{"id":826,"date":"2026-10-20T18:02:58","date_gmt":"2026-10-20T18:02:58","guid":{"rendered":"\/news\/?p=826"},"modified":"2026-10-20T18:02:58","modified_gmt":"2026-10-20T18:02:58","slug":"ai-business-productivity-what-works-2026","status":"publish","type":"post","link":"\/news\/ai-business-productivity-what-works-2026\/","title":{"rendered":"AI and Business Productivity: What Works in 2026"},"content":{"rendered":"<p><span style=\"font-weight: 400;\">In 2026, AI is already improving individual productivity, but not yet that of most companies. According to <\/span><a href=\"https:\/\/www.mckinsey.com\/capabilities\/quantumblack\/our-insights\/the-state-of-ai\"><span style=\"font-weight: 400;\">McKinsey&#8217;s global survey published in August<\/span><\/a><span style=\"font-weight: 400;\">, 80% of respondents see individual productivity gains. However, only 37% can attribute any impact on operating income (EBIT) to AI, the same proportion as a year earlier.<\/p>\n<p><\/span><\/p>\n<h3><b>Where is AI being applied to gain productivity?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Today, AI performs best in text, data, and repetitive customer service tasks. These are high-volume jobs with clear rules and a human who can quickly review the output.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The most widespread uses in companies are:<\/span><\/p>\n<ul>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Writing and summarisation:<\/b><span style=\"font-weight: 400;\"> draft emails, sales proposals, meeting summaries, and long document overviews.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Marketing and sales:<\/b><span style=\"font-weight: 400;\"> content pieces, database segmentation, and client call preparation.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Customer service:<\/b><span style=\"font-weight: 400;\"> suggested agent responses, ticket classification, and searching the internal knowledge base.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Software development:<\/b><span style=\"font-weight: 400;\"> code generation and review, testing, and technical documentation.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Analysis and reporting:<\/b><span style=\"font-weight: 400;\"> natural language data queries and creating reports that previously took hours on a spreadsheet.<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">The change in 2026 is <\/span><b>agents<\/b><span style=\"font-weight: 400;\">: systems that do not just respond, but execute a sequence of steps (search, complete, send, log). McKinsey notes that 40% of large companies are already scaling agents, compared to 22% of smaller ones. Where they report the most cost savings is in supply chain, service operations, and manufacturing.<\/span><\/p>\n<p><span style=\"font-weight: 400;\"><\/p>\n<p><\/span><\/p>\n<h3><b>Why isn&#8217;t individual productivity reaching the bottom line?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Because saving 30 minutes per person changes nothing if the process remains the same. The freed-up time gets diluted in more meetings, more emails, or simply another unmeasured task.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Macro data shows the same tension. A <\/span><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0954349X26000937\"><span style=\"font-weight: 400;\">study by Katz and Jung<\/span><\/a><span style=\"font-weight: 400;\"> published in Structural Change and Economic Dynamics concludes that generative AI raises productivity across all worker groups, between 2.5% and 5.7% once adoption reaches critical mass. However, the <\/span><a href=\"https:\/\/www.stlouisfed.org\/on-the-economy\/2026\/jul\/ai-productivity-what-firms-say-earnings-calls\"><span style=\"font-weight: 400;\">Federal Reserve Bank of St. Louis<\/span><\/a><span style=\"font-weight: 400;\"> notes that total factor productivity in the US grew by just 0.07% over the four quarters ending in March 2026.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The reasonable interpretation: AI works at the task level, and the bottleneck lies within the organisation. McKinsey confirms this from another angle. Only 6% of companies qualify as high performers (more than 5% of EBIT attributable to AI), and that group stands out for redesigning entire workflows instead of adding tools to old processes. Furthermore, 20% already see the costs of operating AI, such as token consumption, as a drag.<\/span><\/p>\n<p><span style=\"font-weight: 400;\"><\/p>\n<p><\/span><\/p>\n<h3><b>How is AI adoption progressing in Australia?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">It is growing, but unevenly. According to the <\/span><a href=\"https:\/\/www.ai.gov.au\/news-and-insights\/blog\/ai-adoption-insights-december-2025-february-2026\"><span style=\"font-weight: 400;\">SME AI Pulse from the National AI Centre<\/span><\/a><span style=\"font-weight: 400;\">, a monthly survey of at least 400 Australian SMEs, 44% were using AI in February 2026. Among those that have already adopted it, broad usage across several areas simultaneously is growing, while single-use adoption is dropping. More complex applications, like agents, remain largely untapped.