Improving Price Perception

Improving price perception of deli items

Role: Lead UX Designer

Duration: 2-3 months

The problem

"How might we showcase the prices of deli items (e.g. cheese, meats) online to improve customer price perception and help bridge the gap between online and in-store customer mental models around shopping deli items?"

The commercial team at Woolworths engaged my squad to lead a discovery piece to improve value perception of fresh produce sold online at Woolworths. This stream of work involved investigating how we can improve price perception of deli items (any items sold behind the counter at store) online and improve sales. I was the design lead managing this stream of work alongside other initiatives. The work required collaboration with a multitude of stakeholders (5+ squads, legal, commercial teams). The outcome was an improved product tile experience that better showcases the price of deli items online.

Understanding the problem

Background

Woolworths was losing key market share across fresh produce to Australian supermarket competitors such as Coles
Market research analysis revealed root cause of market shift was change in value perception driven by factors such as inflation and customers seeking more value from their shop
Currently online pricing of deli items makes Woolworths appear more expensive than retailers such as Coles. Woolworths displays "maximum weight pricing," while other retailers such as Coles display "minimum weight pricing," for equivalent deli items. This difference in display is driven by historical reasons in how Woolworths charges and sells weighted items online.

When customers order deli items online, in-store pickers strive to match requested quantities, but variable weights may lead to differences. For example, a customer may order 5kg of chicken breast, but the final amount picked in-store could be more or less depending on the variable size of chicken breast pieces. To address this, Woolworths displays "maximum amount and price" online, and always refunds the difference. This ensures customers won't be overcharged based on the final pick.
However, displaying max prices can make Woolworths deli products seem more expensive than competitors, deterring customers from purchasing items. This is often because customers don't read the "unit price" and are used to buying deli quantities in smaller amounts (100g vs 1kg of salami).
Estimated value of improvements in price perception was estimated to generate upto $12M revenue

Our key goals:

Improve sales of critical deli lines through improvements in price perception
Find solutions that can be incrementally tested and deliver value (low effort, high value)
Ensure solution works across end-to-end journey (product tile changes impact 15+ squads in terms of touchpoints across the whole end-to-end shopping journey. Our solution needs to account for impacts across web and app journeys)

1. The difference in showcasing price (max price vs min price) is creating a price perception problem with customers

2. Adding messaging to indicate price is "max price" will help mitigate any price perception problems

3. Changing the price display to showcase "min price" will help improve price perception with customers

Our hypotheses

How we did it

Iterative design process

Given the timeframe of the project, I made the decision to start testing the hypotheses early with experiements before investing too much time into the discovery process. After this point, I went through a typical research, design and development cycle.

Before going into deep research and discovery, I wanted to test the two hypotheses below through experimentation.

1. The difference in showcasing price (max price vs min price) is creating a price perception problem with customers

2. Adding messaging to indicate price is "max price" will help mitigate any price perception problems

Google Experiment

Collaborated with the Google Ads team to change the description on products displayed to include price/KG to try and reduce sticker shock value of max pricing.
Resulted in 37% uplift in clicks, 28.7% uplift in impressions and improved click rates
These improvements were seen only after 2 weeks
In addition, the revenue and units sold increased compared to the year before, with close to $5000 increase in sales

CRO A/B Test

Use A/B testing to see if adding the description "price based on max weight" will help reduce sticker shock value of items and increase add to cart
Resulted in slight increase in cart additions, but not at high confidence levels
There was also a higher save to list action, an increase in orders, revenue and AOV for the selected random weight products

Experimentation

Learnings

Both experiments demonstrated that changing the way pricing was displayed to communicate max pricing improved add to cart and sales even over a short period of time. This provided confidence to continue investigation into the issue further and conduct proper discovery and research into the best solution

I wanted to understand what competitors were doing both from a local and international perspective. I conducted a holistic competitor review across 15 ecommerce stores to understand trends and best practice. The competitor analysis supported various workstreams, requiring a broad review scope. The work was completed in collaboration with another senior UX designer, concurrently working on a parallel stream of work.

Key trends:

Most supermarkets use minimum weight and price on product tiles
Most product tiles include messaging that indicates price is variable "final cost by weight"

Competitor research

Based on competitor research and seeing that the majority of retailers employ minimum weight and price, I wanted to test similar concepts with Woolworths customers. The primary objective was to understand their comprehension of the final price paid and the final amount provided using minimum weight. Additionally, the testing aimed to gauge preferences between the current and future state experience. There are different types of deli items sold behind the counter so we also wanted to test different scenarios (sliced items such as ham, piece items such as chicken).

