Artificial intelligence for small business: nonsense or reality?

Helen Curash for

When it comes to artificial intelligence many people imagine large and massive systems, supercomputers as IBM Watson or huge science centers. But in reality AI has been available for solving many routine tasks for a long time already. Today we are going to tell you about how neural networks can be used by, for instance ….. a bakery. As they say, why not?

What could be more obvious than a small business, maybe even a family business, producing for local consumers? To start working you need to obtain the status of entrepreneur and to hire several people. What AI are we talking about? However, the experience proves the opposite, as AI is capable of improving many processes even for a small business.

What do neural networks have to do with small business?

Artificial intelligence is applicable in any field dealing with a significant amount of data. No matter if the business realizes that or not, the data accumulates from the first working day and every little detail helps in machine learning and can show AI how to improve your business processes.

Speaking of a bakery, this micro-business is able to collect the following data:

- possibilities to produce different types of bakery products;

- sales volumes depending on the time of day and on days of the week;

- the impact of holidays and weekends on demand and sales;

- tracking of customers preferences depending on the season;

- flour and other ingredients prices fluctuations;

- staff performance and productivity depending on the shift duration.

If you implement any loyalty system, for instance, offer a discount for a registered phone number or offer loyalty cards (still works despite being an archaism), you receive even more data:

- Individual purchase story tracking;

- individual preferences of the customers;

- response on different promos and discounts;

- customers’ evaluation of new items on the menu.

Thus, any small business has enough data to implement AI technologies. And even if you haven’t been collecting the data for many years there’s no need to be upset. You can start today — the time is now! As soon as you launch the data collection process AI will get the basis for your business optimization.

Customized offers for the clients

Based on customers’ preferences analysis, their purchasing history, participation in promo actions you are able to come up with the most appealing offer for each customer. For instance, it may turn out that a retiree Mr. Ivanov will be happy to buy an extra poppy seed bun every day if you offer him a 5% discount, and Mrs. Sidorova will buy brown bread twice as often if it comes in a set with white bread. It is absolutely impossible to check all these hypotheses manually, but AI is not only capable of creating possible offers (within a given framework) but also of choosing the right target customers for them.

The best will be to have followers in social networks and distribute discounts or announce the giveaways there. In that case AI will be able to analyze your customers’ profiles and draw conclusions about the behaviour of each of them. Thus, you’re going to have special promos for birthdays and other personal holidays, as well as pastries selected especially for the customer according to their mood.

Return the customers who have gone

It is extremely useful for a small business to know their clients, at least register their telephone numbers. If a customer’s started to buy less often or doesn’t come by anymore at all, AI will come up with some ideas about how to make this customer come back.

Having additional data from social networks also helps AI understand why that has happened. Maybe the person is sick so we could offer delivery. Maybe the customer has moved and then there is no point in bombarding him with texts via telephone or instant messengers.

Reviews processing

Today any business has its reputation. Customers are willing to give feedback using various platforms, sending messages directly to the management and posting in social networks. Mobifitness isn’t a bakery of course but the point remains the same. They’ve used hundreds of real customers’ reviews to teach the neural network, as a result AI is now capable of evaluating the tone of the review and sorting out the most important ones.

AI can assign emotional rating to the review, as well as assess the aggression level of negative comments. It is useful for businesses of any size, as it is not easy to respond to the reviews not only when the business is growing but also when you have limited staff. Neural networks approach saves time of the employees and takes care of the company’s online reputation even if you have neither time nor energy to do it yourself.

Production optimization

Even if you don’t have many employees their productivity depends directly on many factors. Thus, someone remains productive for 5 hours while others can work efficiently for 8 hours. Someone is better at baking rolls while the other one’s specials are cabbage pies. And you can determine that by analyzing the sales. Based on this data AI is able to create optimal work schedules for you to do more, spend less and cook better.

Development strategies elaboration

Every general manager sooner or later wonders “What are the possible

directions for my business development?” Should you bake more sweet rolls? Should you start baking rye bread? Wouldn’t it be better to go for delivery or to open another bakery somewhere else in town? AI is able to analyze all the data the company has as well as open source available data about your town and come up with some business development proposals. Moreover, you can get the updates on a daily basis, as well as the percentage rate evaluation of potential success level of every possible innovation. That makes planning your business a lot easier.

Supply optimization

Most likely you may work with various goods and services suppliers. For instance, you can purchase different flour and yeast, you can work with different shipping companies. But usually a small business has a contract with the same trusted supplier and that leads to a considerable margin lost.

Thus, it may turn out that today the flour of the “” brand (the name is invented, any resemblance to reality is pure coincidence) is cheaper from one supplier and the next day a new distributor is going to offer a discount for a batch over 500 kilograms.

No CEO is able to analyze all that manually while AI is capable to propose new procurement options for various categories of goods on a daily basis.

Helen Curash for

Loyalty programs implementation

Loyalty program is a positive thing. Although if not customized it is not going to be very efficient. In order to make the loyalty program work it is necessary to interact with each client personally. Only AI can help implement it, tracking the customer’s preferences and building up his loyalty by creating customized mailings, sending greetings, offers and simply pleasant content.

Search for clients in social networks

Finally, you can find new clients not only in the nearest neighborhood, but also in social networks. To do this, you need to create your own group, post pictures of beautiful bread and satisfied customers (now everyone loves Instagram), and invite new followers. AI ​​will also help you to find them. Analyzing the data of open profiles, neural network algorithms come up with recommendations, and there is just one thing left to do — the entrepreneur should send out the invitations.

Automated orders processing

If you receive a lot of orders via telephone the AI automation would let you save a lot of time and money. AI is capable of answering hundreds and thousands calls simultaneously so you wouldn’t miss a single order. Essentially all customers’ calls can be reduced to standard cases: orders registration, complaints in case of delivery delay, requests to change payment method or delivery method, orders cancellations, and, finally, complaints about the staff.

Voice AI is already highly developed and can easily define the tone of the request, the purpose of the call and is capable of processing a major part of the incoming calls, letting the operators deal with non standard requests. That system has been implemented by “Dodo pizza” which automated the processing of 250 thousand calls. That led to 500 000 rubles of cost cuts per month as AI managed to determine the reason of 67% of calls and successfully processed 97% of those.

Is that all available?

People keep asking us about how a small business can find enough money for AI implementation. Nowadays it’s not obligatory at all! The existing algorithms successfully tested in other fields can be applied to new areas without any extra cost. Of course, you can order a new solution elaborated “from scratch” and then implement it in your business, or the bakery owner can use AI in form of SaaS, i.e. to set up data transfer to the partner’s cloud in order to receive ready-made forecasts and analytics, paying for the access to the service.

Here in SmartCoders we went even further so that we offer our clients to start using AI without any initial payment! We just agree that some part of the profit received after AI implementation or a percentage of the savings will be transferred to our company. This approach is beneficial for both parties — we keep improving our technologies and small businesses benefit from using AI technologies at minimal cost.

SmartCoders is a company specialized in the field of artificial intelligence and clients’ behavior analysis. Our team has many years of experience in customized