Insights
From conversation volume to conversation value: What AI means for customer relationships
Insights
AI is changing the operating model for customer communications. Conversations are no longer limited by agent capacity or business hours. That shift fundamentally changes what customer conversations can achieve, and the value they create for each individual consumer.
But many brands haven’t adjusted to that reality.
Businesses often use AI to optimize for speed and scale – handle more inquiries, end interactions faster, reduce cost. Their mindset hasn’t changed, even though the technology has evolved. Too often, many brands treat conversations as something to manage rather than moments that can strengthen relationships.
The problem with this approach is that scale without context creates false confidence. When AI scales interactions without intent, it automates low-value engagement. Customers get answers, and they may even get resolution, but they don’t get progress or relevance. In 2026, that’s just not good enough.
The next phase of AI in customer communications won’t be defined by how many conversations brands can have – it’ll be defined by if brands meet customer expectations in every interaction. This perspective builds on themes we explored in our Sinch Predictions 2026 on customer communications, looking more closely at what those shifts mean for how conversations are defined, measured, and valued.
In 2026, consumers expect that their conversation history is carried forward to all their interactions with a brand. If they checked an order status over chat yesterday and then called today to update their delivery time, they expect the brand to connect those moments. They also expect the brand to anticipate what they need next, without being made to start over.
Sinch’s State of customer communications 2025 report found that 59% of consumers say it’s important that information flows between channels like email, text, chat, and voice. That context is the difference between a conversation that’s useful and one that’s just forgettable. A customer conversation has evolved into becoming an ongoing thread that carries context, memory, and intent across the relationship.
And when it doesn’t, the experience breaks down quickly: 81% report a negative reaction when they’re forced to repeat information, with many citing frustration and lost trust.
of consumers say it’s important that information flows between channels.
of consumers report a negative reaction when they’re forced to repeat information.
The problem? Most organizations aren’t set up to meet those expectations.
According to Vandita Arora, Group Product Manager at HubSpot, even when a business has multiple channels in place, fragmented data prevents conversations from feeling continuous or coherent.
“What still feels fundamentally broken is the isolation of data… even though you have all these channels, the context isn’t shared, and every interaction starts from zero again.”
Consumers expect conversations that recognize who they are and remember what’s already happened. Once conversations become continuous and contextual, success can’t be measured by how quickly they’re shut down. The moment conversations stop being isolated, speed-based efficiency metrics start telling the wrong story.
Most of the metrics used to measure customer conversations were designed for a world of scarcity. There was more inbound demand than available agents, and more conversations than time. In that environment, metrics like Average Handle Time (AHT) and deflection made sense. Speed was the only way to keep up.
AI changes that equation.
As conversations become less constrained by capacity and more connected across channels, optimizing purely for speed misses the point. AI can shorten interaction time, but that doesn’t always make them successful. In many cases, it does the opposite by compressing conversations that actually need more time and context dedicated to them.
And that gets to the heart of why legacy efficiency metrics don’t cut it anymore. When AI is used primarily to drive deflection or reduce handle time, teams may hit their numbers while missing the outcome entirely. A quick interaction that ignores context can feel efficient internally while creating friction for the customer.
What matters more now is whether a conversation creates progress:
Measurement has to evolve accordingly. Not towards more metrics, but towards better questions. Instead of asking how fast a conversation ended, leaders need to understand whether it created continuity, trust, and long-term impact.
When conversations are designed to create progress, the role of customer service starts to change. Service interactions stop being treated as costs to minimize and they start to shape how customers feel about the brand long after the issue is resolved.
A well-handled conversation can restore confidence after a problem or helps a customer make a better decision. It can also surface opportunities that genuinely benefit the customer, like a timely reminder to reorder something they rely on or a recommendation for a plan that better fits how they use the service. That carries value even when no immediate sale is involved. Over time, those moments compound, influencing influence loyalty, retention, and long-term value.
That’s the shift from volume to value, where success is defined by what each conversation contributes to the relationship.
Seen this way, service becomes a growth engine because it consistently earns trust by acting at the right moment, one conversation at a time.
Once brands start thinking of service as a driver of long-term value, the question will stop being how conversations perform – and who those conversations are actually between. Today, most customer communications are designed around a person reaching out on one side and a company responding on the other, whether through a human agent or a basic chatbot. That model is already starting to change.
The next inflection point will be the shift toward agent-to-agent communication. In these interactions, the “customer” initiating the conversation won’t be a person at all, but a personal AI assistant acting on their behalf.
Instead of a customer checking an order status, following up on a missing package, or comparing options themselves, their AI assistant will handle it. The assistant will contact the relevant brand or delivery systems, exchange information, and return with a clear answer. The consumer never has to navigate multiple touchpoints again because their assistant does it for them. And even though these conversations happen machine-to-machine, the outcome still directly affects how a person experiences and trusts a brand.
This has meaningful consequences for businesses. When AI systems are interacting directly with other AI systems, brands need to be explicit about responsibility. A brand’s AI agent will need to know what information it can share, what actions it’s allowed to take, and when a situation requires human involvement.
Over time, this points towards an emerging “internet of agents,” where AI assistants begin to directly interact with brand systems in controlled ways. Some brands will manage those interactions themselves, while others will participate through defined interactions and permissions. In both cases, how an AI agent behaves becomes a core design decision.
Early signals of this shift are already visible in open-source projects like OpenClaw’s AI assistants that can interact directly with other AI systems, exchanging information and acting on a user’s behalf without human involvement. Just as importantly, these projects highlight how important security and secure guardrails are for agent-to-agent interactions to be safe and trustworthy at scale.
The brands that succeed in this next phase are the ones that define authority early and set clear boundaries for their AI agents. In an agent-to-agent world, trust will be earned through predictable behavior and systems that act responsibly on a brand’s behalf. Brands that start planning for this now will be far better positioned as these interactions become part of everyday customer journeys.
AI is changing how customer conversations happen. The next step for leaders is deciding how those conversations are meant to perform.
As interactions become continuous and increasingly automated, success needs to be defined around outcomes. Leaders need to understand whether conversations create progress, reduce uncertainty, and build confidence over time. That’s only possible when context is preserved. Speed and efficiency still matter, but without context they only show how fast conversations end, rather than if they create continuity, trust, and meaningful outcomes.
And as AI agents take on a larger role, they will begin to represent the brand in every interaction. Their responses are shaped by the context they’re given, and directly impact trust. Their decisions have consequences. Clear intent, boundaries, and accountability are required to ensure those interactions remain consistent and reliable as volume grows.
The challenge ahead is managing growth without losing control, and scaling conversations without eroding trust. Leaders who approach AI with that discipline will be better positioned for what comes next.
Want to learn more? Watch Sinch’s on-demand webinar, What’s in the cards for customer communications, to hear directly from industry leaders on how they’re thinking about the next phase of customer communications.
Author:
