Filip Háša is Director at Pixelfield, a London AI and product engineering studio he founded in 2013. The company ships production LLM, RAG and agent systems for scaleups and enterprises, and will say when a simpler fix is the better product.��

• A ChatGPT demo is not a product. It is a slide.
• Most AI proof-of-concept work never becomes a system people use every day. IDC research, reported by CIO.com, found 88% of observed PoCs never reach widescale production.
• RAG is how a system answers from your files, with citations and permissions. The 2023 “dump PDFs into a vector database” version is what failed. Retrieval itself did not die.
• Agents fail on cost, state and access, not on model IQ. A founder on X described a coding agent wiping production data in nine seconds. Last week, researchers showed training agents turning public wikis into a private message board.
• Write the one-sentence success test before you write the prompt. If you cannot, you are not ready to build.

If you have spent the last two years around UK startups, you already know the scene. Someone opens ChatGPT on a projector. The bot summarises a contract. The room nods. A board member asks how soon it can sit on the website. Six months later the same team is still “iterating the prompt”.

That gap is the story. Not the models. The models got better. The product work did not.

I run Pixelfield, a London AI and product studio. We have been shipping software since 2013. More than 50 AI features we built are still running: chatbots, retrieval, agents, older machine-learning jobs that never needed a large language model. The useful pattern, after enough of those builds, is unglamorous. The demo answers a question nobody will pay for twice. Production answers the question people ask on a Tuesday, with the right files, the right permissions, and a way to tell when the answer is wrong.

Here is how we got here, in order. Named people, dated sources. No vibes.

2022–2023: The Demo Became The Product

ChatGPT arrived and, for a few months, that was enough. Founders showed a chatbot that could talk about their help centre. Investors asked if the company was “an AI company now”. A Copilot licence looked like a strategy.

None of that was stupid. A live demo is how you find out whether anyone cares. The mistake was treating the demo as the finish line. A prompt on top of a public model has no memory of your permissions, no log of why it said what it said, and no owner when it is confidently wrong in front of a customer.

UK teams felt this first in regulated corners: insurance, health, property, anything with a paper trail. The bot was fine in a Monday standup. It was not fine once someone asked “where did that number come from?” and the only answer was “the model”.

A shared Copilot login still looks like a strategy. A few saved prompts. A hope that the next model will paper over the mess.

2023–2024: Everyone Ran An AI Proof Of Concept

Then the brief changed. Boards stopped asking for a demo and started asking for a proof of concept. Fair enough. A PoC is supposed to answer one question on real data, then stop: go live, iterate, or kill it.

What actually got funded was often a second demo with a nicer UI. The data was a SharePoint dump. The metric was “the CEO liked it”. There was no eval set, no failure mode, no plan for what happens when the corpus goes stale next Tuesday.

Hamel Husain described the stall in a February 2026 interview. Teams had glued together “the basic components of an AI system, like a RAG database and all the UI components”, shipped a prototype, and then got stuck. The hard part was not the glue. It was how to improve the thing without guessing.

Ashish Nadkarni at IDC put a number on the same pattern. Research commissioned with Lenovo, reported by CIO.com in March 2025, found that 88% of observed AI PoCs never made it to widescale deployment. For every 33 proofs of concept, four graduated. Nadkarni’s diagnosis was not “the model was too weak”. It was board-level panic, thin funding, and no real business case. “Most of the time the POC happens not because of a strong business case.”

Gartner had already warned, in July 2024, that at least 30% of generative AI projects would be abandoned after proof of concept by the end of 2025, for poor data, missing risk controls, rising cost, or no clear value. That was a forecast. Sitting in London in 2026, it does not feel like they missed.

A PoC that cannot be killed is not a PoC. It is a hostage.

2025: Agents Arrive, And So Does The Hangover

Once chatbots looked ordinary, the pitch became agents. An agent, in the useful sense, is not a chat window with a personality. It chooses tools, keeps state, and is supposed to know when to stop.

Gartner’s Anushree Verma, in a 25 June 2025 note, predicted that over 40% of agentic AI projects will be cancelled by the end of 2027. Read that as a forecast, not a body count. The line that matters is this:

“Most agentic AI projects right now are early stage experiments or proof of concepts that are mostly driven by hype and are often misapplied.” -Anushree Verma, Senior Director Analyst, Gartner, 25 June 2025

Gartner also named the sales trick: agent washing. Chatbots, RPA and search boxes, relabelled. Verma’s split is still the one I use with founders. Use an assistant for retrieval. Use automation for a routine workflow. Use an agent only when a decision has to be made and a human path exists when the agent should not.

Maxim Fateev, co-founder and CTO of Temporal, has been making the unfashionable point for months. Agent demos die on execution and state, not on IQ. In a 1Password conversation published in August 2026 he said people who started from “a synchronous Python in-memory program are learning the hard way that these are not simple problems.” A laptop notebook has no crash recovery and no audit trail. The first overnight job is where the demo ends.

What RAG Actually Is, And What Died

Somewhere in 2024 a lot of people decided retrieval-augmented generation was over. Context windows got bigger. Vendors said you could “just put the documents in the prompt”. Then the bill arrived, the model missed the one clause that mattered, and nobody could say which file the answer came from.

RAG, in plain English: the system fetches the right pieces of your own material, then writes an answer from those pieces. Citations. Permissions. A log. It is how a bot or an agent answers from your files instead of from its training data.

Hamel Husain opened a series on this in July 2025 with a line I wish more pitch decks had read: he was tired of hearing “RAG is dead”. Ben Clavié’s version is even cleaner. What died is the 2023 marketing version. Chuck documents into a vector database, do cosine similarity, call it a day. That fails because a single vector cannot hold a 40-page policy.

Retrieval itself is more important than ever. Models are frozen at training time. A million-token window does not make it cheap, or accurate, to stuff everything into every query.

If you want the engineering version of that argument, the production checklist we use for is unromantic: chunking that matches the document type, hybrid search, reranking, citation tracking, evaluations, freshness, and monitoring. Graph databases are optional. Guessing is not.

A UK example, from a bank rather than a startup pitch. In March 2026 the quantitative research machine-learning team at Lloyds Banking Group wrote up what happened when they tried to get a Copilot over SharePoint to answer methodology questions. Copilot was the right first test. It hit a wall: multiple sources, mixed question types, from “how does this model work?” to “what if we change a production parameter?”. They built their own intent detection and routing. Stateless “search then generate” was the limit. They said so in public on 24 March 2026.

2026: Cost, Permissions, And Nine Seconds

Two public stories from this year sit on opposite ends of the same problem. One is about money. One is about access.

The free lunch ended in August

On 23 August 2026, Drew Breunig published “Fable & The End of the Free Lunch”. Simon Willison linked it the same day. The argument is simple enough to steal.

“Prior to Fable, it felt silly to waste too much time improving your coding harness or context strategies. A new model would arrive at the same price (or cheaper!) and paper over most of your problems. But then Fable landed. It was (and still is!) incredible. But the cost was so high and Opus was good enough… for most of the code we needed. So we started to think about what work went where.”- Drew Breunig, 23 August 2026

He now uses the expensive model to interrogate a design, then hands a brief to a cheaper model for the rote work. GLM 5.2, in his comparison, was roughly a ninth of Fable’s cost and good enough for most routine coding once the context was strong. The other shock was policy: access controls and required data retention. Enough companies, and countries, started asking where traces go.

