In Part 1, we showcased five areas in which artificial intelligence was having a major impact on restaurant ops. These included sales demand forecasting, delivery driver co-ordination and kitchen robotics.
For Part 2, our focus shifts to facial recognition, menu optimisation, equipment maintenance, customer feedback and food photography.
The benefits of fine-tuning your menu are numerous and far-reaching. From improving profitability and reducing food waste, to boosting customer satisfaction and streamlining kitchen operations, a well-curated menu can be an absolute game-changer.
Menu optimisation isn’t the easiest of processes though. There are a number of moving parts to consider such as cost per serving calculations, demand fluctuation, customer preferences, external events, costs of goods sold analysis and so on.
If you run a busy operation or don’t have access to proper data, the task becomes extremely difficult if not impossible. To overcome this long-standing industry problem, restaurant owners have now started to rely on artificial intelligence to do the legwork for them.
With the use of machine learning, AI can sift through vast amounts of data to identify patterns and trends, while predicting customer behaviour based on past purchases, demographics and location. Some of the more advanced AI menu management systems also incorporate ABC/XYZ analysis into their calculations.
Without getting too technical (the subject requires its very own article) the ABC/XYZ model is loosely based on the 80/20 Pareto principle which asserts that 80% of sales can be attributed to 20% of products.
Thus, it’s possible for managers to prioritise the 20% of menu items that account for the most revenue, while disregarding those that yield little. And according to the XYZ method, items can be organised according to their sales performance over a set period of time, helping to accurately measure variability of demand.
Don’t assume that most restaurants have got things covered when it comes to menu engineering. Many don’t. In fact, it’s an area that’s commonly overlooked by owners. So in essence, having this kind of data to hand is akin to being in possession of gold dust.
A restaurant lives and dies by the reliability of its equipment. To ensure consistent operability, a traditional approach in the F&B industry is to use a combination of regular inspections and preventative maintenance, backed up by staff checks, cleaning and record-keeping.
Unfortunately, maintenance sometimes becomes reactive, even with effective practices in place. The resulting disruption can prove very expensive.
To prevent downtime, many businesses are turning to predictive AI. And there's a raft of solutions on the market that are proving increasingly popular among F&B operators, among them Sensemore and Workpulse.
Typically, these systems use AI to analyse data from a variety of sources including maintenance histories, performance logs and energy consumption. From this data, they’re able to identify and track patterns to help predict when equipment will likely fail. This effectively enables owners to be proactive in their maintenance practices, thus avoiding equipment downtime and improving overall efficiency.
Facial recognition (FR) has proliferated in a big way over the past few years. Primarily used for security checks and log-in processes, this technology has become commonplace in our every-day lives. And now, more personalised solutions are being introduced, some of which work well in the F&B sector.
Biometric specialist, Pop ID has developed a series of customer-facing POS and kiosk terminals that include embedded cameras. With the use of FR, these terminals can provide personalised ordering experiences for customers based on past orders. They can also automatically sign customers up to loyalty programs while also taking cardless payments.
The benefits of FR terminals are fairly obvious. In addition to simplifying the ordering processes, transaction times are faster which, according to POP ID’s owners, can result in increased sales of up to 4%.
The only downside here is the continued resistance to FR among customers concerned about the sharing of personal information. Because of this, facial recognition is highly regulated in the UK by the GDPR. So it remains to be seen just how disruptive this technology will become.
For restaurants that choose to rely on photos to showcase their dishes, be it on a menu or promotional materials, the visual representations need to be appealing. Sadly, a lot of restaurants and food outlets fail quite miserably in this area.
Due to issues such as poor lighting, the use of smartphones instead of cameras (professional photographers are expensive) and a lack of styling, the photographs don’t come across as being especially appetising.
AI-powered tools such as Adobe Firefly, DALL-E 3 and Mid Journey help F&B operators to get around this problem. Of these, Mid Journey is probably the best of the bunch, featuring an impressive suite of tools that can enhance existing photos, create idealised versions of a photograph, change backgrounds and even simulate professional techniques.
To use something like Mid Journey, you merely enter a text prompt describing the kind of image you’d like, say ‘a professional photograph of a juicy hamburger on a dark background’. An image is then outputted that will, in all likelihood, precisely match your request.
For time-limited restaurant owners unable to afford professional photographers, these tools provide an increasingly powerful solution. The results are already astounding to say the least.
That concludes our two part series covering AI's impact on restaurant operations. Given the speed in which AI is taking hold of the F&B sector, we may need to write a part 3 in the coming months.
Reputation management is now a real challenge, no matter what industry you’re in. With a multitude of social media channels to choose from, customers have endless options for leaving reviews about your brand.
Manually keeping track of this feedback can be understandably daunting. But to protect and enhance your digital reputation, it’s essential to react regardless of whether the feedback is positive or negative.
Social listening tools are the most obvious solution here. And there are plenty on the market such as Brand24, Mentionlytics and Hubspot. These tools scan blogs and social media channels, tracking brand names, keywords, phrases and hashtags.
Some listening tools have also started to use natural language processing to analyse social media conversations for intent and sentiment. So if a conversation about your brand starts to become negative, the system alerts you, allowing you to take action. It’s an effective way to ensure your brand doesn’t get torched by disgruntled customers or malicious trolls.
Thus, it becomes much easier for businesses to identify specific pain-points. The insights gleaned can prove invaluable in driving informed decisions about customer experience and brand development.
And that concludes our two part series on the disruptive power of AI. Given the speed at which machine learning is permeating the food and beverage sector, you can expect part 3 in the coming months. So be sure to check back!