next few Growth Playbooks, we're going to get our hands dirty. No fluff, no memes. Just solid, practical steps on how startup operators can harness Gen AI to boost their marketing, sales, and customer service ops.

Before we dive into functional topics, we'll kick off with some universal use cases. Today, we'll look at content analysis with ChatGPT with examples of how to turn market reports into valuable content and how to decode customer interviews for product planning.

‍

KEY TAKEAWAYS

A clear objective and well-constructed prompts are key to getting good outputs from ChatGPT.

Our top tips for content analysis:

‍Format matters. Make sure that the content you wish to analyse is in the right format. PDF plugins are still limited, so best to stick to text-based content, or use the Bing integration for URLs (note these are Plus features, which cost $20 per month).‍
Don't forget when the models were trained. GPT-4 has a knowledge cutoff date of September 2021. You'll get half-truths and hallucinations if you ask the platform for information on more recent events. Bing's integration can help reference more recent content, but remember that the model hasn't been taught on that data, so you prompts should be adjusted to fit.‍
Leap those content length hurdles. The tool can process roughly 2500 words at once. But it's possible to upload content in chunks, then query it as one. Tools like ChatGPT Splitter and ChatGPT File Uploader will save you time and pain.‍
Chain prompts FTW. A 'zero hit' prompt is a simple 'question' - 'answer'. But when have you got everything you want from a conversation on first go, particularly to a complex ask? Chain prompts together in a process to get better results with granular control. Once you've tuned an analysis of a topic, ask the platform to build an article by defining an objective (and key phrases), then outlining a structure, then writing the content.

‍

🧠 Turn reports and research into killer content.

Use this to:

Enhance your market research or competitor analysis.
Craft thought-leadership articles or social media posts based on the report or research.
Share new market insights with your board or team via a succinct email.
Create a slide using the key stats for your investment deck.

1. Kick off with great content. Find a high quality content or a report from a reputable source. Tweak the format so ChatGPT can read it. Then upload it using a service like ChatGPT File Uploader.

2. Ask ChatGPT for:

An overall summary of the key points within the content.
A summary by section.
Key stats or figures.
More information on the thought leaders within the report.
A specific question you have about the report.

3. Follow up by:

Asking ChatGPT to craft a selection of social media posts. Make sure you specify the objective of the content, channel you want to use, your company's tone of voice and the style (you can reference a favourite prominent author). Give it feedback on the copy and ask it to redo the task if you're not satisfied (with clear feedback to improve on).
Asking ChatGPT to craft a short email. Make sure you specify objective (sales, or marketing, nurture or action, for instance) the length and the recipients of the email, to effectively and quickly summarise the report/research.
Asking ChatGPT to tell you what information potential investors might like to see in your investment pitch deck to create a shortlist for your presentation.

4. In action ⚡️

A Marketing Manager wants to include more topical content from industry thought leaders on his social feeds. He's a small team of one, so he wants to use ChatGPT to help him extract the right data from McKinsey's The state of AI in 2022—and a half decade in review and turn them into engaging social content He also wants to summarise the findings for his very busy CEO.

>> See the report summary here

>> See the LinkedIn posts here

>> See the CEO summary email here.

🗣️ Craft your product roadmap with customer interviews.

1. Use this to:

Enhance your customer feedback process and insights.
Categorise insights to help inform your product roadmap and future development.
Craft summary emails to send back to your customer, outlining what you learned and proposed next steps.
Provide summaries for senior stakeholders in your business about what this research uncovered.

2. Kick off with great content:

Do you have transcripts from customer discovery interviews? Reviews Trustpilot or G2C? Surfacing great insights rely on great inputs, so data is key. Collate these, categorise them and feed each tranche of data into the platform separately to get best results.

3. Ask ChatGPT for:

A summary of positive and negative feedback.
An overall sentiment score out of 10 for the feedback.
Key product wants/ needs.
New feature ideas based on customer feedback.
How to approach testing the customer requests.
Whether your customers offer similar features.

4. In action ⚡️

A Product Manager is conducting customer interviews, as it's been 6 months since they launched their latest product release. She has over 20 interviews to get through and wants support in analysing the findings and forming her product roadmap recommendations.

>> See the customer interview transcript summary here

>> See how to validate the customer recommendation here

>>See how to prioritise two customer suggestions here.

>> See how to layer competitor analysis into the findings here.

>> See how to iterate competitor analysis for more in depth insights here.

As we conclude, it's essential to highlight that Gen AI tools, like ChatGPT, can be an effective asset for startups, streamlining operations and improving efficiency. It enables you to turn complex market reports into digestible content and deciphers customer conversations for better product planning. However, it is important to acknowledge that ChatGPT is an evolving tool, constantly under development, with its capabilities expanding over time.

While ChatGPT is designed to enhance human capability, not replace it, its increasing sophistication hints at utility in a wider range of tasks, thus presenting even more opportunities for start-ups. To leverage it effectively, start-ups need clear objectives and well-thought-out prompts. Familiarising yourself with its specific features and limitations, such as its knowledge cut-off of September 2021 and a 2500-word content limit, is vital.

Despite these limitations, ChatGPT proves to be a significant resource for startups seeking to save time and optimise their operations. By following the practical steps outlined in this series, start-ups can harness the potential of Gen AI to drive their marketing, sales, and customer service operations forward.

🗞️ What's got our attention?

Automated content generation is getting good. Really good. So good that Wordpress just rolled out a tool which can create a blog post for you from scratch. Read more on The Verge here. Consider taking Notion AI for a spin too, it's integrated into your workflow, and feels like magic.
Could it take a decade for economies to reap the productivity benefits of generative AI? Check out this article in the Financial Times to find out more.
Large corporates are taking ChatGPT very seriously, with McKinsey reporting that over 50% of their staff globally are using the tool. Read more on VentureBeat here.

Written By

Ann-Marie Rossiter

Senior Marketing Manager

Ann-Marie has almost a decade's worth of experience building and scaling startups in the UK tech industry. From luxury travel and corporate events to robotic process automation, she loves challenger brands with a killer idea. Passionate about ESG initiatives, Ann-Marie has implemented many programmes in her previous roles to support charities tackling homelessness and helping young people and women into careers within the tech sector.

At Forward, Ann-Marie is responsible for our brand and building experiences that founders love, whilst supporting the growing businesses in our portfolio via the Studio.

Building a startup? Stay ahead of the pack with The Forward Fortnightly.

Subscribe today to get tech news and trends, growth tips and advice and exclusive event invitations, straight to your inbox every fortnight.

Mastering startup launch PR: Dos & Don’ts.

read

No items found.

Takeaway