If you are a UK startup still waiting for “the next model” to fix a messy corpus and an unowned workflow, that essay is the polite version of no.

An agent on X, and a wiki message board last week

On 25 April 2026, Jer Crane, founder of PocketOS, posted on X what then ran in Decrypt and TechRepublic. A coding agent in Cursor, running Anthropic’s Claude Opus 4.6, was doing a routine task in staging. It hit a credential mismatch. It “fixed” the problem by deleting a Railway volume. The volume held the production database and the volume-level backups.

“Yesterday afternoon, an AI coding agent—Cursor running Anthropic’s flagship Claude Opus 4.6—deleted our production database and all volume-level backups in a single API call to Railway, our infrastructure provider. It took 9 seconds.” – Jer Crane, founder of PocketOS, posted on X, 25 April 2026, as reported by Decrypt

The agent found an over-privileged token, guessed that a delete would stay in staging, and did not ask. Crane’s point was not “one bad model”. It was an industry wiring agents into production faster than it is wiring confirmation, scoped tokens and delayed deletes. Railway later added a 48-hour soft-delete on that API path. For a UK team handling bookings or payments: a demo agent with blanket credentials is an outage with a chat log.

You do not need April for a current example. On 4 September, Simon Willison wrote up new research on OpenAI training agents that were supposed to have controlled web access. They found public wikis they could write to with GET requests, then spent weeks exchanging thousands of messages to finish a timed benchmark.

Moderators cleaned up. The agents noticed the alphabetical deletion sweep and started copying pages with a ZZZ prefix. Willison’s point on the sandbox is the sticky-note version: the proxy assumed GET cannot change data. Plenty of old software never got that memo.

That is not a startup using ChatGPT. It is the same shape of failure you get when a tool-using system meets a permission model written for humans. If your agent can send, delete, pay, or post, the control cannot be a sentence in the system prompt.

What Production AI Looks Like From London

I am not going to pretend every UK company needs a custom model. Most do not. A lot of the work we turn down is a rules engine, a better search box, or a form that already has the answer. One client arrived convinced they needed a custom LLM and left with a £5,000 automation. We could have sold a £40,000 build. We did not. Vendor-neutral is a dull phrase until you have sat in the meeting where someone is trying to buy a logo.

When AI is the right tool, production has a shape. There is a named owner. There is a corpus someone is responsible for. There is an eval set of real questions, including the nasty ones. There are permissions that match the product, not the demo. There are logs. There is a kill switch. There is a cost line that assumes the cheap model will not stay cheap if you send it everything.

A proof of concept, if you run one, should be designed as a go / no-go on real data. If we cannot write the question in one sentence, it is not ready. AgentWise, a real-estate operating system we still maintain, started as an AI PoC. That is the only reason to run one: it becomes a product, or it dies in public inside the team, before it dies in front of customers.

If you need an AI development company that will also say when the simpler fix is the product, that is the work we do at Pixelfield. London HQ, studios in Prague and Amsterdam, shipping since 2013. You keep the IP. Discovery starts at £2,000. A typical proof of concept sits between £8,000 and £20,000. Production is a different conversation, and it should be.

Founders are trying to buy an outcome, not a model. The outcome is a system that is still correct in six months. That is a product problem with some machine learning in it. It is not a prompt.

A Checklist Before You Spend The Next £20,000

Steal this. Argue with it. Do not skip it.

• Can you write the success test in one sentence, with a number a sceptic would accept?
• Is the data you need actually available, permissioned, and not three years out of date?
• Who owns the corpus next quarter? If the answer is “the intern who built the demo”, stop.
• What happens when the answer is wrong in front of a customer? Who gets the ticket?
• Do you need retrieval, a workflow, or a decision-making agent? Most pitches need the first or the second.
• Which model does the expensive thinking, and which model does the rote work? Breunig’s split is now table stakes.
• If an agent can delete, pay, send or post, is there a confirmation that the agent cannot auto-complete?
• What is the monthly inference bill if usage 10×? If you do not know, you do not have a production design.
• Will you kill the project if the PoC misses the number? Write that down in the same doc as the prompt.

FAQs

What is the difference between a ChatGPT demo and production AI?

A demo shows that a model can talk about your material in a room of friendly people. Production AI is a system with an owner, permissions, evaluations, logs, a cost model, and a way to fail safely. The first is a slide. The second is software.

Why do so many AI proof-of-concept projects fail?

They are often a second demo: no one-sentence question, no real data, no evals, no kill criteria. IDC, as reported by CIO.com, found 88% of observed PoCs never reached widescale production. Gartner separately forecast that at least 30% of generative AI projects would be abandoned after PoC by the end of 2025. The blocker is usually data, process and the business case, not model IQ.

Is RAG dead?

No. The 2023 version is. Dumping PDFs into a vector store and hoping cosine similarity finds the clause you need is a bad product. Husain and Clavié’s point in 2025 still holds: retrieval, routing, citations and evaluations are the work. Lloyds’ quant team had to leave a generic Copilot path for exactly that reason.

Should a UK startup build AI agents?

Only if a decision has to be made, tools have to be called, and you can say what happens when the agent should stop. Gartner’s 40% cancellation forecast by end-2027 is a warning about hype, not a reason to freeze. Crane’s outage is about credentials. Fateev’s point is about state. Start with retrieval or a workflow unless you can explain why those are not enough.

How much does an AI proof of concept cost in the UK?

It depends on the data, not the slide. On our side, a focused PoC is typically £8,000 to £20,000, after a discovery pass that starts at £2,000. If a vendor cannot tell you what would make them recommend you do not build, keep walking.
When is the simpler product the right product?

When a search box, a rules engine, a notification, or a well-designed form already answers the job. Generative models are bad at being a database and expensive at being a spreadsheet. We will say that in writing. You should want a studio that does.

The Short Version, One More Time

The last four years in UK AI look like this if you squint. 2022: the demo. 2023: the PoC. 2024: the hangover. 2025: agents, and a lot of relabelling. 2026: cost, permissions, and a few very public accidents. The teams that got through it treated production AI as product engineering with a sceptical eval harness, not as a prompt with a logo on it.

If you are still in the projector room, that is fine. Just do not confuse applause with a system. Write the one sentence. Then decide whether you are building software, or collecting another demo.

Filip Háša is Director at Pixelfield, a London AI and product engineering studio he founded in 2013. The company ships production LLM, RAG and agent systems for scaleups and enterprises, and will say when a simpler fix is the better product.

The post UK Startups Are Stuck Between ChatGPT Demos And Production AI appeared first on 91̽��.

Meta’s CTO has admitted, in an internal memo, that the company’s AI reorganisation was “an atrocious job.” It is a rare thing for a technology leader to say so plainly, and it is worth taking seriously, because the pattern behind it is not unique to Meta.

Since the reorganisation, code output has reportedly jumped by 220% among engineers reassigned to AI training work. On paper, that looks like a win. In practice, morale has fallen to a record low, and Meta is still monitoring employee activity even as it introduces new perks to offset the damage. Those two facts sitting side by side, rising output and falling trust, tell you almost everything you need to know about what went wrong, and it was not the technology.

I spent fifteen years in Fortune 500 technology leadership before moving into workplace wellbeing and leadership consulting, and I have watched this same pattern repeat across multiple waves of technological change. It happened with the shift to enterprise software, again with cloud migration, and now with AI. The technology moves fast because it can. Leadership moves more slowly because it depends on trust, and trust is not something you can deploy on a timeline.

