How to Remove the Guesswork from Trend Forecasting

How to Remove the Guesswork from Trend Forecasting

Written by Phung Yi JunAugust 22, 2020

How to Remove the Guesswork from Trend Forecasting

Before Covid-19 upended the world, the industry often resorted to the same fashion trend analysis methods to gauge market demand. Trend forecasting services like WGSN are typically used to guide the fashion design and buying processes, leading to the homogenised state of the fashion landscape. 

There are also other common methods used, such as…

  • Historical sales performance
  • Fashion runways 
  • Social media

But the pandemic has impacted how we operate in unprecedented ways. 

Trend forecasting reports or services are aspirational-focused, with little geographical relevance. We benchmarked the top trends forecasted for S/S 2020 by WGSN against data – and found that some trends failed to translate in the context of local markets. 

When it comes to trend adoption, consumer readiness is a huge factor, which is lacking in these reports.

Historical sales is not a reliable source in today’s climate either, as it is no longer indicative of future performance. Runway styles were once prominent, but are now risky if they fail to translate to commercial pieces.

And for the retailers that adopt social media trends, they’d find themselves at risk of inventory glut as likes do not equate to sales numbers. 

In today’s changing climate, how should merchandisers, buyers and designers perform more predictable trend forecasting? 

Validate, Validate, Validate

To meet demands, it’s important to distinguish promising trends and short-lived fads – especially as the latter is increasingly common in a world powered by social engagement.

What’s more, consumer shifts are ever-changing, which means having a firmer grasp of a trends’ lifecycle is all the more crucial. 

Data validation is key. By monitoring and validating trend movements, you can create a commercial assortment, minimising guesswork and increasing confidence in decision making. 

Layering trend forecasting with data analytics reflects true demand, no matter if it’s social media trends, runway reports or forecasting services.

A Data-Driven Approach to Trend Forecasting

Cottagecore is the next mega-trend propelled by the pandemic. The most popular example is Lirika Motoshi’s “strawberry dress” which became a viral sensation on TikTok.

Lirika Motoshi’s “Strawberry Dress”. Credits: @lirika.matoshi

The demand for strawberry-themed products began rising in March and spiked again in June as the dress became increasingly popular across social media.


Source: 456 data points were analysed on products retailing online from fast fashion retailers in the US, UK and Europe from January to August 2020, as tracked on the Omnilytics Dashboard.

Fashion publications such as Refinery29 too recently reported on popular summer dresses – which featured the same strawberry dress.

But the real question remains: does this reflect true demand? Should retailers invest in this trend or look into other summer styles?

The need for visibility is urgent – here’s how to conduct accurate fashion trend analysis. 

Fashion market insights platform, Omnilytics, surfaces trends from 44 different countries across 400 retailers by category, colour, pattern and material.

Identifying Trends

The first step here is to have a high-level overview of the dresses in the market. By filtering the top brands and retailers for dresses in the UK, you can easily identify uprising subcategories.

Uptrending and downtrending dresses, as tracked on the Omnilytics Dashboard.

Across the board, only a few subcategories are trending, such as A-Line Dresses and Swing & Trapeze Dresses. Interestingly, the latter was forecasted by WGSN for 2020, with the theme for womenswear being ‘Modern Femininity’. 

Retailers like Farfetch and Nordstrom aggressively brought in new stock last year, but was met with an underwhelming response. 

Now, the trend is picking up once more, in-line with the current demand for loose dressing. What’s more, with its small contribution, it signals a gap in supply.

Uncover Emerging Patterns

Looking at the trend’s trajectory, Swing & Trapeze Dresses saw a few spikes throughout the year, and exceeded the total category’s trend line from May to August 2020, proving its strong popularity. 

Trend performance for Swing & Trapeze Dresses, as tracked on the Omnilytics Dashboard.

Here, you can also spot key patterns, colours and materials of the subcategory Swing & Trapeze Dress. Florals held the highest percentage of top patterns, followed by animal prints and dots. 

But even after identifying Swing & Trapeze Dress as a trend, it’s important to have a complete overview of demand.

Validating Trends

A crucial element to trend validation is to analyse its trade performance. Did it sell out at full price – or mostly at a discounted price? What’s the full price contribution? How often is it replenished?

Based on the sections above, we’ve identified Swing & Trapeze Dresses as a trend. But is the category really worth investing in – or is it just a short-lived fad?

Trade analysis for Dresses, as tracked on the Omnilytics Dashboard.

Here, we can see that while the category Swing & Trapeze Dresses did not have the biggest contribution, it performed the best at full price, exceeding the category average of 30.74%.

Its Total Sell Out at Discounted Price was also below the average, proving its popularity of this trend. In other words, there’s a demand for Swing & Trapeze dresses.

Another indication that this is not a short-lived fad is also in its price positioning – it had the highest median price at $36.81, proof that consumers are more than willing to pay for this style. 

Trending Patterns and Colour Performances

Next up, we need to look for design and buying inspirations – what patterns and colours are the best-performing for Swing & Trapeze Dresses?

Top patterns for Swing & Trapeze Dresses, as tracked on the Omnilytics Dashboard.

To identify the bestselling patterns, layering the sell-out contribution against the total number of SKUs gives a more comprehensive view.

As depicted by the yellow line, florals had the strongest sell-out, and had the highest contribution for all retailers (with the exception of Missguided). The pattern is also in-line with summer, a key staple for the season.

There are opportunities for polka dots and stripes too, with a higher sell-out contribution than the product contribution – which signals a gap in the market.

Top colours for Swing & Trapeze Dresses, as tracked on the Omnilytics Dashboard.

As for fashion colours, blue led with both a strong product and sell-out contribution, at 15.50% and 15.46% respectively. This validates Pantone’s announcement of Classic Blue as 2020 Colour of the Year.  

Other summer hues, such as green and yellow, showed potential too, given its high sell-out contribution. 

Changing the Current Climate 

Trends do emerge from runways, social media or even services like WGSN, but building an assortment by relying solely on what’s trending, without validation, leads to a negative impact on cash flow – especially with consumers’ changing demands. 

Validation is key. Having a data-driven approach to your fashion trend analysis not only adds confidence in your decision-making but also increases accuracy in meeting demand. 

Receive Weekly Industry Roundups

Sign up to receive the latest industry insights, news and weekly updates delivered straight to your inbox.

You might also like

Navigate the crisis with Omnilytics

Learn how our fashion retail clients make better decisions with ecommerce analytics