A Maxymiser white paper.
After years of debate, the era of truly synergistic multi-channel retailing has arrived. While Wal-Mart plans to use its 10,000 retail stores to fulfil online orders, following the Amazon Locker model, Amazon is reported to be opening its own bricks and mortar stores or creating partnerships with existing retailers to respond to customer demand for support and direct access to goods. Add in the rapid evolution of mobile retail from research to transaction and the concept of the totally integrated retailer is being realized.
With in-store, online and mobile interactions now affecting consumer behaviour across every aspect of the business, it has become essential to understand multi-channel performance. Online testing is no longer just about improving online conversion; it is about tracking delivery strategies and exploiting customer insight to enable behavioural targeting.
Maxymiser believes it is essential for retailers to consider not only optimizing the web site but also the role of online changes in optimizing business processes in a joined up online and offline world.
Over the past two years online optimization techniques have transformed conversion rates. From improving the check-out process to stripping back content, retailers have exploited the ability to present dynamically multiple variants of content, design and navigation to determine the optimal combinations for increased conversion rates.
Testing is now set to get far more complex as retailers look to explore the influence of online changes on every aspect of the business. From transforming delivery options to behavioural targeting, the changes made online can have a dramatic impact on every aspect of the business – from logistics to in store behaviour.
Retailers need now to build on the proven bedrock of improving conversion rates to attain more insight into the way new strategies, such as the introduction of Click & Collect or Product Ratings, affect consumer behaviour both online and in store. This information is no longer the preserve of the marketing department, but has real value to a raft of business decision makers.
One of the most marked shifts in online retail over the past 18 months has been the emphasis on customer delivery and fulfilment options, not least the huge growth in Click & Collect and the increasing interest in Collect Plus where goods are delivered to a nearby hub, such as newsagent.
The option to buy online and collect in store is compelling for a number of different reasons.
- Provides consumer choice and avoids the need to wait for deliveries – or rely on a friendly neighbour.
- Overcomes the problems associated with failed deliveries, which cost the industry £851m in 2012 according to IMRF.
- Drives the consumer back into the high street store where experienced sales staff can up- and cross sell products, leveraging the existing retail estate and the skills of sales staff.
Click & Collect also fits into the new multi-channel model by supporting the evolution from siloed multi-channel strategies towards proactively driving business between channels to maximize overall, rather than channel specific, revenue.
There are significant opportunities to differentiate the brand based on a flexible, timely and efficient fulfilment model. Retailers can improve conversion by optimizing the way these options are presented to the consumer. But this is a fast changing business model, with consumer expectations changing in line with new delivery choices. While some retailers now boast 70% click and collect rates, others – especially those with smaller products – are achieving just 30%. And while growing numbers are experimenting with Collect Plus, this is not a model that works well for all brands.
The first step for any retailer must be to optimise the way delivery options are presented to the consumer. Raising the profile of Click & Collect on the site, ensuring customers can rapidly identify the nearest store and understand availability options can have a significant impact on uptake. But changes to any delivery option should not be considered in isolation.
- Test market elasticity by varying the minimum price at which free shipping is applied.
- Ascertain the threshold which prompts the largest uplift in customer order value.
- Understand how Click & Collect performs in comparison to free delivery.
- What is the impact on Click & Collect uptake if the free delivery threshold is changed?
- How do different delivery models affect the returns profile? Are individuals more likely to return goods in store if they Click & Collect – but will they also opt to try on and buy something else?
Testing is the only way a retailer can truly gain insight into how customers react in real time to changes in delivery and fulfilment options.
Fulfilment optimisation is clearly far more complex than ensuring the check-out process is effective, or product page correctly presented. Changing the free shipping threshold, for example, will have an impact on the delivery demand, impacting the logistics processes. It is therefore essential to carry out tests in conjunction with the logistics team to ensure any spike in demand is expected and managed; whilst tests must also extend throughout the entire buying lifecycle, including returns, to understand true conversion costs.
It is also important to ensure in-store staff are geared up to respond to a push on Click & Collect, for example, from managing returns to proactively selling additional items.
With far more variables that can influence customer behavior and affect the cost base, organizations need to combine multiple measures. It is in depth analysis of delivery costs, the in-store time spent managing collections, as well as online conversions that will provide true insight into the cost/value equation of different fulfilment options.
Ratings & Review
It is also important to reflect the growing external influence on consumer behaviour. From comparison sites to Facebook likes, consumers are becoming savvy when it comes to assessing the value of a new purchase. They don’t simply want more product information: they want the feedback of peers and the experiences of previous purchasers. From the fit of this season’s trousers to the real battery life of a new tablet, consumers are increasingly influenced by external factors.
However, assuming ratings and reviews will give an uplift is a mistake: the Amazon review model does not work on every site. To maximize uplift and avoid an expensive mistake it is essential to ascertain via testing just how important reviews are to each organisation’s customers.
The way these reviews and ratings affect purchasing behaviour varies dramatically based on a number of factors, from consumer demographics and product type/value to the position of the rating/review on the site and, of course, the quality of the reviews.
- Is it better to offer a long Amazon style list of reviews on each product page or a single average product rating star on the product list page?
- Is the latest, the average or the most useful review that provides the best uplift?
- Is there the critical mass of reviewing customers required to give a credible product rating?
- Or is it simply better to let customers use the raft of independent review sites that are available rather than offer a rating/review option?
The response to different fulfilment and delivery options as well as ratings and reviews will vary considerably, from consumer profiles to product mix. To maximize performance it is therefore essential to test consumer response to different delivery options. There are multiple approaches to this:
A hypothesis based approach enables retailers to discuss, consider and then test specific hypotheses associated with customer behaviour. Using a customer’s previous online activity, from products viewed to goods purchased or abandonment during the check-out process, the organization can test an array of approaches to assess how best to increase customer conversion rates and improve customer value.
Changes in content can range from the subtle, such as raising the prominence of the check-out button, to the blatant – presenting the customer with a detailed list of items previously put into a basket but not purchased. This hypothesis based model enables organizations to gain insight into trends in behaviour and the response of specific segments to different targeting models.
Alternatively, using mathematical models, the process can be automated to deliver continually enhanced content to customers. The behavioural targeting tool can measure visitor behaviours and responses to content and offers. It rapidly evolves and adapts the content and offers to consumer interactions to drive the highest engagement, conversion rates and revenue.
As the online experience evolves, it will become increasingly important to track the behaviour of customers across a raft of additional areas – not just product browsing and purchase. Issues such as the use of ratings and reviews and the response to delivery options will also provide valuable insight that can be used to improve the relevance of both content and offer.
For example, responses to ratings and reviews should be assessed for gender or demographic difference in response; whilst testing the value of ratings and reviews on different customer segments, based on demographics or previous behaviour, can fine tune the strategy.
Indeed, there are opportunities for organisations such as Wal-Mart that can already boast a deep customer understanding, from order history to online behaviour, to extend this model. Using both historical data and current data, a retailer can present the right content to the right people – effectively becoming an invisible personal shopper. This will require the use of content and data in real time, as the customer browses the online store, to predict what will shape each customer’s experience right now.
The growing overlap between online and offline business activity creates the need for new performance insight. Customers now expect a consistent experience from the time of order to the time of pickup, and every other touch point along the way including customer service. Tracking and testing the performance is essential to understand the changing influences on performance.
This model also provides a retailer with the opportunity to personalize not only each individual’s online experience but improve each subsequent experience online and in store. The key is to automate this process by using the same model of targeting online to inform the content presented to each customer across every other channel: the visitor profile is at the centre of the cross-channel experience. Exploiting big data through effective testing and behavioural targeting across every channel will transform the consumer experience and retail performance.
To optimise this complex delivery, fulfilment and consumer experience model online retailers need to create the big picture. Combining existing offline information resources with online testing into a single dashboard can provide clear insight into the way different aspects of the business are affected by online optimisation strategies:
- How will changing logistics costs alter the overall cost base?
- What is the impact of raising the free shipping threshold on to the impact on sales and acquisition costs?
- Are consumers more interested in Collect Plus type services today than 12 months ago?
- More consumers are now buying via mobile – if they can look at a product in store and order cheaper online for immediate or next day delivery what is the implication for the existing delivery model?
- How can behavioural targeting be used to present different consumer segments with an array of delivery options – perhaps based on previous behaviour?
Again, without testing it is impossible to prove the impact of changing delivery option on the company’s strategic direction.
Proving the Business Case
With such radical shifts in customer service experience now becoming possible, the pressure to get the right delivery and fulfilment processes in place is increasing fast. The ability to test new processes and delivery options in real time is becoming increasingly key to tracking the development of new logistics models and maximising the investment in new these services.
It is clear that many factors influence consumer behaviour. But hypothesising is not enough. Assumptions need to be tested; and insight needs to feed back into an increasingly complex big data analytics model to reflect the influence of strategic change on every aspect of the business. It is only by testing new ideas and analysing the impact cross-channel that retailers can effectively master the complexity of a completely integrated strategy.