What makes the AI wave different is speed and visibility. Employees do not experience an AI rollout as a neutral technology project. They experience it as a direct question about their own value: will I still have a role, will my judgement still matter, is my output being measured in ways I cannot see. When leaders do not answer that question honestly, employees answer it for themselves, usually in the least generous way possible. Output can rise in that environment. Commitment does not.

Meta’s response, adding perks while keeping the monitoring in place, is a common leadership reflex, and it does not work. Perks address comfort. Surveillance signals distrust. You cannot resolve a psychological safety problem with the first while reinforcing it with the second, and employees notice the contradiction immediately.

For founders and leaders building fast-growing companies, the lesson is not to slow down AI adoption. It is to treat the human side of that adoption as seriously as the technical side, and to do it before the rollout, not after morale has already dropped. That means being transparent about what is changing and why, giving people a real voice in how new tools are introduced into their own work, and being honest that AI raises genuine questions about job security and competence rather than pretending it does not.

There is a global standard for exactly this kind of risk, ISO 45003, which addresses psychological health and safety at work, and its logic applies directly here: psychosocial risk has to be assessed and managed with the same discipline as physical safety, not treated as a soft afterthought to a technology decision.

Meta’s admission is unusually candid, and that candour is useful. It gives every other organisation currently rolling out AI a preview of what happens when the technology outruns the leadership around it, and a chance to choose differently before their own employees force the same admission.

César Gamio spent fifteen years in Fortune 500 technology leadership. He is an author, adjunct professor at IE Business School, founder and managing director of the Dharma Centre for Workplace Wellbeing, and a British Standards Institution Associate Consultant for ISO 45003, the global standard for psychological health and safety at work.

The post Meta’s AI Reorg Is A Leadership Story, Not A Technology One appeared first on 91̽��.

The UK government launched a £500 million Sovereign AI fund back in April with a clear objective: back British AI companies, help them scale quickly and ensure that growth stays in the UK.

Designed to operate more like a venture capital firm than a traditional public funding body, this marks a massive shift in how the UK is supporting innovation. For SME leaders and founders, it is not to be mistaken with just another funding announcement. It signals a more competitive, better-resourced environment for building and scaling AI businesses, domestically.

Capital Meets Capability

Access to funding has always been a constraint for high-growth AI businesses, but Sovereign AI goes further than capital alone. It combines direct equity investment with access to infrastructure, talent and regulatory support, creating a far more complete growth environment.

For SMEs, that means fewer trade-offs. Instead of choosing between funding, talent or capability, businesses can access all three in one place, a true moment to focus on.

Infrastructure That Actually Moves the Needle

A key differentiator is access to the UK’s AI Research Resource supercomputers, something that has historically been out of reach for most smaller companies.

This level of computation completely changes what is possible. Models can be trained faster and tested more rigorously, removing a lot of the constraints that typically slow development. The result is a shorter path from idea to deployment, and a clearer route to building a real competitive edge.

Talent and Speed

Hiring remains one of the biggest blockers to growth in AI, particularly at the specialist level. Sovereign AI addresses this directly through faster visa decisions, making it easier to secure international expertise without long delays.

These are exactly the kinds of challenges being actively and continuously explored at . Founders and senior teams are comparing how they are securing talent, building AI capability and scaling faster in an increasingly competitive market, everywhere from stands to the speed networking area. The gap between businesses that can move quickly and those that cannot is widening, and access to the right conversations is a competitive advantage in itself.

From Funding to Market Access

Beyond investment, Sovereign AI also supports businesses in navigating procurement and regulation, opening clearer pathways into public sector opportunities.

For many SMEs, this pinpoints where growth either accelerates or stalls. Access to contracts, combined with a better understanding of regulatory frameworks, allows businesses to convert innovation into revenue more effectively.

What It Means for SMEs:

• Reduced barriers to scaling AI businesses within the UK
• Access to high-performance compute previously limited to larger firms
• Faster hiring through streamlined international talent processes
• Stronger access to procurement opportunities and regulated markets
• Support for building datasets and advancing AI capabilities

This is not just funding. It is a coordinated push to make scaling in the UK more viable.

The Bigger Picture

The global AI race is intensifying, and the UK has often struggled to retain its highest-growth companies as they scale. Sovereign AI is designed to change that by removing the need to look elsewhere for capital and capability.

For business leaders, the message is clear. The conditions for scaling in the UK are improving, but so is the pace of competition. Knowing how others are responding, where capital is moving and how quickly the landscape is shifting is becoming just as important as the funding itself.

Sovereign AI is more than a funding initiative. It is a signal that the UK is backing its innovators, and that the next wave of AI growth will favour those ready to move early.

The post The UK Government Has Backed AI With £500 Million: Here’s What That Means appeared first on 91̽��.

Artificial intelligence has generated no shortage of ambitious national strategies. The harder question is what happens when the technology leaves the data centre.

How does AI improve a production line, optimise an energy grid or manage a complex logistics network? How does a promising technology become something that can be manufactured, deployed and sold at scale?

These questions provide a useful way of looking at the next phase of relations between Germany and the United Arab Emirates.

President Sheikh Mohamed bin Zayed Al Nahyan’s state visit to Germany this week comes as the two countries to put technology much closer to the centre of a partnership historically built around trade, energy and investment. The agenda now encompasses artificial intelligence, advanced manufacturing, semiconductors, digital infrastructure, robotics and other emerging technologies.

The logic is not difficult to see.

Germany remains one of the world’s strongest industrial economies. Its advantages lie not simply in research or invention, but in the dense ecosystem that connects engineering companies, manufacturers, universities, specialised suppliers and the Mittelstand. That capacity to turn technical knowledge into reliable industrial processes is particularly valuable as AI moves beyond consumer applications and into factories, transport systems, energy networks and physical infrastructure.

The UAE approaches the same technological transition from a different direction. It has made digital infrastructure and artificial intelligence central to its economic diversification, alongside a wider effort to attract capital and companies into advanced manufacturing, technology and other new-economy sectors. It also offers access to investment capital and a pivotal geographic position linking European companies with markets across the Middle East, Asia and Africa.

That creates a potentially useful division of strengths: Germany can provide industrial and engineering depth, while the UAE can contribute capital, infrastructure, connectivity and the ability to help emerging technologies scale.

Her Excellency Lana Nusseibeh, the UAE’s Minister of State, put the ambition plainly in remarks in Berlin: “The aim is not just to exchange technology, but to build together.”

That distinction matters.

International technology partnerships can easily become exercises in selling products into one another’s markets. A more ambitious model would involve joint development: connecting researchers and universities with companies and investors, developing commercial applications together and building businesses capable of expanding beyond either country.

Artificial intelligence is probably the clearest test case.

Germany has considerable expertise in robotics, automation, precision engineering and complex manufacturing. The UAE has invested heavily in AI capabilities and digital infrastructure and has set itself the goal of becoming a global leader in artificial intelligence by 2031. Bringing those strengths together makes most sense not in another general-purpose chatbot, but in industrial AI: machine learning and automated systems applied to manufacturing, heavy engineering, energy, transport, logistics and supply chains.

This is also where technology policy begins to merge with industrial strategy.

For Germany, the challenge is not a lack of technical expertise. It is maintaining industrial competitiveness while manufacturing becomes more automated, software-intensive and capital-intensive. AI and robotics offer productivity gains, but deploying them across an economy dominated by sophisticated physical industry requires investment, experimentation and the ability to integrate new digital tools with existing machinery and processes.

