News and Insights

Customer churn can feel like the company equivalent of being ghosted. Where did that customer go, did you even see them leave the party? It can be an incredibly frustrating part of the customer lifecycle, and at scale, it has a direct impact on growth. The good news is that through the use of AI, companies are better placed to understand their customers and anticipate when they are likely to leave, and why. Max Jackson from the ECI Commercial Team shares his top five top tips for using AI to keep customers from doing the digital Irish goodbye: 

Predictive analytics: the crystal ball of churn

AI can now be used to spot a ghoster before they even start fading. By analysing patterns like a drop in usage, a spike in support tickets, or suspiciously quiet buying habits, machine learning models can be used to identify high risk customers and flag when they look like they are about to leave, allowing you to intervene before it is too late.

Personalised engagement: because no one likes generic

Used properly, AI doesn’t do “Dear Valued Customer.” Instead it can be used to segment users into micro-groups and serve up hyper-personalised content - think Netflix recommendations, but for your product. Companies that leverage this ability to segment and customise relevant content will find their marketing efforts driving much higher engagement.

AI chatbots: the customer whisperers

AI-powered chatbots are always on and always polite, and as users get more comfortable talking to a bot there is likely to be further growth in trust and, in turn, use of these tools.  Chatbots are evolving beyond handling FAQs; they can engage inactive users, triage customer queries and suggest next steps, keeping the conversation (and the relationship) alive. The data and insights they capture around customer conversations can also be used to drive product and marketing decisions. 

Sentiment analysis: what are customers really thinking

Natural language processing tools can now be used to scan emails, chats, reviews, and social posts to detect shifts in customer sentiment. Whether it’s subtle frustrations or full-blown disappointment, AI can identify emotional cues across multiple channels and over time, helping you pinpoint where attention is needed most. 

Proactive retention: the digital equivalent of “don’t go!”

The best AI tools won’t just sense a customer is drifting, they can act or prompt a salesperson with the best next action. Whether it’s a timely discount, a helpful nudge, or a reminder on product features and roadmap, there are an increasing number of tools to help reel customers back in before they’re gone for good.

Georgia Ling recently attended the Economist Impact’s AI in Health Summit 2025, where policymakers, healthcare and MedTech providers, and investors convened to discuss how to unlock the potential of AI in healthcare. The real question to debate here is, what is going to tip the balance to drive AI adoption at scale in the health sector?

1. No time to waste on data build

“Good AI” depends on “good data.” Yet what constitutes “good” varies depending on the problem at hand. We all know that progress in healthcare is slower due to legacy systems, siloed and inconsistent data entry across trusts, the multi-modal nature of clinical information, and the sensitivities of patient privacy. This serves to highlight the scale of the opportunity and the appetite for solutions, but also the reality of the challenges at hand. The £1.9bn Frontline Digitalisation Programme and the £10bn NHS Digital Transformation budget for 2025–2029 are steps in the right direction, whereas cross-industry partnerships can help to accelerate standard-setting and data interoperability. HealthTech firms that can help solve the data problem first will be the most valuable.  

Once these foundations are in place, it will open up new data streams (and revenue streams) through the rise of wearables and privacy-enhancing technologies – creating trust and enriching doctor-patient relationships, as well as health outcomes.  

2. Expectations befall reality for AI adoption

Has the promise of AI been realised? At the Summit one hospital reported access to 120 AI tools yet active use of only 5, largely because most were poorly integrated into workflows or misaligned with clinicians’ needs. Adoption of AI tools hinges on expectations and taking people on the journey. What is the promise that is being made? Is it cutting down bureaucracy, improving insight, or more time for patient care? Involving clinicians early in product design helps surface the real pain points and accelerates adoption.   

The same logic applies across the healthcare ecosystem. Payers, providers, and pharma companies are drawn to AI tools that streamline decision-making, reduce uncertainty, and improve pricing or commercial efficiency. When a solution directly accelerates speed to market, sharpens negotiation outcomes, or increases ROI, the incentive to adopt becomes self-evident. Every unused tool is a sunk cost - and a missed opportunity for better care and outcomes. 

