Self-service channels are standard in service architectures, but they often do not fully resolve enquiries. The combination of uniform customer data, intelligent speech/text automation and AI-supported context provision increases the success rate of automated solutions, reduces follow-up work and improves service quality and cost-effectiveness.

Starting point: usage versus completion rate

Although customers frequently use self-service channels, the proportion of cases that are completely resolved there remains low. Gartner notes that ‘the average self-service customer support success rate today is just 14%’ (Gartner, 2024, para. 3). This gap causes repeat contacts, higher operating costs and negative effects on the customer experience. (Gartner)

Why pure automation is not enough

The cause is rarely technology alone, but usually fragmented data and a lack of integration. Twilio emphasises that many companies overestimate their understanding of their customers: ‘81% of brands believe they deeply understand their customers—yet only 46% of global customers agree’ (Twilio, 2024, para. 12). Without up-to-date, connected customer profiles, bot responses often remain inadequate and lead to escalation to live agents. (Twilio)

How AI increases speed and quality

AI works in three areas: (1) Intent and sentiment recognition reduces misrouting, (2) context enrichment (summaries, profiles) shortens processing times, (3) virtual agents close standard cases. McKinsey puts it this way: ‘AI-enabled customer service is now the quickest and most effective route for institutions to deliver personalised, proactive experiences’ (McKinsey, 2023, p. 1). These mechanisms increase first-contact resolution and efficiency while reducing cost-to-serve.

Voice automation (IVR) as a quick lever

Intelligent IVR strategies (with NLU/ASR) increase containment rates and reduce post-processing. Twilio quotes McKinsey with specific figures: a 5–20% improvement in IVR containment rates and a 15–25% increase in authentication rates can reduce call centre costs by 10–30% within three to six months (Twilio, 2024, para. 13; McKinsey, 2023, p. 4). This underscores the rapid ROI of well-implemented voice automation. (Twilio)

Practical examples with measurable effects

Universidad UK (education): By consolidating fragmented profile data and using an agent co-pilot system, automation rates increased significantly: ‘Virtual agents are now able to resolve up to 70% of cases, which has led to a 30% reduction in average handle time’ (Twilio, 2024, para. 11). The combination of unified profiles and AI assistance thus improved speed and satisfaction. (Twilio)

Toyota Connected North America (automotive/telematics): After introducing an integrated platform (Twilio Flex + CRM integration), after-call work decreased by approximately 13% and average handling time by approximately 18%; at the same time, the solution ensures automatic reconnection to the same agent in the event of disconnections, which creates continuity (Twilio, 2024, para. 15). (Twilio)

Marks & Spencer (retail): The switch to voice-based routing (speech-to-text + intent recognition) enabled high routing hit rates (approx. 90%) and significantly reduced manual switching efforts; the project went live in a short period of time (Twilio / M&S, 2018, para. 4–5). (Twilio)

Operational metrics as an indicator

Twilio reports the following figures: ‘In 2023, Twilio helped process over 927 million contact centre tasks, up 35% from 2022’ (Twilio, 2024, para. 12) — an indication of the volume that is already being managed via cloud-based platforms today. The potential for efficiency gains through better automation is correspondingly high. (Twilio)

Recommendations for decision-makers

1. Build unified profiles: Link CRM, transactions and interaction history in real time to enable personalised self-service experiences (Twilio, 2024, para. 9–12). (Twilio)
2. Use automation in a targeted manner: Automate standardised process steps and reserve complex, value-adding cases for employees (McKinsey, 2023, p. 2–4).
3. Modernise IVR: Implement NLU/ASR-supported IVR and better authentication; measure containment rate regularly (Twilio; McKinsey). (Twilio)
4. KPIs and monitoring: Continuously track containment rate, first-contact resolution, after-call work and CSAT/NPS; perform A/B testing for bot responses (Gartner, 2024, paras. 3–6). (Gartner)
5. Modular, API-based platform architecture: allows easy integration of AI modules, CRM and communication channels (Twilio; McKinsey). (Twilio)

AI does not automatically make self-service successful — but it is only through the combination of a clean database, contextualised automation and a modern IVR strategy that noticeable gains in efficiency and quality can be achieved. Decision-makers who systematically combine these elements can reduce costs, shorten processing times and increase customer satisfaction in the long term.

Selected quotes

1. Gartner: ‘The average self-service customer support success rate today is just 14%.’ (Gartner, 2024, para. 3). (Gartner)
2. Twilio: ‘81% of brands believe they deeply understand their customers—yet only 46% of global customers agree.’ (Twilio, 2024, para. 12). (Twilio)
3. McKinsey: ‘AI-enabled customer service is now the quickest and most effective route for institutions to deliver personalised, proactive experiences.’ (McKinsey, 2023, p. 1).
4. Twilio (Universidad UK): ‘Virtual agents are now able to resolve up to 70% of cases.’ (Twilio, 2024, para. 11). (Twilio)

Literature / Sources
Gartner, Inc. (2024). Self-Service Customer Service: Key Capabilities and Strategies. Section Customers expect a lot from self-service, and too few get what they want or need (para. 3). Retrieved from https://www.gartner.com/en/customer-service-support/topics/self-service-customer-service. (Gartner)

McKinsey & Company. (2023, 27 March). The next frontier of customer engagement: AI-enabled customer service (pp. 1–6). Retrieved from https://www.mckinsey.com/~/media/…/the-next-frontier-of-customer-engagement-ai-enabled-customer-service.pdf. (See p. 1 for the quoted basic statement).

Twilio. (2024, 18 July). Transform CX with customer-centric AI strategies. (See paras. 11–13 for Universidad UK, IVR figures and 2023 volume statement). Retrieved from https://www.twilio.com/en-us/blog/transform-cx-customer-centric-ai-strategies. (Twilio)

Twilio. (2024). 2024 State of Customer Engagement Report. Retrieved from https://www.twilio.com/en-us/lp/Twilio-State-of-Customer-Engagement-Report-2024. (Twilio)

Twilio / Marks & Spencer. (2018). Marks & Spencer calls on Twilio to transform its communications with customers (press release). Retrieved from https://www.twilio.com/en-us/press/releases/marks-spencer-calls-on-twilio-to-transform-its-communications-wi. (Gartner)