Due to growing cost pressure and the difficulty of finding suitable staff, more and more companies are turning to automation in service. It is often not clear where to start. A study provides insights into this.
The automation of customer interfaces does not always run entirely smoothly and is sometimes very poorly received by customers. Nevertheless, we already know many examples of successfully automated processes from everyday life, which are also perceived very positively by all those involved. For example, self-check-in has long been available at airports, which simplifies processes for passengers and in turn saves providers money.
A classic form of automation in the After-Sales however, represents the Implementation of self-service solutions which has been described elsewhere.
In which areas do projects make sense?
At this point, we want to ask ourselves which processes can (or should) be automated under which circumstances and which are better not. To this end, researchers Andreassen, van Oest and Lervik-Olsen have developed a model that can be used to determine the optimum degree of automation for an activity.*
High-quality services should rather be provided by people
The researchers define three quality dimensions: Relationship quality, process quality and outcome quality. In the case of relationship quality, the focus is naturally on the human being as the contact person. Human communication partners are preferred, especially if the customer has high expectations of the quality of a service.
However, this can of course change if process and result quality can be improved through automation. However, you should only automate services with high quality requirements on the part of the customer if the implementation of your solution offers significant added value in terms of process and result quality. This is because the dimension of relationship quality is given a very high weighting by the customer, especially in the case of demanding services.
The uncertainty factor plays an increasingly smaller role over time
As is often the case with the introduction of new technologies, there is often an initial defensive attitude towards the innovation when it comes to automating customer interfaces. This uncertainty is initially reflected in the acceptance of the service.
The study shows that this initial reluctance makes it considerably more difficult to sell an automated service at the beginning. However, it was shown that this effect diminishes over time and no longer exists once the new technology has become established.
The topic Building trust in the age of the Internet of Things has already been highlighted elsewhere in the ServiceLobby.
The role of price sensitivity in service automation
The customer normally assumes that automated services are cheaper, as their provision by the provider is generally associated with lower costs. Nevertheless, according to the researchers, there are scenarios in which a higher price can also be charged for automated services.
This works if customers have very high expectations of the quality of a service and low price sensitivity. If this is the case and the automation also enables a significantly better and faster service to be provided than the service previously provided by humans, you can definitely achieve a price premium compared to the non-automated service.
Recommendations for action for service automation:
- Always think long-termEven if automation solutions usually suffer from low acceptance at the beginning, the costs of provision can often be greatly reduced. As long as the quality of your offer is right, a habituation effect will occur and sales figures will normalize again, ultimately allowing you to achieve an acceptable plus. If the cost of sales is only slightly lower, automation should be reconsidered, as it may not be possible to compensate for the short-term slump in the long term and the investment will not be recouped.
- Offer incentives to buyIn order to mitigate the effects of the initial lean period after the introduction of an automated service, you can think about creating additional purchase incentives in good time. For example, credit notes on related products or services from the existing portfolio or a starter discount would be worth considering. These incentives should then be reduced over time and, above all, communicated as a temporary discount so that you don't ruin the price in the long term. Otherwise, the customer will expect to receive the discounted service permanently and the business case will not pay off for you
- Evaluate the right automation leversIf the quality of the automated service only marginally exceeds that of the service provided by humans or, in the worst case, is even lower or just as good, you may ruin your business. The uncertainty of the customer and the high price sensitivity of automated services will ensure that your project does not pay off. Demand will fall drastically. You can only compensate for this if automation is many times cheaper than manual provision. In most cases, however, this is not the case to the extent required. However, if you have a solution whose quality clearly exceeds that of manual provision, you should definitely move towards automation.
Source: *Andreassen, T. W., van Oest, R. D., Lervik-Olsen, L.: Customer Inconvenience and Price Compensation: A Multiperiod Approach to Labor-Automation Trade-Offs in Services, 2018
