How Hypotenuse AI produced consistent, on-brand content for ecommerce business
3
93%
800+
Our customer is a mid-sized ecommerce company ($100M+ annual revenue) with 3 distinct brands focusing on athleisure wear, sports accessories, and wellness products. While brands are encouraged to follow a corporate brand guide, each struggled to create product and marketing copy that was on-brand for their specific market segment and target audience. Over a 4-week period, Hypotenuse helped develop 3 distinct custom AI writers for the brands, each trained on the specific nuances of the brand context and tone of voice.
The Problem
Inconsistent product copy when companies own multiple brands
Each of our customer's 3 subsidiary brands has its distinct target audience, positioning, and messaging requirements, making it challenging to find the right balance between alignment with corporate guidelines while also allowing room for brand-specific nuances.
Writing product copy is time- and labour-consuming
As copy volume increased beyond 400 pieces of content per month, the in-house content team struggled to keep up. During peak seasons, the high volume of content led to more copy errors and inconsistencies which negatively impacted conversion rates and sales revenue. Prior to using Hypotenuse AI, year-over-year sales had declined by 16%.
Content marketing and product copy is misaligned
A 12-person content marketing team handled everything from advertising to web copy without coordinating with the product team. This resulted in inaccuracies and quality issues in product descriptions. To create a unified product and brand experience for customers, there needs to be alignment between the product and content teams.
The Solution
Trained AI model to generate on-brand copy
Hypotenuse AI utilized internal files and documents from our client to coach our AI and produce 3 distinct tailored AI models that learned the brands' unique voice, tone, style and format.
More specifically, we trained the models on 20 existing product descriptions from each subsidiary brand provided by the client, and over a 4-week period iterated on the output directly with the customer's product and content teams, refining the AI models to accurately and consistently replicate the key elements that defined each subsidiary brand's distinct voice and personality. Feedback directly from the customer helped further fine tune the models.
The end result was 3 highly specialized AI content models capable of producing unique and on-brand product descriptions that mirrored the voice of each individual brand and adhered to their product and marketing requirements.
High quality copy at scale
Bulk workflows power Hypotenuse AI to generate product descriptions for entire catalogs quickly. This allows ecommerce brands to scale to large volumes of high-quality descriptions without compromising on content quality.
Using Hypotenuse AI, businesses can generate product descriptions for entire catalogs in a matter of minutes, thanks to the implementation of efficient bulk workflows. These workflows are specifically designed to cater to the needs of enterprises with substantial product volumes, allowing for seamless import and export processes using CSV files or APIs, and bulk generation of thousands of products at one go. This not only saves time and effort for copywriters but also ensures that the quality of the copy is consistently maintained.
Bulk workflows powered by custom AI models helped our customer significantly reduce time spent writing on-brand descriptions. This resulted in higher conversion rates and sales.
On-brand AI-generated marketing content
Hypotenuse AI has 20+ templates specific to marketing content needs, including for blogs, ads, landing pages, social media and comms.
Once we develop a bespoke AI for a brand, it is available across all templates on our brand. Now, with 3 custom AI models, our customer was able to scale up their marketing content without additional resourcing while still preserving each brand’s unique voice.
The AI-generated content matched each brand’s established tone and brand voice, producing blogs, ads, web content, and social media copy that closely mimicked those written by the in-house team. This consistency across online and distributed content strengthened and reinforced each subsidiary's brand identity and messaging.
Furthermore, our collaborative workspace allowed for several departments to work closely together on the same projects. This meant that the product and content marketing teams could seamlessly collaborate on a single platform to develop on-brand product copy and messaging.