Last year, I talked about how Browser Extensions are changing the way we work. Reducing the number of places we need to go during our work day can directly improve company performance metrics and help your business scale faster and more efficiently, all while creating a better experience for your customers. Some examples include:

– Minimizing the time it takes to reply to a customer support issue reduces your issue resolution rate and improves NPS.

– Answering prospect questions without leaving your inbox. The faster you can respond to inbound leads, the more likely you are to connect.

– Knowing when your prospects are engaging in the content you send them to better understand their interest.

In December 2015, Slack launched its platform, allowing anyone to develop a bot on top of Slack. Since then, the number of bots for the workplace has exploded, creating a robust ecosystem of tools that help businesses work faster and more efficiently. Additional platform offerings were announced from Facebook, Microsoft, and most recently Google and create the potential for new bot-driven experiences at work. Like browser extensions, bots live where you work, providing exciting new ways for teams to transform their process.

Bots are not without their challenges

Bots are entrenched in the upward slope of the Hype Cycle these days, and many people like to talk about how they’re failing. However, when you boil down most of the negativity, it has more to do with the scope of the problem bots are trying to solve than the bot itself.

For example, we can’t (yet) expect a bot to “replace” a human. And while I won’t conjecture whether that will ever be possible, the fact of the matter is that most bots don’t succeed because they attempt to go far too wide with the problems they are trying to solve. Because bots (like other fields of AI) rely on NLP (natural language processing), they are only as good as their training data.

There is a direct correlation between the breadth of the problem you want to solve and the amount of high-quality training data you need access to. Many of us have probably experienced first-hand the limitations of broadly defined “personal assistants.” We try to treat the bot like an assistant and then it doesn’t work and we get mad! It’s not the bot’s fault. It’s a mile-wide and inch-deep problem, and the bot doesn’t have a chance of success because it is being asked to do too many things. When it comes to humans, we don’t simultaneously try to teach ourselves to code, sell, do customer support, etc. We tend to optimize and get really good at a relative few things. Same idea applies to bots.

Bots in the workplace are already working

But in the workplace, it’s a different story. The approaches being taken with bots that help us do our jobs better are already showing early signs of success. Why?

For decades, enterprise software has been seen as the “ugly and hard to use” kind of software, desperately trying to compare itself to “beautiful and easy to use” consumer software. But it’s this reality that actually makes bots an immediate success in the workplace. Think about the enterprise software you use today to do your job. And the bigger the company you work at, the worse it is! The products are confusing, hard to use, slow, etc. Now imagine a bot that allows you to complete a task in a fraction of the time it normally takes, without leaving Slack, in a few keystrokes.

The value of the bot here is the comparable experience; you are transforming a task that was long, distracting, and unpleasant into a much better, faster, simpler experience. And this is the opportunity that is so exciting. Think about all those tasks across the workplace where this “before and after” opportunity exists. It’s everywhere right? Everything is up for grabs.

In the early 2000s, everything in the enterprise software world started to change with the move to the cloud. Company workflows all got rewritten as vendors could now offer their customers a simple web browser experience. Compared to the on-premise client server alternative we were used to, this new world of SaaS “just worked.” Oh, and upgrades “just worked,” as well. But there are still many enterprises stuck on old unsupported software stacks because they were customized in a one-off manner, in such a way that they can never be upgraded without major surgery to critical company business processes. In the SaaS world, this was not possible. You had to productize the ability to customize so that it worked across all your customer channels. It’s very hard to do, but once it’s done, there’s a massive upside on the other end as you can just keep releasing upgrades and your customers get the benefit immediately.

However, while we certainly improved the user experience in this last rewrite, we didn’t make it that much better. Many of these SaaS apps are still complex, with many modules and features that only grow more complex over time. While we did get good at the “productizing customization” mentioned above, it came at the price of bloating the UI’s of those SaaS apps, with more and more features getting added over time. This meant an ever-growing checklist and list of options to choose from so you can accommodate an ever-expanding customer base.

So what if there was no “UI”?

What if you didn’t have to use these interfaces at all? What if you could have a simple conversation to accomplish your task? Something like filing a receipt for your expense report could be as simple as uploading a picture to your Concur expense bot. Need to update your pipeline? A simple conversation with the Troops bot, and you’re done. Does your team run daily standups? Let the Standuply bot take that over and replace the manual process of asking for updates, capturing notes, and sending them out.

These examples are all real, working bots. You can try them today. The other important thing you will notice as you try some of these things out is they are very purpose-built and narrowly focused on improving a specific workflow. As mentioned above, this is an important factor to success. These bots are taking a few workflows and really working to make them much better.

I think it’s an exciting time. Just like with the move to the cloud, once again all of these workflows are being rewritten. This time, we will see a dramatic change in the user experience. Combining bots with AI brings our daily tasks right into the places we are already working. It guides us through what we need to do proactively, and it helps us get answers to questions that previously would have required huge manual effort across numerous internal systems.

Rick Nucci is the cofounder and CEO of Guru, which helps companies like Airbnb, Square, Optimizely bring company knowledge into the workflows of their sales and support organizations.