3. Balancing innovation with trust

Regulators know that they are playing catch-up to innovators in AI. They recognise the need to facilitate innovation and to encourage experimentation, but equally the need for guardrails in place as patient and clinician trust remains the bedrock for adoption. Thoughtful debate about what constitutes an “acceptable” risk profile in different settings is therefore central to the future of AI adoption. However, given the pace of change and development, will regulators realistically keep up with the cutting edge?  

Conclusion

The Summit highlighted the enormous opportunity ahead if the promise of AI is unlocked. Better data foundations, a deep understanding of the pain points in healthcare and regulatory clarity will be critical enablers, but trust, collaboration, and a willingness to rethink entrenched processes are foundational to innovation, while ensuring patient safety and adoption.

For more information on how we are supporting companies in the healthcare sector, please contact us via the form below.   

Remember when "just Google it" was the default response to any question? For Gen-Z that phrase is starting to feel as dated as looking something up in the Yellow Pages. The digital discovery landscape is splintering, with a significant generational shift.    

Recent data from a session hosted by our portfolio company, Croud, at the ECI Unlocked Digital Growth Summit reveals that search is fragmenting. This is both in relation to channel, with 51% of Gen Z now choosing TikTok over Google as their search engine, but also with LLM aided search it means that two thirds of searches now result in no click whatsoever.  This fragmentation is creating both challenges and  opportunities for businesses that understand how to navigate this new landscape. 

1. How is search fragmenting?

Traditional search engines still dominate by volume, so this isn’t a case of ignoring Google as a channel, but it’s important to be aware that discovery is changing:  

• Within search itself: The rise of zero-click searches means users increasingly get their answers without leaving Google. 52% of users now use full-length queries when searching, fundamentally changing how people are looking for you online.   

• In new environments: From social platforms like TikTok and Instagram to AI-powered tools like ChatGPT and Perplexity, discovery has become a multi-channel journey.   
  
• Through new behaviours:  Users, particularly Gen Z, are searching with different expectations. They want discovery over quick answers, visual-first experiences, and social proof rather than single "correct" results from traditional search engine. This can make it harder for businesses to predict or track exactly how and where their customers will find them online.  

2. Why it matters for your business 

For B2B software companies, click-through rates for key industry keywords have declined by around 30% since AI-generated summaries began appearing at the top of search results. Meaning that when potential customers can get their questions answered without clicking through to your website, traditional conversion funnels break down.  

Yet this creates opportunities for forward-thinking companies. New discovery patterns mean new ways to get in front of your potential customers. Compliance software providers are finding prospects engaging on specialised forums, HR tech companies are seeing research happening through LinkedIn communities, and cybersecurity firms are watching buyers consulting AI tools for vendor comparisons. Most companies are researching how they are appearing on LLMs and creating action plans to ensure that is representative of their GTM strategy.   

3. Search strategies for growth companies

The winners won't be those who optimise for one platform. They'll be companies that build intelligent, consistent, and relevant presence across multiple discovery channels. This requires three fundamental shifts:  

• Get close to your audience: Understanding where your specific customers search has become more critical than general search optimisation. More than half of Gen Z use social media for brand information, compared to 45% who use search engines. Consumer behaviour tracking and social listening now provide more actionable insights than traditional keyword research.  

• Track brand perceptions:  With 47% of UK consumers using AI for personal research purchases, monitoring how AI systems represent your brand has become essential. Tools like Croud's BrandAI help by analysing how major language models describe your brand compared to competitors, performing thousands of reviews to identify your positioning in AI-driven responses and highlight key drivers of purchase intent in your category. 
  
• Boost your brand visibility: Search engine optimisation remains important, but it's no longer sufficient. Digital PR, social media presence, and platform-specific content strategies now work together to create comprehensive discoverability.  

4.  Don’t ignore Google  

While Google’s monopoly of the search market is challenged by this fragmentation, abandon it at your peril. Google still accounts for 9% of all web visits, and is still the dominant mode of search. Google won’t let itself be left behind either, the platform continues to evolve, integrating AI capabilities and adapting to changing user behaviours.  

Source: Google Trends (Business Services category)  

Graph shows how Google search volume for business services keywords has remained stable from 2022-2025, indicating consistent demand despite search fragmentation across platforms. Since this data was collected, Google has introduced AI Mode to the search experience, which is only likely to reduce click-through rates further.

