The tax and accounting professional’s guide to everything AI

As a tax and accounting professional, you know the importance of having a forward-looking gaze. Business may be good today, but you must always be looking for ways to fast-track to tomorrow. Business operations are always a good place to start, and often offer critical insight into how your firm will bode in tomorrow's world. Streamlining the way your firm does research is one of the key methods to improving efficiency, but streamlining is only good so far as the accuracy of your research. Enter AI-enabled research software, such as Checkpoint Edge.

The first question to consider here is the motivation behind the push for AI in tax and accounting research. Are tax and accounting professionals taking a serious look at AI-enabled research because they are known for their tech-savviness and love of new software? Probably not. Do most tax and accounting professionals require AI-enabled solutions? That's where the answer isn't quite as clear. AI-enabled research software is seen as a means to an end; faster, more reliable, accurate answers to researchers' questions. This is always something firms are seeking to maximize, and AI continues to be the best way to get the job done. Not only is AI-enabled software going to find you the answer you're looking for, it's going to give you related news, content, and regulations you might be interested in. More on that piece to come.

The second question is what exactly makes Checkpoint Edge the best tax and accounting research software? It starts with the implementation process. Checkpoint Edge is the result of thousands of hours spent developing algorithms that learn from your search habits as you use it. The more questions you search, the more the program picks up on your language, search habits, and what sorts of content you like to see in relation to your questions. All of this is to provide a more well-rounded search experience, giving you not only the answer to your question, but deep dive articles, related news, or Thomson Reuters expert commentary. 

There are mixed opinions when it comes to the presence of AI in the workplace, and understandably so. We hear countless messages about AI replacing jobs and eliminating the need for administrative and bookkeeping roles in tax and accounting firms. But let's put the noise aside for a moment and focus on what AI is truly capable of, and what it's not.

Today, AI stands somewhere in the middle of narrow AI and general AI, where machines are capable of more than just data entry but cannot think for themselves as humans do. AI has the power to automate repetitive tasks, such as entering client data into a return or scheduling client visits. AI does not sleep, eat, make mistakes, or take vacation time. In this sense, time is saved on day-to-day tasks, human error is eliminated from the equation, and firms are granted more time to focus on growing their business.

In terms of research, AI has already taken a foothold in the industry. It used to be that if you had a tax question, you'd spend considerable time searching through volumes of books that serve better as distinguished décor than user-friendly resources. Now, the process has been completely transformed, thanks to AI. Software like Checkpoint Edge uses AI to deliver quick accurate answers for the most complex questions facing tax and accounting professionals. How is this done exactly?

To start, let's think about Google. When you search a question, you can get over a million results all containing the phrase or keywords you searched, tailored from most to least relevant, date most recently published, and/or most viewed based on a similar search. AI-enabled tax and accounting research platforms are essentially no different. Built into Checkpoint Edge are algorithms that have incorporated millions of user data points, including usage patterns and prior questions, which point you to the answer you're looking for much faster than traditional research tools would. A new algorithm in Checkpoint Edge is designed to recognize natural language phrases, so you can type a question as you'd ask it of a coworker, and the program can pick up on the phraseology to provide an answer, eliminating the time spent on trial and error keyword searches. As you're typing in a question to the search, the program also now features predictive questioning, where a recommended or similar search suggestion will appear below the search bar. These predictive questions have been formulated based on the over 55 million unique searches that have been done on Checkpoint in the last five years and are designed to minimize the amount of time a user spends searching for answers. The difference you’ll notice between a Checkpoint Edge search and a Google search of a tax or accounting question is the quality of results. Backed by the knowledge and integrity of all Thomson Reuters products, a Checkpoint Edge search will give you only trusted, vetted, and reliable results. Oh, and was it mentioned that it's self-learning? Checkpoint Edge collects anonymized user data to inform developers of user habits and better fine tune the software. We’ll get into the relationship between AI and data in a subsequent chapter, but to summarize in terms of Checkpoint Edge, data collected from users is used by humans, who develop the algorithms, which in turn give users of Checkpoint Edge a refined research experience. This process creates a positive feedback loop that only leads to better results and more time saved on tax and accounting research.