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">\u00a0<\/span><\/p>\n<table style=\"width: 390px;\">\n<tbody>\n<tr>\n<td style=\"width: 209.262px;\"><b>AI use in Australian SMEs<\/b><\/td>\n<td style=\"width: 167.137px;\"><b>% of adopting companies<\/b><\/td>\n<\/tr>\n<tr>\n<td style=\"width: 209.262px;\"><span style=\"font-weight: 400;\">Content generation<\/span><\/td>\n<td style=\"width: 167.137px;\"><span style=\"font-weight: 400;\">54%<\/span><\/td>\n<\/tr>\n<tr>\n<td style=\"width: 209.262px;\"><span style=\"font-weight: 400;\">Data analysis<\/span><\/td>\n<td style=\"width: 167.137px;\"><span style=\"font-weight: 400;\">54%<\/span><\/td>\n<\/tr>\n<tr>\n<td style=\"width: 209.262px;\"><span style=\"font-weight: 400;\">Cybersecurity and threat detection<\/span><\/td>\n<td style=\"width: 167.137px;\"><span style=\"font-weight: 400;\">48%<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><span style=\"font-weight: 400;\">The barriers are not about cost. Around 65% of SMEs not using AI distrust its decisions or prefer to maintain human control, and 54% believe it is not relevant to their business. Another 19% simply do not know how to use it. The <\/span><a href=\"https:\/\/www.oecd.org\/en\/publications\/empowering-smes-in-the-age-of-ai_bf5a9816-en.html\"><span style=\"font-weight: 400;\">OECD&#8217;s D4SME 2026 survey<\/span><\/a><span style=\"font-weight: 400;\">, covering more than 2,000 SMEs across 12 countries, shows a similar pattern: most use off-the-shelf products, but integration into operations remains uneven.<\/span><\/p>\n<p><span style=\"font-weight: 400;\"><\/p>\n<p><\/span><\/p>\n<h3><b>What do companies seeing results do differently?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">They select a few processes and redesign them end-to-end. Instead of &#8220;giving AI to everyone&#8221;, they take a specific workflow, such as processing supplier invoices, and ask which steps disappear, which an agent automates, and where a person must decide.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">They also measure from day one. The <\/span><a href=\"https:\/\/www.stlouisfed.org\/on-the-economy\/2026\/jul\/ai-productivity-what-firms-say-earnings-calls\"><span style=\"font-weight: 400;\">Federal Reserve Bank of St. Louis<\/span><\/a><span style=\"font-weight: 400;\"> analysed around 490,000 earnings calls from US-listed companies and found that nearly 95% of statements regarding AI and productivity speak of future, not achieved, gains. Without a baseline (hours per task, response time, errors), it is impossible to know if the tool pays off.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">A typical example is an administration team testing an agent to reconcile supplier invoices. Before starting, they record how many hours a week they dedicate to that task and how many errors occur; during the trial, they measure the same metrics. At the end, they have a concrete number to decide whether to scale, adjust, or discard it.<\/span><\/p>\n<p><span style=\"font-weight: 400;\"><\/p>\n<p><\/span><\/p>\n<h3><b>How to start using AI for productivity in your company?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Start with a process, not a tool. This order avoids the most common mistake: paying for licences nobody knows how to use.<\/span><\/p>\n<ol>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Choose a repetitive, measurable process.<\/b><span style=\"font-weight: 400;\"> Something that currently consumes hours, follows clear rules, and has an easily reviewable outcome.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Measure the baseline.<\/b><span style=\"font-weight: 400;\"> How long it takes, how much it costs, and how many errors occur today. Without that number, there is no way to prove improvement.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Organise the data the AI will use.<\/b><span style=\"font-weight: 400;\"> Documents, spreadsheets, and systems must be accessible and up to date. Without organized data, AI has little to work with.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Redesign the workflow with a human in control.<\/b><span style=\"font-weight: 400;\"> Define what the AI does, what a person reviews, and who responds if something goes wrong.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Run a 4 to 8-week test and compare.<\/b><span style=\"font-weight: 400;\"> If the key metric improves, scale it; if not, adjust or discard without guilt.<\/span><\/li>\n<li style=\"font-weight: 400;\" aria-level=\"1\"><b>Train the team and set rules of use.<\/b><span style=\"font-weight: 400;\"> Which data must not be uploaded to external tools, what must always be verified, and how errors are reported.