Rather than replacement, we're witnessing expansion. The shift toward multi-platform search brings platforms like TikTok, Reddit, and Instagram that complement or even enhance traditional search engines.   

"The search volume for business services has remained remarkably stable since 2022, which tells us that demand for information hasn't decreased," says Duncan Nichols, Director of Strategy & Planning at Croud. "Instead, it's been distributed across more touchpoints, creating additional opportunities for businesses willing to adapt their approach."  

For more information on how we are supporting companies navigate the changing search landscape, please contact us via the form below.   

As AI usage becomes standardised across cyber providers, the real question is: what sets one cyber firm apart from another and how does that translate to value? Providers with the best future growth potential are constantly evolving their go-to-market strategies, defences, and vertical-specific solutions. Daniel Bailey looks at 5 key areas within the evolving value proposition in cybersecurity:

1. AI is the new baseline. So, what’s the differentiator? 

AI is now standard in cyber - used by both attackers and defenders. Generative AI powers phishing and reconnaissance; defenders counter with machine learning to detect threats and automate containment, and both sides continually use learnings to optimise.  

With AI now a baseline, the differentiator is all about service levels, expert oversight and integration with existing stacks or layering on top of service providers like Microsoft or Google. Leading SOCs stand out by delivering measurable outcomes, utilising MITRE ATT&CK-aligned detection engineering and proactive threat hunts to improve Mean Time to Detect (MTTD) and Mean Time to Respond (MTTR). They also combine AI and human oversight to identify and contain threats, delivering security confidence to customers.  

2. Go-to-market strategy is the real innovation 

Smaller cyber software providers can’t match big tech on R&D spend, so a focussed product strategy is important, as is an effective go-to-market strategy, which can become the real battleground. The most successful firms focus on integration, not replacement - working with existing tools to enhance resilience. There’s a growing demand for zero-trust service models that offer scalable and proactive defence, and for bespoke monitoring tailored to industry, company size and tech stack. Rather than selling a one-size-fits-all SOC, leading providers differentiate by how well they adapt to the customer’s environment. 

3. Operational Technology: The next frontier?

Operational Technology (OT), like factory floors and smart infrastructure, is fast becoming the next frontier in cyber security. Unlike IT environments, OT systems often lag in cyber maturity, creating vulnerabilities as IoT adoption grows and connectivity between OT and IT increases. This opens the door for advanced persistent threat actors, such as Volt Typhoon, a group engaged in cyberespionage reportedly on behalf of the Chinese state.  

Providers that specialise in OT security - using network segmentation, device authentication and privileged access controls - have a solid growth opportunity. This is especially critical in sectors like manufacturing, energy and healthcare, where infrastructure is both complex and often regulated. 

4. Vertical-specific value and regulatory readiness

Different industries face different cyber threats - and sectors like healthcare and finance are especially exposed to sophisticated attacks from advanced persistent threat actors from the likes of Russia, China, Iran and North Korea. At the same time, regulations such as NIS2, DORA, and the upcoming UK AI Act are reshaping compliance expectations. Providers that offer sector-specific solutions and help clients navigate these frameworks are better positioned to grow. For many cyber firms, regulatory change isn’t just a challenge; it’s a growth lever. Those that make compliance a core part of their USP often see stronger valuations. 

5.  People remain the most significant point of vulnerability

Despite advances in automation, human behaviour remains the biggest vulnerability in cybersecurity. AI-driven deepfakes and spear phishing have made social engineering more convincing and harder to detect. That’s why cyber firms that can support customers to upskill users as well as overlay verification tools are so valuable. 

Cyber firms that focus on that change, for example, through tailored, data-driven awareness programmes, build stickier customer relationships as they are delivering behavioural change. Measuring metrics like phish-prone rate and time-to-report, as well as integrating training with existing security tools, helps foster a positive security culture. We’ve seen this in action at ISMS, whose focus on both automation and human-centric defence has proven critical as high-profile attacks continue to exploit user error. 

If you would like to discuss the unique opportunities we see in the cyber sector, please get in touch.