Call Middle Jobs Are Declining, however 14,000 Calls Present Humans Nonetheless Outperform AI
New analysis from Bold Agent Kit finds that human receptionists delivered roughly 4 instances stronger inbound consumption conversion efficiency, regardless of AI answering extra calls.
Artificial intelligence is quickly altering the economics of customer support. Companies are deploying AI receptionists, automated name handlers, and voice brokers to cut back labor prices, prolong working hours, and deal with rising name volumes with out increasing their workforce.
And the influence on employment is starting to indicate.
An October 9, 2026, report from Andreessen Horowitz (a16z), citing employment knowledge from Revelio, highlighted a major decline in international call-center employment development. After greater than a decade of largely constant growth, call-center employment development has turned unfavourable throughout revenue teams.
The report raises an necessary chance: AI might already be contributing to the displacement of call-center employees.
But at Bold Agent Kit, a platform that permits companies to construct and deploy AI receptionists and communication brokers, we needed to look at a special query.
Just as a result of AI is changing human employees, does that imply it’s performing their jobs higher?
Our evaluation of roughly 14,000 buyer calls suggests the reply is much extra sophisticated than the business’s enthusiasm for automation would possibly suggest.
The Study: 14,000 Calls, Human vs. AI
Working with our purchasers, Bold Agent Kit evaluated roughly 14,000 inbound calls to match the efficiency of AI-powered receptionists with human receptionists.
Of these calls, 6,343 have been dealt with by AI brokers, whereas the remaining roughly 7,657 have been assigned to human receptionists.
Our evaluation targeted on three enterprise outcomes: profitable inbound consumption conversions, calls deserted with out significant progress, and missed calls.
The findings challenged among the assumptions driving the speedy adoption of AI voice brokers.
Key Findings
1. Human receptionists delivered roughly 4x stronger inbound consumption conversion efficiency.
In our evaluated calls, human receptionists considerably outperformed their AI counterparts at changing inbound inquiries into profitable consumption outcomes.
This is doubtlessly essentially the most consequential discovering for service-based companies.
For a regulation agency, contractor, medical follow, or different service supplier, the aim of a receptionist shouldn’t be merely to reply the cellphone. It is to know the caller’s wants, construct belief, gather correct info, and assist transfer that individual towards the subsequent applicable step.
Answering a name is an exercise. Converting that decision right into a significant enterprise final result is the consequence.
Our findings point out that, a minimum of among the many workflows we examined, human receptionists have been significantly simpler at producing that consequence.
2. AI-handled calls skilled 700% extra no-progress hangups.
Our evaluation additionally discovered that AI-handled calls had a reported 700% greater incidence of callers hanging up earlier than significant progress was made.
That is a very regarding final result for companies investing in automation to enhance their buyer expertise.
An AI agent might reply instantly, ship knowledgeable greeting, and comply with a rigorously designed script. But if callers disengage earlier than finishing the consumption course of, that interplay might produce little enterprise worth.
There are a number of potential explanations for this conduct, together with conversational friction, response delays, issue dealing with surprising questions, or callers’ preferences for talking with an individual. Our research didn’t isolate which components prompted the distinction.
Whatever the reason, the result issues.
A name that’s answered however deserted with out progress mustn’t routinely be handled as a profitable buyer interplay.
3. Human receptionists missed roughly 15% extra calls.
AI did exhibit a significant operational benefit: availability.
Human receptionists in our research missed roughly 15% extra calls than AI brokers. AI receptionists could be configured to reply incoming calls persistently, together with in periods when workers are unavailable.
That benefit is important.
Missed calls can characterize missed earnings, significantly for companies the place a potential buyer is unlikely to go away a voicemail or name again.
But our knowledge highlights a important distinction: answering extra calls shouldn’t be essentially the identical as producing extra enterprise.
An automated system that persistently solutions the cellphone however struggles to transform callers might not create the financial return a enterprise expects.
Conversely, a human receptionist who converts inquiries successfully however can not reply each name additionally presents a chance for enchancment.
The problem is discovering the appropriate stability.
The Hidden Cost of Replacing Human Receptionists
Much of the enterprise case for AI receptionists focuses on labor financial savings.
A enterprise would possibly examine the month-to-month value of a human receptionist towards an AI service that may reply calls across the clock at a fraction of the expense.
Viewed strictly as a staffing expense, the AI various can look compelling.
But that calculation is incomplete.
For many service companies, a receptionist shouldn’t be merely an operational value. The receptionist additionally contributes on to earnings technology.
Consider a enterprise the place every profitable consumption conversion may result in a buyer relationship value tons of or 1000’s of {dollars}.
If switching to AI materially reduces the share of callers who change into prospects, the misplaced earnings might outweigh the discount in staffing prices.
A system that prices 80% much less to function shouldn’t be essentially extra worthwhile if it additionally converts considerably fewer alternatives.