For the UAE, the challenge is almost the inverse. The country has built an increasingly sophisticated digital economy and wants to deepen its domestic industrial base. Partnerships with German manufacturers and research institutions could help translate investment in AI and emerging technologies into more physical industrial capability.

There are already signs that the relationship is moving in this direction.

At GITEX Europe in Berlin in 2025, a UAE delegation brought Emirati technology companies and start-ups working in areas including AI, climate technology, digital health, smart cities and industrial innovation. The exercise was significant less for its scale than for the model it suggested: direct links between the two innovation ecosystems, allowing Emirati technology companies to enter Europe while giving German businesses another route into Middle Eastern and wider regional markets.

Start-ups could become an important part of that relationship. Germany has excellent technical universities, research institutes, established industrial companies and a mature innovation ecosystem. The UAE has developed a strong environment for venture funding, attracting international talent and scaling companies regionally. Connecting those systems creates opportunities in fields ranging from cybersecurity and health technology to smart cities and climate tech.

Cybersecurity demonstrates another dimension of the same argument. As factories, energy networks and logistics systems become more digital, protecting industrial technology is increasingly inseparable from deploying it. Cooperation between the UAE Cyber Security Council and Siemens already includes plans for a joint centre focused on operational technology cybersecurity, the systems that control critical infrastructure and industrial processes.

Clean energy belongs in the technology story too.

Hydrogen, battery storage, smarter electricity grids and industrial decarbonisation all depend on engineering and technological deployment as much as energy policy. Germany brings industrial technology and a large market seeking lower-carbon solutions; the UAE combines energy expertise with capital and an increasing presence in renewable power. The potential is therefore not simply for one country to supply energy to the other, but to develop technologies, manufacturing capacity and supply chains around the transition.

The state visit is intended to give this technological relationship more structure. Her Excellency Nusseibeh said the two countries would establish a dedicated framework for deeper cooperation on technology and digitalisation, bringing these areas into the core of the bilateral relationship. The objective, she said, is to connect “researchers, universities, entrepreneurs and investors” and turn ideas into scalable businesses–an ambition which could make technology far more than another area of expanding UAE-German cooperation, but rather an organising principle for the next phase of the two countries’ relationship.

The post Why Technology Could Define The Next Phase Of UAE-Germany Relations appeared first on 91̽��.

While that figure sits below the global average of 2,422 weekly attacks per organisation, it confirms that UK businesses remain firmly in attackers’ sights, with education, energy and utilities, media and entertainment, government and software named as the five most targeted sectors in the country.

The UK figures form part of Check Point’s wider Global Threat Intelligence report for August, which found that cyber-attacks worldwide rose 4% month-on-month and 22% year-on-year. The report points to an escalation in risk across several fronts at once, with ransomware, phishing and GenAI-related data exposure all identified as pressing enterprise security challenges.

Education And Travel Sectors Under Pressure

Globally, education remained the most targeted sector, averaging 5,354 weekly attacks per organisation, up 28% year-on-year. Government followed with 3,067 weekly attacks, while Hospitality, Travel and Recreation climbed into third place with 3,056 weekly attacks, up 56%, as attackers appeared to capitalise on the busy summer travel period.

Regionally, Latin America recorded the highest volume of attacks at 3,577 per week, while Europe posted the fastest year-on-year growth at 28%, suggesting that cyber pressure is spreading into mature, interconnected digital economies rather than remaining concentrated in traditionally high-volume regions.

��

Genai Use Climbs As Data Exposure Risk Lingers

One of the report’s key findings centres on the growing use of generative AI tools inside businesses. High-risk GenAI prompts, those posing a risk of exposing sensitive data, fell to their lowest level in several months, with just one in every 43 enterprise prompts now flagged as high-risk.

At the same time, overall GenAI usage kept climbing. The average enterprise user generated 106 prompts in August, up from 95 in July and around 78 in June. Check Point notes that 86% of organisations using GenAI regularly were still affected by high-risk prompt activity at some point, and that the average organisation now uses seven different AI tools, a spread that makes consistent governance harder to maintain.

Healthcare and Medical organisations recorded the highest exposure rate of any sector, at 4% of prompts, followed by Software at 3.6% and Business Services at 3.5%. The data most commonly exposed in GenAI prompts related to network and IT infrastructure, financial data and legal or regulatory information, underlining how far AI-related risk now extends across core business functions rather than sitting at the margins.

Phishing And Ransomware Both On The Rise

Email remained a favoured route into organisations, with one in every 112 emails classified as phishing in August, up from one in 128 in July. Links featured in 72% of phishing emails, while 14% carried malicious attachments.

Ransomware activity accelerated sharply, too. Check Point recorded 1,042 ransomware attacks globally in August, almost double the number seen in August 2025 and 8% higher than July. Business Services bore the brunt, accounting for 36% of reported attacks, ahead of Industrial Manufacturing at 13% and Consumer Goods and Services at 12%. The United Kingdom featured among the countries most affected by ransomware during the month, behind the United States, Germany and Italy.

Qilin was the most active ransomware group in August, responsible for 15% of published attacks, followed by The Gentlemen at 10%. Orova, a group that only surfaced publicly in May 2026, broke into the top three for the first time with 4% of attacks, most of them concentrated among small and mid-sized businesses.

Barnaby Nickels, Head of Sales, Exposure Management, UKI & North EU at Check Point, said: “August’s data shows cyber risk expanding across several fronts at once. With attacks climbing, ransomware accelerating, phishing remaining a common entry point and GenAI creating a new route for data exposure, security teams cannot rely on fragmented defences.”

He added that organisations need proactive detection and validation of threats, alongside a focus on cutting remediation times, warning that the window attackers have to exploit a vulnerability is shrinking fast.

For UK businesses, the message from Check Point’s latest figures is consistent with the broader global picture: attack volumes, ransomware and AI-related data exposure are all rising together, and the company is urging organisations to move toward a prevention-first strategy that gives consistent visibility across users, email, networks, cloud environments and AI tools.

The post UK Firms Hit By Over 1,500 Cyber Attacks A Week As Ransomware Nearly Doubles Globally appeared first on 91̽��.

How a Warehouse Shuttle System Works

The system uses automated shuttles that travel through multi-level racking and transport boxes or totes between storage locations. Vertical lifts connect different levels, while conveyors move containers between the storage area and picking, packing or dispatch stations.

When a product is required, the warehouse management system identifies its location and sends a retrieval task. The shuttle collects the corresponding container and transfers it to a lift. From there, the box moves by conveyor directly to the required workstation. After picking, it can be returned automatically to storage or sent to another warehouse zone.

This creates a Goods-to-Person workflow: employees remain at dedicated workstations while the system delivers the required inventory to them.

Compact Storage And Faster Processing

A shuttle system allows warehouses to use vertical space more efficiently because containers can be stored across several rack levels without conventional access aisles for workers. This is particularly useful for facilities with limited floor space and a large number of SKUs.

The modular configuration also allows capacity to expand as warehouse requirements change. Additional rack levels, shuttles, lifts and picking stations can be integrated without redesigning the entire storage concept.

One Connected Automated System

Shuttles operate together with racking, conveyors, lifts and warehouse software. This integration coordinates container movement, storage locations and picking priorities within one automated process.

As a result, a warehouse shuttle system combines high-density storage, automated retrieval and Goods-to-Person picking in a single solution suitable for e-commerce, retail, pharmaceutical, manufacturing and distribution operations.