Testing round 1

Hypotheses:

Customers understand how deli meat (i.e. chicken, ham) will be sold and charged by weight
Customers prefer to see lowest weight on deli products

Task:

We used a prototype with chicken and salami as the items for purchase. Customers were asked to pretend they were purchasing these deli items behind the counter at specific amounts. They were asked questions to check comprehension around price paid and amount received. We also asked open ended questions to obtain insights into preference between current and future state.

For the first iteration of the design, I experimented with the format of putting the minimum weight next to the key price on the product tile during the browsing stage and removing it in cart. I made this decision as at the add to cart stage, customers will have entered their purchase amount and will be seeing the real-time calculation of the item. At this point in the journey, the customer no longer requires the minimum weight and price information as the price dynamically changes to match the quantities entered. In the browsing stage, customers will need to calculate the final amount manually as the pricing information displayed is static, not dynamic.

Qualitative research

Results:

5/5 customers understood how much they would be charged and understood final quantities provided
All customers preferred the new design over current state
For both items, customers preferred the minimum weight amount to sit below the unit price ($/KG) to reduce cognitive information overload
In particular for salami, customers noted that shopping by minimum weight was more realistic to how they would purchase the item in-store (i.e. purchasing 100g salami more realistic than purchasing 1kg salami). Aligning the in-store and online mental models helped customers better understand how much they would be charged

Testing round 2

While customers indicated wanting to see minimum weight displayed under the unit price, this created technical complications for the product tile. This included:

Product tiles have been developed with fixed information to be turned on/off. Addition of new slot for minimum weight creates high development effort (need to account for over 20+ variations of tile and ensure it doesn't break across multiple touchpoints across the whole journey in app and web)
Product tiles are already tall in height and take up a lot of space in key journey touchpoints, adding extra line underneath will increase height of the tile and impact downstream touchpoints

Based on these challenges, I collaborated with the UI designer to experiment with different designs of the initial design (changing weight, font, style, placement) to reduce cognitive overload. The final two versions are below. We shared both designs with developers and there was a strong preference for version 2 due to greatly reduced technical efforts. Due to this strong preference, we repeated testing with version 2. At the end of the test we completed a preference test between version 1 and verison 2.

Results:

5/5 customers understood how much they will pay and be charged
5/5 customers preferred version 2 over version 1
Customers noted that with chicken items it was necessary to include the description "min.250g" as it is not always clear what the min. weight of a chicken piece is. They didn't deem this information necessary for sliced meats like ham

With testing producing a strong weighting towards version 2 and clear comprehension from customers. We presented the findings to senior leadership and proceeded to hi-fidelity design and development

The outcome

Hi-Fi design and development

The process of finalising hi-fidelity designs required additional stakeholder management and analysis.

Legal consultation

Woolworths has quite strict regulations and rules around how pricing is displayed on products. In our current state, when product descriptions exceed a certain character length, we truncate the description. Legal asked us to investigate whether we can truncate descriptions in the middle and keep the "min X weight" description visible. Due to a large amount of tech requirements to complete this, I conducted a character count analysis (1800 SKUs) to understand the impact of truncation across different scenarios. After showing conducting the analysis, I concluded the impact was on a very low percentage of items. I shared the findings with legal and they were comfortable with us proceeding with our initial solution without needing to truncate in the middle of the description

E2E Impact analysis

The product tile is a component that is complex due to the many variations that exist across the end-to-end journey in web and app. As part of parallel initiatives I was able to map out how changes on one tile would propagate across the journey using a service blueprint, this output helped with stakeholder collaboration and end-to-end design.

Hi-fidelity design

Together with my UI designer, we designed the final version of the product tile across all touchpoints and channels, taking into account different scenarios (e.g. promotions and needing to stack other information on the tiles). I collaborated with stakeholders across all areas to get final approval for development. We worked with our developers to help assist in selecting the best technical strategy for development. I worked with analytics and commercial teams to set up metrics for success and what data we need to track once the designs are launched

Next steps and reflections

Development of the final designs are due this quarter for release. The biggest learnings from this project are that sometimes small changes can have unexpectedly better experiences for customers. Initially coming into this project, I was skeptical about the impact of changing pricing display on shifting customer value perception. In the process of completing the research and discovery, it was rewarding to discover a solution that not only benefitted commercial teams but also better served customers and aligned to their mental models of how they prefer to shop deli items online and in-store.