6 min read
Opinions expressed by Entrepreneur contributors are their own.
When AI was first launched, it was perceived as a technology that could revolutionize business models by creating a full understanding of a clients’ profile and their needs. It would be a must-have technology for large businesses to boost their balances and a magic tool for small business owners to remove barriers between their product or service and the final user — using the right data, of course.
At first, AI appeared to be something that could almost “magically” make perfect predictions about the success or failure of any product just by analyzing different consumers’ profiles and their purchasing behaviors. Startups imagined a world where market traction could be reached simply by running a Python script, rather than optimizing a minimum viable product.
Financial investors have welcomed AI as the beginning of a new financial cycle pumping large amounts of money at any stock labeled AI-based. The “dot-com era” that started in the 90s — and with it, the stock market crash due to excessive speculation — seemed to be replaced by the “AI-era.” Images of humans resting while robots worked on their behalf were the collective image of how this new era would look alike. FOMO (fear of missing out) did the rest, with large companies jumping into this technology without fully understanding it, or in many cases, without even knowing what to do with it. Python developers quickly became among the top hiring needs of many companies.
Since initiation, however, AI adoption has proven to be very challenging for large companies and presented huge barriers for small businesses; from Amazon Alexa and Facebook’s embarrassing moments to China’s police confessing of shaming the wrong person (who turned out to be a billionaire woman) caused by trust of their facial recognition system. Machines have proven to be less than perfect and behind humans in many cases, and to date, their glitches have indicated that they are simply not quite there yet. Hence the question: If large businesses haven’t yet mastered AI, how can small businesses succeed in this task? And even before that, a more generic question: Can small businesses benefit from AI, or is this something that only large companies — with large datasets, programmers and data analysts — benefit from?
1. Automate your boring tasks
Before you are told that small businesses simply don’t have enough data for jumping into AI and/or that not hiring a data scientist will seriously compromise your budget, clarification is needed. AI is often confused with Machine Learning. AI is a broad umbrella term that includes any application from text analysis to robotics, and Machine Learning is just a subset of AI. While AI is meant to do human tasks faster and more precisely, (hence removing the human-error factor), Machine Learning helps in predicting outcomes and making estimations.
Although Machine Learning has become much more approachable with platforms such as Amazon Pre-trained AI Services or Google Colab — which allows you to build a basic model in a matter of minutes without the need for hiring data scientists, you might not have the skills or more likely the data to run a trustworthy model. You still have the option, however, to use AI to automate your daily routines. Any repetitive task you can think of can be fully automated thanks to AI. For instance, you can record a set of actions on your computer and use AI tools with image detection and on-screen text recognition for activating them automatically. Or you might want to use advanced AI-organized text snippets and typing auto-completion to avoid writing the same things more than once.
2. Leverage AI in your pitches
Let’s face it, AI is a magic buzzword that gets people’s attention. Labeling your product or service as “using AI” gives some sort of higher credibility and it makes people more likely that will hear what you say. Any business can find a way to implement AI in their processes, whether that means using images or text recognition, matching algorithms, communicating bots, smarter classifications, natural processing language, etc.
My company tested this hypothesis. I own a PR agency and as such, we match brands with journalists, or at least, we pitch stories hoping to find journalists interested in what we tell them. We built an automated scraper to collect all the latest stories published by targeted journalists, saved them to a database, then used AI to classify the data. The goal was to get a precise idea of what each journalist was more likely to be interested in writing about. Frankly, results haven’t been that much better than when we did all this manually. In some cases, results came up worse, but that allowed us to pitch ourselves differently. From a traditional PR agency, we transformed ourselves into an AI-platform. We had just to create a landing page that would summarize our “new” approach. Adding “using AI” did all the rest. The result of this? Our conversions — measured as the number of people that would sign up for our program — basically doubled.
3. Use AI to interact more with your clients
One of the biggest advantages of AI, when applied to customer service, is that it can run many processes simultaneously. Imagine if you had one person running a customer services desk; they would be expected to handle one customer at a time, maybe two. AI gives a chance to serve more people at once. AI is not taking our jobs, but rather helping us to perform our roles more efficiently.
Take the messaging pop-up boxes on websites. Previously they were manually operated, meaning there was someone typing to you. Nowadays AI can take over and do some of the work for you, such as prescreening your clients or providing them with some common answers to their questions. Many times the chatbot can solve the client query or if it cannot find a solution, a human operator can step in.
The advantage of this is multiple queries can be dealt with at once, whereas before only one query could be dealt with at any given time. Saving businesses from having long queues at customer service counters or long telephone wait times is never good for a company’s image. The AI process frees up time for departments to focus on the customers who need it most.
AI does a smart classification of the information so that in complex customer service environments (such as highly specialized IT departments) they can forward the client query to the right customer service agent. This means that specialists can immediately step into complex issues without having the client being redirected from one department to another.