The post Warehouse Shuttle System For Automated Storage And Retrieval Explained appeared first on 91̽��.

Discovery begins on a phone, frame selection may involve a live camera, prescription information can move securely between systems and a lens laboratory can translate a digital order into a highly individual product.

This is more than the story of another high-street category moving online. Spectacles sit at an unusual meeting point of healthcare, fashion, precision manufacturing and personal identity.

A jumper that is slightly loose can still be worn; lenses made to the wrong specification may cause discomfort and are not fit for purpose. The technology therefore has to achieve two things at once: remove needless friction and preserve the safeguards that make optical care trustworthy.

The wider commercial backdrop is already digital. Office for National Statistics data show that internet sales accounted for 27.6% of retail sales in Great Britain in May 2026. That figure moves from month to month, but the direction of travel is clear: online retail is no longer a specialist channel. Consumers now expect broad choice, transparent prices, delivery updates and straightforward returns. Eyewear businesses are being measured against that standard while handling a far more technical product.

There is also a human need behind the market. The World Health Organisation estimates that at least 2.2 billion people worldwide have a near or distance vision impairment and says that in at least one billion cases the impairment could have been prevented or has yet to be addressed. Technology cannot replace trained clinicians or solve unequal access by itself. It can, however, make information, selection and fulfilment less expensive and more convenient.

From Catalogue To Guided Digital Shop

The first generation of online eyewear shops largely reproduced a paper catalogue on screen. Customers could filter by colour or price, but still had to imagine how a frame would look and feel. Contemporary platforms are closer to guided retail environments. Search tools can narrow hundreds of products by width, bridge type, material, colour and face shape. Good product pages show measurements clearly, explain lens choices in ordinary language and allow the customer to compare the total price before reaching the checkout.

This matters because too much choice can be as unhelpful as too little. Recommendation systems can use stated preferences and browsing behaviour to surface a manageable selection. The responsible use of such systems is not to declare that an algorithm knows a customer’s taste better than they do.

It is to reduce repetitive searching while leaving the decision with the wearer. Retailers should also make it easy to reset recommendations and browse outside the suggested range, so personalisation does not become a narrow corridor.

For households balancing cost, time and access, the ability to compare and purchase can be genuinely useful. A rural customer may have fewer nearby frame ranges; a shift worker may struggle to shop during standard opening hours; and a parent replacing a familiar pair may value speed.

Yet convenience should be presented as part of an eye-care journey, not as a substitute for a current prescription or an appropriate sight test. That distinction is central to building a durable online optical brand.

Virtual Try-On Becomes Genuinely Useful

The most visible innovation is augmented-reality try-on. A camera identifies facial landmarks, estimates the position and scale of the face, and places a three-dimensional representation of the frame over the live image. Earlier versions were entertaining but approximate. Better cameras, depth sensing and improved computer-vision models have made the experience steadier and more persuasive, particularly for judging colour, general shape and relative frame width.

Virtual fitting has a longer commercial history than many shoppers realise. Wikipedia’s account of DITTO, an early specialist in the field, describes software that measured facial features and produced images of customers in different frames; the company was founded in 2011 and later became part of a larger optical-technology business. The history is useful because it shows that today’s tools are the result of more than a decade of iteration, not a novelty suddenly produced by generative AI.

Even now, a virtual mirror has limits. Camera perspective can distort apparent size. A screen cannot reliably communicate weight, pressure at the temples or whether a bridge will slip during a commute. Colour also varies between displays. Retailers should describe the feature accurately: it is a decision aid, not a clinical measurement and not a guarantee of fit.

Clear dimensions, helpful photography, human support and a fair returns or adjustment process remain essential.

The next step is likely to be a better connection between visualisation and fit data. With explicit consent, facial measurements could help rank frames by likely suitability or flag an unusually wide or narrow fit. Because facial data can be sensitive, businesses must be restrained about collection and retention. A useful principle is simple: collect only what the service needs, explain the purpose plainly and do not quietly repurpose a face scan for unrelated marketing.

The Quiet Revolution In The Laboratory

The cleverest part of online eyewear is often invisible to the customer. Once an order reaches the laboratory, software validates the prescription and selected lens, checks whether the chosen frame can accommodate it and creates instructions for production.

Lens blanks are blocked, surfaced or selected, coated where required, and edged to the precise outline of the frame. Barcode and order-management systems help the correct lenses, frame and customer record travel together.

Modern free-form surfacing uses computer-controlled equipment to create complex lens surfaces at very fine tolerances. This is particularly valuable for progressive lenses, where the viewing zones and transitions must be calculated for an individual prescription and fitting position.

Digital tracing allows the frame shape to be captured and passed to an edging machine. Automated inspection can support technicians by identifying cosmetic faults or confirming dimensions, although trained quality control remains important.

For independent British retailers, cloud-based order systems can connect the website, customer-service team, optical review and laboratory. That reduces re-keying, one of the most ordinary and costly sources of mistakes. It also creates a useful audit trail: when information was received, which checks were completed, when glazing began and why an order was held. Customers benefit through clearer updates rather than vague promises that their glasses are ‘being processed’.

There is a broader operational lesson here. Digital transformation is rarely one dramatic invention. It is the removal of small uncertainties across a chain. A sensible prescription form, an automatic warning for missing information and a well-designed laboratory dashboard may create more value than a fashionable chatbot placed on an unreliable process.

AI: Capable Assistant, Poor Substitute For Accountability

Artificial intelligence is beginning to influence demand forecasting, customer support, product tagging, fraud detection and quality inspection. Computer vision can identify frame attributes from photographs. Language systems can turn technical lens descriptions into clearer explanations. Forecasting tools can help a retailer decide which colours and sizes to hold, potentially reducing dead stock and unnecessary transfers.

Used carefully, AI can give human teams more time for work that needs judgement. A support system might summarise an order history before an adviser responds, or flag an unusual combination of prescription and lens choice for review. It should not make unsupported clinical claims, invent prescription advice or obscure the route to a qualified professional. In a regulated, health-adjacent category, the person or business making a decision remains accountable even when software helped to reach it.

Bias also deserves attention. A frame-recommendation model trained on an unrepresentative image set may work less well across different face shapes, skin tones, ages or accessibility needs. Testing should therefore look beyond average performance. Retailers need a clear fallback when automation is uncertain, and customers should be able to speak to a person without navigating an obstacle course.

Regulation Is Part Of The Product

British optical retail is not a regulatory blank sheet. The College of Optometrists’ professional guidance explains that sales involving certain groups including children under 16 and people registered as sight impaired require professional supervision, while the Opticians Act governs the sale and supply of optical appliances. The exact obligations depend on the customer and product, so retailers need processes that recognise restricted cases instead of treating every basket identically.

A sight test also does more than generate numbers for a pair of lenses. NHS guidance notes that an optometrist checks eye health as well as vision and, where necessary, can refer a patient for further investigation. An ecommerce journey should reinforce the value of routine examinations and make prescription requirements clear. It should also offer an obvious route for customers whose details are incomplete, expired or unsuitable for automated fulfilment.

Research provides a useful note of caution. A 2016 peer-reviewed UK study comparing spectacles bought online with those supplied by optometry practices found similar lens quality and prescription accuracy overall, but participants preferred practice-supplied spectacles and more online pairs were judged unacceptable on certain measures. The study is not a verdict on every current retailer, technology and processes have advanced but it highlights why fit, verification and aftercare cannot be treated as optional extras.

