The short version: AI agents already run customer service at Klarna and Bank of America, screen job candidates before a human sees them, chase sales leads on WhatsApp, and fix code overnight while engineers sleep. Most of what gets sold to small businesses as an “agent” is a chatbot with a memory, but the genuine ones are cutting response times from hours to minutes and, in a chunk of cases, quietly getting switched off after 90 days because nobody planned for what happens when they go wrong.
What counts as an AI agent
Before the examples, a quick line in the sand, because the term gets stretched to cover everything. A chatbot answers a question. An AI agent takes an action, checks the result, and decides what to do next without a human clicking “approve” each time. If it can only reply, it’s not an agent. If it can look something up, make a decision, and then do a second thing based on that decision, it is. I go into the mechanics of building one from scratch in this breakdown of what an AI agent is, but for this post I want the businesses that are already running them, warts and all.
Customer service: Klarna, Bank of America and the numbers that got published
Klarna is the case study everyone quotes, and it’s worth quoting because the numbers are unusually specific for a company. In February 2024 Klarna’s AI assistant, built on OpenAI’s models, handled 2.3 million conversations in its first month alone, doing the work of what the company said was roughly 700 full-time customer service agents. Average resolution time dropped from 11 minutes to under 2. Klarna also said the assistant matched human agents on customer satisfaction scores, and it’s now handling a large share of the company’s chat volume in 23 markets, around the clock.
Bank of America’s Erica is the older, quieter version of the same story. Since launching in 2018 it has passed 2.5 billion client interactions and serves more than 42 million users, mostly for things like flagging a duplicate charge, explaining a fee, or nudging someone toward a savings goal. Nobody calls Erica an “AI agent” in the marketing copy because Erica predates the hype, but functionally it is one: it monitors an account, decides something needs flagging, and acts on it before the customer asks.
Sales and lead qualification: the client story I keep coming back to
A recruitment agency I worked with in Manchester was losing candidates simply because nobody could screen applications fast enough. Their consultants were spending around 40 minutes per candidate reading CVs, checking basic fit, and drafting a first response, and with 200 to 300 applications a week for busy roles, good candidates were sitting in an inbox for up to nine days before anyone replied. By the time a human got to them, the strong ones had already taken another job.
We built a simple agent using a workflow tool connected to an LLM: it read each CV against the job spec, scored it against five fixed criteria, drafted a personalised reply within minutes, and only flagged the top-scoring candidates for a human to review and book in. Screening time dropped from 40 minutes to roughly 4 minutes of human oversight per candidate. Time-to-shortlist went from 9 days to 2. It didn’t replace a single recruiter. It just meant the recruiters stopped doing the part of the job that was pure admin and started spending their time on the calls, which is where they were good.
That’s the pattern worth noticing: the agents making real money for real businesses right now are rarely doing the flashy, headline task. They’re doing the boring middle step nobody wanted, which is exactly why the humans downstream get faster and better at the part that needs a human.
Several of these examples came from Queensland clients; the Brisbane AI consultant page has the local detail.
Coding and engineering agents that ship overnight
Cognition Labs’ Devin was pitched as “the first AI software engineer” in 2024, and while the initial claims were dialled back after independent testers found it struggled with anything outside a narrow band of tasks, the underlying pattern has stuck and matured. GitHub’s Copilot Workspace and similar coding agents now routinely take a bug report, write a fix, run the tests, and open a pull request for a human to review, with no engineer touching the keyboard until the review stage. Teams using these tools report the agent handling small, well-defined fixes (a broken link, a failing test, a minor UI bug) end to end, while anything architecturally messy still needs a person from the start. The lesson holds across every category in this post: agents are strong on narrow, repeatable, checkable work, and weak the moment judgement calls or missing context are involved.
Recruiting and HR agents screening before a human ever looks
Beyond the recruitment story above, larger platforms like Eightfold AI and Mercor run agents that source candidates, rank them against a role, and schedule interviews automatically, some processing tens of thousands of applications a month for a single enterprise client. The uncomfortable bit that rarely gets said out loud: these systems inherit whatever bias sits in the historical hiring data they’re trained or tuned on, and a badly built one will quietly filter out good candidates for reasons nobody can fully explain after the fact. Any business running one of these needs a human spot-checking a sample of rejected applications every month, not just the shortlisted ones.
Finance and back-office agents doing the reconciliation nobody wants
Invoice processing and reconciliation is where agents are quietly saving the most money with the least attention, because nobody writes a case study about matching purchase orders. Agents built on tools like UiPath or Microsoft’s Copilot Studio now read an incoming invoice, match it against a purchase order and delivery note, flag mismatches over a set threshold, and pass only the exceptions to a person. One mid-sized manufacturer I spoke with cut its accounts payable processing time from around three days to same-day for roughly 80% of invoices, with the remaining 20% (the mismatched or unusual ones) still going to a human. That 80/20 split shows up again and again across finance, HR and operations: agents clear the routine majority, humans handle the exceptions, and the ratio rarely goes above 80% without accuracy dropping.