The capabilities of AI are almost endless … almost. There are some sophisticated features that are yet to be developed in the world of AI. One of these features is understanding of the user's context. Pushkar Bhoopalam, vice president of product management at Thomson Reuters, gives this example to explain the potential gaps in AI research: let's say you use an AI-enabled platform, such as Google, to search 'Italian food'. The system does not know whether you're looking for take-out delivery food, top rated Italian food, or recipes for Italian food. The search does not contain this context, so all related results will appear, making for a somewhat overwhelming experience. But if instead, you search 'best Italian food near me', you're providing the system with more detail to work with, and your results will be far more applicable to your question. Similarly, let's say you want to find information on the PPP. Searching 'Personal Paycheck Protection' will pull up hundreds of articles, deep dives, and news reports. But searching 'PPP for small businesses in Minnesota' will give a much more tailored result. As Pushkar explains, "In the first case, the user is not putting all of the information into Checkpoint Edge, so the more context the user can provide, the better the algorithm understands the question, and the better it can point the user to not only the right document, but the right section in the document as well." This gap between user language and AI understanding is slowly being filled in as more user data is catalogued and the algorithms become better, but for now, users will get better search results by searching specific key words or natural language phrases.

As tax and accounting professionals, you've come to dread the following: "We're implementing a new system.” Implementation often means hours spent without the ability to access your work, weeks of tutorials, and who knows how long troubleshooting. It is a wrench in the system, and there's no time for it. Certainly, the implementation of an AI-enabled research solution is no different, given its enormous capabilities, right? Wrong.

Here's the thing — Checkpoint Edge is a cloud-based solution, so when you sign up online, that's it. You're in. You're done. There is no implementation process, so to speak, meaning you can go from signing up to fully enabled user in a matter of hours. Now, there are plenty of features your firm may want more guidance on, and there's plenty of that available, but as far as time taken from a firm's busy schedule, you're not going to find that with Checkpoint Edge. It is designed to be usable from the start and allows your firm to take advantage of powerful AI technology right away. So, with busy season behind us and almost no set-up time required, there is really no better time to be evaluating AI solutions for your firm. Take this time to go through the steps of evaluating an AI solution, and set your firm up for the best year yet in 2021. Since the implementation process is so painless and easy, there isn't a bad time to adopt AI solutions. You should consider your staff’s ability to navigate new software, no doubt, but Checkpoint Edge is not a tool that will take hours just to get the hang of. It was designed to be user friendly and can be implemented at any time of the year.

While there is flexibility in when you implement AI, it's not a can you want to kick down the road, way into the future. As Pushkar explains, "You don't want to wait until everyone else already has AI solutions in their firm. Not only are you going to be behind the eight ball in terms of your practice overall, but you're going to have demotivated staff on your hands. They will see their friends at other firms doing research so much faster and so much more accurately than they are and they're going to say, 'Wait a minute, I want that', so don't be the last firm in town to adopt this sort of technology." We are all used to AI in our daily lives — Spotify, Amazon, Siri, and the list goes on — but when tax and accounting professionals go to work and see a non-AI solution, it slows them down because they're simply not used to it and it's not intuitive. They will turn to something that is AI-enabled, like Google, to do their research and end up with uninformed, unauthenticated answers. Don’t wait, take the leap and invest in your firm and stay at the cutting edge of all AI has to offer.

While there are nearly countless things individual firms must consider before jumping on the AI bandwagon, we've compiled a list of the top three things to consider when you're looking at your AI-enabled options. These are things all firms should ask themselves and seek out answers to in order to get the most out of their software.

1. Identify your biggest pain points: Before beginning the search for AI-enabled research, ask your firm "What are we hoping AI will do for us?" Identifying needs will help determine not only features you should be looking for in different solutions, but also whether AI is even the right answer. Instead of looking for a solution that has AI, ask yourself if AI can solve your pain points. For example, if your concern is new team members coming into the firm and not being able to find answers quickly, or finding the wrong answers, AI has the potential to drastically reduce the amount of time spent searching for the right answers. In this case, problem solved. However, let's say a chief concern is retention within the firm. Is implementing AI going to make a large difference? It might cause people to stick around long enough to explore AI, but it is not a solution for firm retention long term. In order to get the most out of your research software, be honest about what you're hoping it will do and how you can measure whether pain points are alleviated.
2. Assess the vendor: What is the history of the vendor? What are the capabilities of the vendor? What are other firms saying in reviews of the vendor? Doing the work on the front end to answer these questions will serve your firm in the long run. Having a good working relationship with your software vendor will make all the difference. We've all been on hold with customer service for a few hours too long at one point or another, and it leaves a bad taste in the mouth when you go back to use whatever product or service you called about. Save yourself on the front end, and make sure your vendor is trustworthy, responsive, and accessible. You also need to consider how new the competitor is in the market. New AI products, while shiny on the surface, are often not as sophisticated as ones that have had time to collect data and fine tune algorithms.
   