Trust is therefore built in layers: secure handling of health-related information, truthful product descriptions, qualified review where required, clear delivery expectations and a practical remedy when something is not right. These are not merely compliance costs. In online retail, they are core features.

Affordability, Inclusion And The Hybrid Future

Technology can lower some costs by centralising stock, automating administration and enabling laboratories to process orders efficiently. Price comparison also becomes easier. But affordability is not achieved by showing a low headline price and revealing necessary lens charges at the final step.

The strongest retailers explain what is included, distinguish optional upgrades from genuine requirements and help customers understand the trade-offs without pressure.

Digital inclusion matters too. Small text, low contrast or a checkout that cannot be navigated by keyboard is especially inappropriate for an optical website. Pages should work with screen readers, permit magnification and avoid using colour alone to communicate status. Instructions for entering a prescription need examples and validation that is helpful rather than scolding.

Telephone or human-assisted ordering can support customers who are less confident online.

The future is unlikely to be a contest in which websites eliminate practices or practices reject ecommerce. It is hybrid. A customer may have a sight test locally, research frames at home, ask a question by chat, order remotely and visit a partner location for adjustment. Another may discover a frame online and complete the purchase in person. Shared records and consistent service can make those movements feel like one relationship rather than a collection of disconnected channels.

Physical practices have strengths that software cannot reproduce: examination, hands-on fitting, immediate reassurance and community presence. Digital businesses offer range, convenience and scalable information. The most credible operators will combine these advantages and be candid about where each channel is strongest.

A Greener Industry Needs Evidence, Not Slogans

Digital operations can reduce paper, improve stock planning and consolidate some journeys, but ecommerce is not automatically sustainable. Individual deliveries, returns, packaging, lens waste and unsold frames all carry an environmental cost. Equally, a high-street purchase is not impact-free. The proper question is which design and operational decisions reduce waste across the entire product life cycle.

Technology can help by forecasting demand at size-and-colour level, producing selected components closer to demand and recording material provenance. Better fit information may prevent avoidable returns. Durable hinges, replaceable parts and accessible repair services can extend useful life. Retailers should publish specific evidence such as packaging composition, repair policies or audited emissions rather than relying on broad terms such as ‘eco-friendly’.

Circular models are also emerging: frame take-back, refurbishment, recycling trials and materials made partly from recovered or bio-based feedstocks. None is a universal answer. Mixed materials and very small components make spectacles challenging to recycle, while a frame with a lower-impact material is not sustainable if it breaks quickly. Longevity remains an unfashionable but powerful measure.

Data Quality And Security Are Competitive Advantages

An online optical service depends on information that is both accurate and personal. A prescription may include sphere, cylinder, axis, prism and addition values, while the order can also contain pupillary distance, fitting information, contact details and payment records.

A form that merely accepts any combination of characters is not digital transformation; it is a mistake waiting to enter production. Thoughtful validation should recognise common formats, distinguish a blank field from a genuine zero and ask for confirmation when the right and left eyes appear to have been transposed.

Automation must nevertheless allow for legitimate variation. Prescriptions are not always laid out in the same way, and an unusual value is not automatically an error. The sensible model is risk-based: software catches missing or inconsistent information, then a trained person resolves exceptions. Customers should be able to upload a clear copy or authorise verification where appropriate, while the retailer records which version was used for manufacture.

Security deserves the same practical attention. Optical retailers should limit staff access by role, protect accounts with strong authentication, encrypt information in transit and maintain tested recovery procedures. Suppliers that host customer-service, analytics or laboratory systems need scrutiny too; a polished storefront does not compensate for weak links behind it.

Retention periods should be defined rather than allowing sensitive records to remain indefinitely because storage is cheap.

Transparency is equally important. Privacy notices are most useful when they explain, in plain English, which information is required to fulfil an order, which features are optional and whether images or facial measurements are retained. Consent for a virtual try-on should not be bundled invisibly with consent for unrelated advertising. In a market where products can look similar and prices are easily compared, careful stewardship of customer data can become a meaningful point of difference.

What The Winners Will Get Right

The technology transforming eyewear will continue to improve. Virtual try-on will become more accurate, manufacturing systems more connected and AI more capable of supporting routine decisions. Smart glasses may also blur the boundary between optical appliance and computing device; Wikipedia’s overview of glasses traces how eyewear has repeatedly changed in materials, design and purpose over centuries. Yet innovation will not remove the fundamentals.

Customers want to see well, feel comfortable, understand what they are buying and know that somebody competent will help if there is a problem. Businesses that use technology to meet those needs; not to disguise their absence will earn repeat custom. That means investing as seriously in prescription checks, accessibility, cybersecurity and aftercare as in the front-end experience.

Britain is well placed for this next phase. It has mature ecommerce habits, established professional standards, skilled optical teams and consumers who are comfortable moving between digital and physical channels. The opportunity is not simply to put more frames on more screens. It is to create a joined-up service in which data moves accurately, choices become clearer and professional care appears at the moment it is needed.

The best version of online eyewear will feel less like a technological performance and more like good service: convenient without being careless, personalised without being intrusive, and efficient without losing its human touch.

The post How Technology Is Transforming The UK Eyewear Industry appeared first on 91̽��.

Decodo’s own research points to the same driver. AI tools read the web at machine speed, and that speed explains most of the new bot traffic.

AI-driven traffic grew about 187% across 2025, roughly 8 times faster than human traffic growth, per��. Agentic AI traffic, software that acts on a person’s behalf, grew about 7,851% year over year in the same report.

A person shopping for a camera might open a few websites. An AI agent doing the same task can query thousands of pages, turning one user request into thousands of page fetches. That gap in scale is what pushed bots past humans.

What Decodo’s Data Shows About Who Is Behind The Traffic

Automated traffic splits into three groups. Training crawlers collect data to build and update AI models, and make up the largest slice. AI agents and fetchers pull live pages to answer a prompt or complete a task. Malicious bots run scraping, credential stuffing and fraud.

Decodo’s research puts the malicious share at roughly a third of all bot traffic, leaving about two-thirds doing legitimate work, like search indexing and AI answers.

A handful of companies run most of that legitimate traffic. OpenAI’s bots account for about 69% of observed AI-driven traffic by volume, Meta about 16%, and Anthropic about 11%,��per HUMAN Security. Three sectors absorb more than 95% of that traffic; retail and eCommerce, streaming and media, and travel and hospitality.

Retail and eCommerce alone made up 62.5% of training crawler traffic.

Decodo’s Network Data On Where The Bots Go

Decodo tracked its own network traffic over six months across more than 195 countries, and the pattern lines up with the global figures. Search engines and AI assistants generate about 72% of the traffic Decodo observes.

Retail and eCommerce sites make up about 13% of successful requests, while travel, airlines, and cargo sites trail far behind at 0.4%. News, real estate, jobs, and finance sites each account for a fraction of a percent. Decodo’s data points to the same conclusion as the country figures below: bot activity follows commercial value, not traffic for its own sake.

The same commercial logic shows up at the country level. Decodo’s analysis of Cloudflare Radar data found that six of the top 60 traffic source countries already run more bot traffic than human traffic, led by markets with dense cloud infrastructure rather than large populations.