<\/span><\/li>\n<\/ol>\n<p><span style=\"font-weight: 400;\"><\/p>\n<p><\/span><\/p>\n<h3><b>Frequently Asked Questions<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">\u00a0<\/span><\/p>\n<p><b>Does AI replace employees?<\/b><span style=\"font-weight: 400;\"> For now, evidence does not show this. The Katz and Jung study did not find that generative AI is substituting some groups of workers for others; it is most common for tasks to shift within each role.<\/span><\/p>\n<p><b>How long does it take to see results?<\/b><span style=\"font-weight: 400;\"> A well-measured test shows signals in 4 to 8 weeks. Impact on business results arrives when the redesigned process scales, not before.<\/span><\/p>\n<p><b>Do I need a technical team?<\/b><span style=\"font-weight: 400;\"> For writing and summarisation tools, no. For agents connected to your systems, it is best to have someone who understands data and integration.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Discover how we at Asta can help you design or select your best AI option: <\/span><a href=\"https:\/\/www.asta.com.au\/ai-development\"><span style=\"font-weight: 400;\">https:\/\/www.asta.com.au\/ai-development<\/span><\/a><\/p>\n<p><span style=\"font-weight: 400;\"><\/p>\n<p><\/span><\/p>\n<h3><b>About Our mission in the digital space<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">\u00a0<\/span><\/p>\n<p><b>Asta is a leading full-service technology and consulting agency. We\u2019re trusted industry leaders, who are committed to advancing businesses through powerful IT.<\/b><span style=\"font-weight: 400;\"> Yet, beyond our IT acumen in software, web and mobile app development, our fit-for-purpose managed IT service solutions and our ground-breaking AI and blockchain technologies \u2013 there\u2019s something more.<\/span><\/p>\n<p><b>At the core of everything we do is our relentless commitment to people.<\/b><\/p>\n<p><span style=\"font-weight: 400;\"><\/p>\n<p><\/span><\/p>\n<h3><b>Contact and social networks<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">\u00a0<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Get in touch with us through our available social channels, and a specialised advisor will contact you to answer all your questions:<\/span><\/p>\n<p style=\"text-align: center;\"><a href=\"https:\/\/www.asta.com.au\/\"><span style=\"font-weight: 400;\">Website<\/span><\/a><span style=\"font-weight: 400;\"> &#8211; <\/span><a href=\"https:\/\/www.asta.com.au\/contact\"><span style=\"font-weight: 400;\">Contact<\/span><\/a><span style=\"font-weight: 400;\"> &#8211; <\/span><a href=\"https:\/\/www.linkedin.com\/company\/asta-solutions\/\"><span style=\"font-weight: 400;\">Linkedin<\/span><\/a><span style=\"font-weight: 400;\"> &#8211; <\/span><a href=\"https:\/\/twitter.com\/astasolutions?lang=en\"><span style=\"font-weight: 400;\">Twitter<\/span><\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>In 2026, AI is already improving individual productivity, but not yet that of most companies. According to McKinsey&#8217;s global survey published in August, 80% of respondents see individual productivity gains. However, only 37% can attribute any impact on operating income (EBIT) to AI, the same proportion as a year earlier. Where is AI being applied [&hellip;]<\/p>\n","protected":false},"author":4,"featured_media":827,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_et_pb_use_builder":"","_et_pb_old_content":"","_et_gb_content_width":"","om_disable_all_campaigns":false,"inline_featured_image":false,"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"fifu_image_url":"","fifu_image_alt":"","footnotes":""},"categories":[1],"tags":[],"class_list":["post-826","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized"],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/www.asta.com.au\/news\/wp-json\/wp\/v2\/posts\/826","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.asta.com.au\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.asta.com.au\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.asta.com.au\/news\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/www.asta.com.au\/news\/wp-json\/wp\/v2\/comments?post=826"}],"version-history":[{"count":1,"href":"https:\/\/www.asta.com.au\/news\/wp-json\/wp\/v2\/posts\/826\/revisions"}],"predecessor-version":[{"id":828,"href":"https:\/\/www.asta.com.au\/news\/wp-json\/wp\/v2\/posts\/826\/revisions\/828"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.asta.com.au\/news\/wp-json\/wp\/v2\/media\/827"}],"wp:attachment":[{"href":"https:\/\/www.asta.com.au\/news\/wp-json\/wp\/v2\/media?parent=826"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.asta.com.au\/news\/wp-json\/wp\/v2\/categories?post=826"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.asta.com.au\/news\/wp-json\/wp\/v2\/tags?post=826"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}