   Thomson Reuters is the most trusted name in tax and accounting research, and Checkpoint Edge is AI-enabled proof. Developed by expert data scientists and AI engineers, and backed by over 20 years in the industry, Checkpoint Edge enables your firm to maximize the efficiency of the day to day research and focus on what matters — your clients. Thomson Reuters is also a leader in customer service and ensuring your firm's software is working at its best. One customer had this to say about the level of care that's gone into the relationship between firm and Thomson Reuters:
   
   "The Thomson Reuters employees have a philosophy that is conducive to the business philosophy that I like and like to work with: they tend to go out of their way to help you solve your problems. They do answer the phone. They don't put you in a phone tree or just refer you to article A or article B. You can get personalized help. We get our questions answered. The people seem to put the clients first and take care of their needs"
   Paul Bogdanoff, CPA, Founder/Partner Bogdanoff Dages and Co, P.C. (2018)
   
   
3. Get to know the solution — In addition to doing your research on the vendor, it's worth your time to do the homework on the solution itself as well. Now, this point may seem obvious, but it's pointed out because it ties back to assessing your firm's needs. After identifying the needs of your firm, and verifying AI is the right solution, look for specific features within the solution that can solve for them. Live demos, a free trial, or asking for a walkthrough are all great ways to gain some familiarity and give an idea of whether the solution is the right one for you. For example, let's say your firm is skilled when it comes to finding the answers to questions they've got, but they need more information to inform their work. Checkpoint Edge has Concept Markers that highlight related information, such as news or pending legislation, that a user might want to explore in order to better inform their research. By the same token, be critical of the features you come across. Some software solutions may have a list of features longer than a legislation book, but it doesn't really matter if the features are not usable or adding value to your firm. When looking at different products, assess whether it makes a researcher smarter. If the answer is no, it's not worth the firm investment. But if the answer is yes, then note that as a solved pain point. Knowing the ins and outs of a software before you make the investment allows you to assess various features and make informed decisions for your firm.
   

The relationship between AI and data has gotten a bad rep over the years. News stories of manipulation between the two, leading to false conclusions and results, have led many to believe you cannot use AI without the risk of a big-brother-like event ensuing. This assumption, however, is a failure to consider their abilities separately and then consider how they can be put to work together.

AI, as we’ve discussed, is the ability of machines to perform tasks that were previously thought to require human intelligence. Tasks performed by AI, such as appointment scheduling and email reminders, do not require data analysis to be working in the background. Data analytics, on the other hand, is a broad term used to describe the process of collecting data and then finding meaningful insights from it. In practice, this is commonly used to inform decisions on customer behavior or price modeling. Data analytics does not require AI. We can perform data analysis and draw conclusions on our own, without the help of AI. But as a firm, you have better things to do with your time than to be collecting data in house and spending hours trying to make something useful of it. The power of systems like Checkpoint Edge comes from using analytics and AI in tandem. Data analytics are used to collect anonymous information on users’ search habits and methods. This data is then used by AI to perform human-like tasks, such as suggest related content or flag sections of a bill you might want to know more about. In this sense, AI creates value from data.

On the surface, artificial intelligence can be complicated and overwhelming. Yet, when it’s broken down, it’s clear AI has the power to transform the way your firm does research, giving you faster, better results. The state of AI has never been better for your firm to take advantage of new ways to explore questions and ensure you’ve got all the information to make informed decisions. Checkpoint Edge is designed to offer your firm a seamless transition into a new, intuitive, and intelligent era of tax and accounting research.

Learn more about Checkpoint Edge, request a free trial, or contact us to speak with a knowledgeable sales rep and see how you can harness tomorrow’s innovation today.  Call +1 800 431 9025.