Marketing agents running the repetitive campaign work
Marketing teams are using agents for the parts of the job that used to eat a Tuesday afternoon: pulling last week’s campaign numbers into a report, drafting the first version of ad copy variants for testing, tagging and routing inbound leads by source and intent, and scheduling social posts based on when an audience is online rather than a guess. I’ve written in more detail about how this looks in practice, with named tools and real workflows, in this piece on AI automation for marketing teams, but the short version is the same as everywhere else in this post: the agent drafts, a human approves, and the time saved goes into the strategy work that moves numbers.
Real estate agents using agents to never miss a lead again
Property is one of the clearest wins because response speed is almost the entire game: a lead that gets a reply within five minutes converts far more often than one answered the next morning. Tools built for this space now qualify inbound enquiries by text or voice, answer the obvious questions about a listing, and only pass a warm, qualified lead to the actual agent. I’ve covered what this looks like role by role, including what a dedicated AI consultant for real estate agents does day to day, because the setup work here is fiddlier than most people expect once you factor in MLS data feeds and compliance around who’s allowed to say what to a buyer.
Where these projects fail, and it’s not the technology
Here’s the part most write-ups on this topic skip. A widely cited 2025 study out of MIT found that around 95% of generative AI pilots inside companies failed to produce a measurable financial return, not because the models were bad, but because the projects were bolted onto broken processes with no one owning the outcome. Gartner made a similar prediction the same year: it expects over 40% of agentic AI projects to be abandoned by 2027, mostly for the same reason, unclear ROI and rising costs once the pilot has to scale past a demo.
I’ve seen this up close. A business I advised launched a customer support agent with real enthusiasm, then quietly turned it off eleven weeks later because nobody had built an escalation path for the roughly 15% of conversations the agent couldn’t resolve, and those customers were leaving angrier than if a human had just answered slowly in the first place. The agent itself worked fine. The plan for its failures didn’t exist.
Agents also hallucinate, confidently, and the businesses getting burned are usually the ones that assumed an agent checking its own work is the same as a human checking it. It isn’t. I’ve written a full explanation of why this happens and what reduces it in this piece on why AI agents hallucinate incorrect information, and it’s worth reading before you let any agent send an email or make a booking without a review step.
How to pick your first use case without wasting three months
- Write down every repeatable task in the business that takes over 20 minutes and follows the same steps every time. That’s your shortlist, not the exciting stuff.
- Pick the one where a wrong answer is cheap to fix, not the one touching money, health, or legal risk, for your first attempt.
- Set a number before you start: hours saved, response time, or cost per unit, and check it at 30 and 90 days.
- Build in a human checkpoint for anything the agent isn’t fully confident about, and define “not confident” in writing before launch.
- Budget for the boring bit, the data cleanup, the escalation rules, the monitoring, because that’s typically 70% of the actual work and almost none of the demo.
If this is your first serious attempt and you’d rather not learn the escalation-path lesson the hard way, working through the first pilot with someone who’s built a few is usually cheaper than the failed one, and I break down what that support should cost in this guide to AI consultant costs.
The unglamorous businesses making money from this right now
Away from the big brand names, plenty of solopreneurs and small agencies are building narrow agents as a service, one that qualifies leads for local trades, one that handles booking confirmations for clinics, and charging a monthly fee for it. It’s a repeatable small business model right now, and I’ve listed it alongside other realistic paths in this list of ways to make money online for anyone weighing it up as a service to sell rather than just a tool to buy.
Frequently asked questions
What is the most common real-world use of AI agents in business right now?
Customer service and lead qualification lead by a wide margin, because the volume is high, the questions repeat, and a wrong answer is usually cheap to correct, which makes it the safest place to start.
Do AI agents replace employees?
Rarely in full. Klarna’s own figures describe the assistant doing the volume of roughly 700 agents’ worth of chats, but the company still employs human agents for complex or sensitive cases, and most successful deployments follow that same pattern of the agent clearing routine volume while people handle exceptions.
Why do so many AI agent projects get abandoned?
Gartner’s research points to unclear return on investment and rising costs once a pilot has to scale, and in practice the most common cause I see is a missing plan for what happens when the agent gets something wrong, not a failure of the technology itself.
How much does it cost a small business to build its first AI agent?
Simple agents built on existing tools like Make.com or Zapier connected to an LLM can cost as little as a few hundred pounds a month in software plus setup time, while custom-built agents for specific industries typically run into several thousand once you factor in data integration and testing.