These are illustrative economics, not a calculated return on asset placement from our research. Actual profitability is dependent upon name quantity, conversion charges, buyer worth, staffing prices, and the forms of inquiries being dealt with.
The bigger level stays: companies ought to consider the fee per profitable final result, not merely the fee per answered name.
Employment Declines Do Not Prove Performance Improvements
The Revelio employment knowledge mentioned by a16z illustrates an necessary transition within the international labor market.
Call-center employment development has deteriorated considerably, with the downturn extending throughout nations at completely different revenue ranges.
AI is a believable contributor to that pattern, though employment statistics alone can not set up precisely how a lot of the decline is attributable to automation relatively than broader financial components.
More importantly, workforce reductions inform us little or no about whether or not customer support is enhancing.
An organization can scale back its headcount, decrease working bills, and course of extra calls utilizing AI whereas concurrently experiencing declining conversion charges, better buyer frustration, or extra unresolved inquiries.
These outcomes are usually not mutually unique.
Efficiency and effectiveness are two completely different measurements.
Our analysis suggests that companies ought to be cautious about equating one with the opposite.
The Opportunity May Be Human-AI Collaboration, Not Replacement
None of those findings counsel that AI receptionists haven’t any worth.
Quite the alternative.
We function a platform constructed particularly to assist companies create AI-powered receptionists and communication brokers. We see the worth of automating repetitive interactions, extending availability, accumulating info, routing calls, scheduling appointments, and dealing with routine communication.
But deploying AI successfully requires understanding the place it performs effectively and the place it at the moment falls quick.
For some organizations, the very best method could also be a hybrid mannequin.
AI can reply calls exterior enterprise hours, deal with easy inquiries, collect preliminary info, and route callers to the suitable individual.
Human workers can concentrate on conversations requiring judgment, empathy, persuasion, nuanced problem-solving, or relationship constructing.
Rather than treating each human-handled name as a price to remove, companies can use AI to make their current groups extra productive.
An AI receptionist doesn’t have to switch a human receptionist to generate a significant return on asset placement.
Sometimes, the very best software of AI is guaranteeing that human workers spend extra time on the conversations the place they supply essentially the most worth.
What This Research Does — and Does Not — Establish
Our findings replicate calls evaluated with taking part Bold Agent Kit purchasers. They shouldn’t be interpreted as proof that human receptionists outperform each AI system in each business.
Call complexity, enterprise kind, agent configuration, caller expectations, and implementation high quality can all have an effect on outcomes.
The evaluation is observational relatively than a randomized managed experiment, and variations between the human-handled and AI-handled name populations might have influenced the outcomes. The headline comparisons ought to subsequently be understood as outcomes noticed on this dataset, not common benchmarks or established causal results.
Future evaluations also needs to look at absolute conversion charges, earnings generated per inbound name, name complexity, buyer satisfaction, and the efficiency of hybrid human-AI workflows.
Nevertheless, the findings increase a query that deserves better consideration as companies proceed adopting AI communication programs.
Are corporations measuring how a lot work AI eliminates, or how a lot worth it really creates?
The Bottom Line: Cheaper Is Not Always Better
AI voice expertise is enhancing quickly, and its economics will proceed to vary.
It might ultimately equal or exceed human efficiency throughout lots of the customer-facing interactions examined on this research.
But companies ought to be cautious about assuming that future functionality is identical as present-day efficiency.
Our evaluation of roughly 14,000 calls discovered that human receptionists delivered considerably stronger consumption conversion efficiency, whereas AI brokers demonstrated a transparent benefit in name availability.
Both findings matter.
For organizations contemplating changing human receptionists with AI, the choice ought to be pushed by measured enterprise outcomes relatively than value financial savings alone.
The purpose of AI adoption shouldn’t be to remove essentially the most workers. It ought to be to supply the very best outcomes for companies and their prospects.
And based mostly on what we have noticed to this point, human beings nonetheless have a considerable function to play.
Try Bold Agent Kit
Bold Agent Kit helps you to construct your individual communication brokers that work throughout cellphone calls, textual content messages, and e mail. Each agent shares one reminiscence of each buyer throughout these channels, and also you determine which conversations it handles and which go to your workforce.
Build your own multichannel agent with Bold Agent Kit
About Bold Agent Kit
Bold Agent Kit is a platform that permits companies to construct, deploy, and handle AI receptionists and communication brokers. The firm works with companies to automate inbound and outbound communications whereas evaluating the real-world efficiency of AI-powered buyer interactions.
Research: Bold Agent Kit shopper name evaluation, roughly 14,000 calls, together with 6,343 AI-handled calls.
External reference: Andreessen Horowitz, This Week in Charts: Are Prediction Markets Politically Biased?, October 9, 2026, part “AI Has Come For Call Centers, After All?”, citing Revelio employment knowledge.