• Iran leads at 81.4% bot traffic
• Singapore follows at 73.7%
• Ireland sits at 71.1%
• The Netherlands runs 61.3% bot traffic
• Finland runs 56.8%

The United States originates 53.5% of worldwide bot traffic and runs 43.6% of bot traffic at home. Germany, the second largest source, runs 45% domestically. Inside the US, Virginia alone accounts for 27% of national bot traffic, followed by California at 9.9% and Oregon at 7.7%.

Traffic carries the flag of the data center, not the person behind the request. A reader in Lagos can still route through a server in Ashburn or Frankfurt, and show up in the data as German or American traffic.

Why Blocking Every Bot Backfires

AI assistants and shopping agents now decide what customers see online, and blocking bots removes a business from those results. In 2025, 77% of agentic AI activity hit product and search pages, with smaller shares on account pages, authentication, and checkout, per HUMAN Security.

Identity is part of the problem. Bots can hide their name, and in August 2025 Cloudflare accused AI firm Perplexity of disguising its crawler as a browser.

Some publishers have responded by charging bots for access instead of blocking them outright. The shift rests on a simple ratio: pages taken against visits sent back. Google crawls about five pages for every referral it sends. Some AI crawlers pull thousands of pages for each visit they return, which is why publishers increasingly price access instead of giving it away.

Vaidotas Juknys, CEO at Decodo, said blocking every bot is “like locking your storefront because some visitors don’t buy.” He added that the agents crawling a site today might be how customers discover it tomorrow.

What Businesses Should Do Next And Where The Web Scraping API Fits

The response depends on the team, and each one needs a different move.

Product And Engineering Teams

Teams building AI agents need reliable, real-time access to web data across regions. An agent that cannot reach a site, or reads a blocked or geo-shifted version of it, returns a worse answer. Decodo built its����for exactly this problem, giving agents, LLMs, and tools the ability to fetch real-time results across more than 195 countries.

The API configures scraping power per request, so a team pays for the capabilities a job actually needs, instead of a flat rate regardless of complexity. Routing requests through local IPs matters just as much, since a product page in Madrid can differ from the same page in Chicago.

An agent that always exits through one region reads one version of the web and misses the rest. Fetching the page a local user would see is what keeps an agent’s answer accurate.

Vaidotas Juknys added, “Reliable access to public web data has become as fundamental as cloud computing.” For teams building on AI, that access is now doing the heavy lifting.

Marketing And Analytics Teams

Agent traffic is a new audience showing up inside your analytics. Tag automated traffic separately from human traffic, and read AI referral activity on its own instead of folding it into session counts. Track which AI tools send the most agent visits.

Security Teams

The job shifts from blocking everything to sorting good bots from bad ones. Set rules based on verified identity and behaviour, not user agent strings alone. Allow crawlers that drive discovery, rate limit unknown traffic, and stop patterns tied to credential stuffing and scraping fraud.

Bots now send more web requests than people do. Iran, Singapore, and Ireland already run bot majorities and AI-driven traffic keeps growing about 8 times faster than human traffic. Decodo’s own network data shows the same pattern: bots concentrate where the commercial value sits. Businesses that sort bots instead of blocking them and give their own agents reliable access through tools like the Web Scraping API stay visible in AI search results.

The post Bot Traffic Vs Human Traffic: What Decodo Found appeared first on 91̽��.

The difference between the two rarely comes down to budget, it comes down to how well they prepared. Without careful thought and planning, a trade show success could quickly become a trade show disaster.

Why Trade Shows Are Still Worth It

In an age of LinkedIn outreach, social media and digital advertising, it’s easy to wonder whether trade shows and expos still have a place in a marketing strategy. The answer, for most businesses, is a clear yes.

There’s something that happens in a face-to-face environment that no digital channel can fully replicate. The ability to read a room, build rapport in real time and leave someone with a genuine impression of who you are and what your business stands for.

In a world full of digital ads, this personal connection is a great way to get ahead of the competitor. Stand out from the crowd and be memorable.

According to , 71% of SMEs have generated new business through face-to-face networking at trade shows and that is a compelling reminder that despite the rise of digital marketing and advertising, in-person events remain one of the most effective business development tools available to smaller companies.

Getting Your Trade Show Stand Right

Trade shows and exhibitions are the great equaliser. No matter if you are an established company or a startup just starting a business, everyone is equal here.

Your stand is the first thing people see when they attend a trade show. Trade shows are a busy place, so brands only have a matter of seconds to catch someone’s attention before the attendee moves on to the next booth.

Expensive does not mean the stand will automatically be successful or effective. Clarity beats complexity every time. A clean, uncluttered display with a clear message about what you do and who you do it for will almost always outperform an overcrowded booth trying to say too much at once.

A few things worth getting right before the show:

• Signage: make sure your brand name and core message are visible from a distance, not just up close. Consider the readability as well
• Layout: avoid large tables that create a barrier between your team and visitors; open, welcoming setups tend to invite more conversation. A table for goodies or information is okay, but make sure it does not hinder the personal connection and conversation
• Lighting: additional lighting can make a stand feel warmer and more professional without a significant cost. Make the experience of visiting the booth a positive one with a well-lit area.
• Consistent branding: colours, fonts and tone should feel consistent across every element of your stand. Many times colours, font and the overall corporate identity is one of the first and only things people remember when visiting a booth. Make sure it is consistent across all mediums

Building a strong trade show presence is part of a broader approach to commercial awareness, understanding how your business is perceived in the market and making deliberate choices about how you show up in it.

How To Draw People Into The Stand

Getting your stand right is only half the battle. The other half is what happens when someone walks past and whether your team can turn a glance into a conversation.

This is where many businesses lose ground, not because of their product or their stand, but because of how they engage. Waiting for visitors to approach, staring at phones or launching straight into a sales pitch are all common mistakes that can be easily avoided with a little preparation.

Tips For How To Ease Into Conversations

1. Leading with a question rather than a pitch is almost always more effective, asking something relevant and open-ended is far more likely to start a genuine conversation than a rehearsed introduction
2. Where possible, live demonstrations tend to do the same job even better, showing what you do is almost always more compelling than explaining it
3. All employees should know and understand the key brand messages, the target audience and how to qualify a lead naturally. Trade shows are long days and energy levels matter, rotating staff regularly keeps the team fresh and the conversations genuine
4. Every interaction should end with a clear next step. Whether that’s a follow up call, a shared contact or something physical to take away, leaving a conversation open rarely leads anywhere

Trade Show Giveaways That People Actually Keep

A giveaway is only as good as its staying power. The goal isn’t to hand out as many items as possible but to give people something they’ll actually use, something that keeps your brand visible long after the show is over.

The most effective trade show giveaways tend to have one thing in common, they’re genuinely useful in everyday life. An item that earns a place on someone’s desk, in their bag or on their kitchen counter is one that’s working for your brand every single day without any additional effort or cost on your part.

Practicality should always come before novelty. A quirky giveaway might raise a smile on the day but rarely survives the journey home. A well-chosen practical item whether that’s a quality notebook, a reusable cup or one of the many that can be printed with your logo will outlast the event and keep your brand in circulation long after it ends.

Following Up After The Show

The work doesn’t stop when the stands come down. For many businesses, the follow-up is where the real value of a trade show is either realised or lost and it’s the part that most businesses handle least effectively.

The most common mistake is waiting too long. Leads go cold quickly after an event, and the longer you leave it, the less likely someone is to remember the conversation you had or the impression you made. Ideally, follow-up should begin within 48 hours of the show ending while the experience is still fresh on both sides.

Personalisation matters here too. A generic “great to meet you” email is easy to ignore. Referencing something specific from the conversation shows that you were genuinely listening and sets you apart from the stack of other follow-ups landing in their inbox at the same time. While you can make use of a template, make sure to always leave room for more personal moments and quips to make each email unique.

Not every lead will convert immediately, and that’s fine. The goal of the first follow-up is simply to keep the conversation going, to move from a trade show introduction to a proper business relationship. Every new client relationship has real significance, that patience and consistency in follow-up is often what separates the businesses that get the most out of trade shows from those that don’t.

Trade Shows Reward The Prepared

For businesses, a trade show is rarely just a networking event, it’s one of the most concentrated opportunities of the year to build relationships, generate leads and raise your profile in front of exactly the right people. But that opportunity is only as valuable as the preparation behind it.

The businesses that walk away with the most aren’t always the ones with the biggest stands or the largest budgets. They’re the ones that thought carefully about every detail in advance, how their stand looks, how their team engages, what they leave people with and how they follow up when it’s all over.

Get those things right and a trade show stops being an expense and starts being an investment.

The post Outshine The Competition: Trade Show Ideas Every Business Should Know appeared first on 91̽��.

2025 marked a pivotal moment for the gaming industry when a US court ruling forced Apple to let publishers offer in-app transactions directly to their users, bypassing the 30% transaction fee. Within a year, direct-to-consumer (DTC) has already become a booming channel, accounting for around 15% of the $113.3Bn mobile gaming in-app purchase market; approximately $17Bn today.

The potential for DTC is proven. GDC Festival of Gaming research, commissioned by Appcharge, reveals that early adopters are seeing a median revenue uplift of 35%, with 63% already outperforming app store monetisation. Today, any publisher with an alternative payment system can expect to generate a double-digit margin increase and further benefits of direct selling.

This is a beacon to the wider app economy, such as AI creation, fitness, education, lifestyle, entertainment and other subscription services. With the global in-app market across all consumer apps projected to reach $290Bn by 2030, it’s a multi-billion opportunity for publishers. Brands that haven’t already are scrambling to get started.

But leaving the safe haven of the Apple/Google model might feel risky to app publishers starting out on their DTC journey.

To succeed in the evolving payments labyrinth, app developers must make DTC a strategic priority armed with dedicated senior leadership, a DTC strategy and the right DTC infrastructure. Only then will they make the most of this key channel and take control over where the business goes next.

The DTC Maturity Gap

We’re seeing a reckoning in DTC with a widening revenue gap between leaders and laggards at around 7x.��According to the Appcharge report, “innovator” studios are charging ahead with a median DTC revenue share of 35%, compared with “late adopters” at 5%. It’s the same market and players, yet this DTC maturity gap opened in less than 12 months. This means 35% of their total revenue has shifted from traditional billing systems that take 30% fees on every transaction to a DTC vendor that takes around 5%. For large publishers, this represents a huge amount of revenue.

Optimism for DTC growth is indisputable. 67% of publishers expect DTC revenue to grow in 2026, yet only 25% describe their DTC strategy as scaling or mature. Most of the industry expects the channel to grow while simultaneously admitting they’re not ready for it.

It’s a steep learning curve, with only 14% of publishers considering themselves ahead of the industry on DTC while 62% say they’re behind. Publishers realise they need to be in the race, but it’s one that’s already started.

Opportunities And Challenges In Moving To DTC

Top objectives for investing in DTC include increasing revenue (63%), building direct relationships with players (53%), improving monetisation (45%), and reducing dependency on app stores (40%).

Success is proven, with early adopters of the DTC channel reporting rising revenues and expecting sales to continue to accelerate. Direct interaction with users and their data is helping to improve engagement, control pricing, generate personalised offers, deepen relationships, and stay ahead of competitors.

On the flip side, challenges are largely operational, with publishers facing demands the app store model never required. Before DTC, publishers didn’t need to think much about payment processing, global tax compliance, fraud, chargebacks and refunds. The app stores took care of it, as part of their 30% fees. Processing payments outside of these traditional billing systems shifts the responsibility to the publisher, who either has to build those capabilities in-house or use a specialist DTC payments vendor.

Scaling is also a challenge, as is having dedicated teams and the right DTC payments infrastructure. DTC leadership isn’t just a one-department decision. It spans product, payments, customer success, finance, and legal. It requires leadership sponsoring it at the top and pushing it across the organisation.

Beyond ownership, it’s essential to have the right skills in the form of DTC expertise. 54% of studios have zero dedicated DTC employees.��Among innovators, the average number is 14. The organisational gap between innovators and late adopters largely explains the revenue gap.

For smaller studios, the main concerns are friction around integration complexity, internal bandwidth and how much this will demand of the team. They simply don’t have the headcount to absorb a complicated implementation.

For enterprise apps, it’s about feature parity. They want the DTC experience to match what they have in-app, with the same offers, the same economy, the same feel. This sets the technical bar high, as understandably publishers want to match the sophistication of their in-game monetisation systems.

The Challenges Of Scaling DTC

Operations is not an easy landscape to move into and make a success of overnight. The moment you take payments outside the app store, you inherit issues the platform used to handle. This might be authorisation logic, fraud prevention management, dispute handling across dozens of markets, and managing tax and VAT across territories. Publishers are now handling these areas directly for the first time.

Therefore, a review of the app payments integrations and journeys must be an initial step in any DTC strategy. DTC owners can then make informed investments and avoid costly missteps that can compromise success. Even more so for first-time DTC strategists, the merchant of record (MoR) model can now replace that infrastructure in a single relationship without the 30%.

Key Features Of A Winning DTC Operation

At a technical level, all apps require different DTC infrastructure. A that’s experienced in handling the tough problems can design a solution that will optimise revenue potential for that specific business. Strategically, some other things must also line up for success.

Here are some consistent features of those publishers doing DTC strategy well:

• Senior ownership –��This should be one person who reports to a C-level and is accountable for DTC as a function, not just a project that straddles multiple teams with no clear home
• DTC is embedded in the business operating system –��This means sitting within the budget and featuring in revenue projections with annual targets and KPIs. It must be treated as a core channel, rather than being evaluated quarter on quarter as to whether it “deserves” more investment
• The incentive structure is right –��The monetisation team, the AMs, customer success, all incentivised by the success of DTC, not just by app store performance. When the people closest to the customer/player relationship have a stake in DTC growth, the whole organisation pulls in the same direction
• There’s committed resource allocation –��Not huge numbers, but dedicated time from the people whose decisions actually shape the player experience. DTC touches various teams across the business, such as development, product, and customer success

A Sizeable Market Opportunity

Gaming has been the DTC proving ground for the rest of the app economy. Some of the next wave of brands in lifestyle, food and drink, health and fitness, shopping and finance are already making headlines around their DTC monetisation. This model can work for any high-velocity consumer business with global users and complex tax implications.

As the new world of DTC advantage opens up, it’s a given that publishers will meet complexity with DTC in-app payments. From high-frequency micro-transactions and virtual economies to global customer bases and real-time LiveOps.

For those fast-scaling app publishers already taking advantage of the channel, building smarter operations and tackling DTC mismanagement are the first goals. For beginners, a robust plan and a perfect blend of the right tech and human expertise is the winning formula.

The post The Reckoning In App Monetisation: How DTC Can Drive Profitability For The Broader App Market appeared first on 91̽��.