[{"data":1,"prerenderedAt":646},["ShallowReactive",2],{"navigation-en":3,"uslugi-slug-en-process-automation":87,"language-switcher-\u002Fen\u002Fservices\u002Fprocess-automation":641},{"services":4,"industries":48,"caseStudies":69},[5,10,14,19,23,27,31,35,39,43],{"slug":6,"navLabel":7,"navGroup":8,"heroHeadline":9},"web-applications","Web applications","build","A web application built around your process",{"slug":11,"navLabel":12,"navGroup":8,"heroHeadline":13},"erp-systems","ERP systems","An ERP system built for your company, not the other way round",{"slug":15,"navLabel":16,"navGroup":17,"heroHeadline":18},"integrations-and-api","Integrations and APIs","improve","Your systems stop being islands",{"slug":20,"navLabel":21,"navGroup":17,"heroHeadline":22},"ai-implementation","AI implementation","AI deployed where it actually saves time",{"slug":24,"navLabel":25,"navGroup":17,"heroHeadline":26},"process-automation","Process automation","Work no one should be doing by hand",{"slug":28,"navLabel":29,"navGroup":8,"heroHeadline":30},"websites","Websites","A site that loads instantly and brings in inquiries",{"slug":32,"navLabel":33,"navGroup":8,"heroHeadline":34},"mobile-applications","Mobile applications","One codebase, both platforms",{"slug":36,"navLabel":37,"navGroup":17,"heroHeadline":38},"code-audit-and-project-takeover","Code audit and takeover","The project stalled. We'll take it over and tell you the truth about its condition",{"slug":40,"navLabel":41,"navGroup":17,"heroHeadline":42},"maintenance-and-support","Maintenance and support","A system needs someone watching it after launch",{"slug":44,"navLabel":45,"navGroup":46,"heroHeadline":47},"discovery-workshop","Discovery workshop","start","Before you decide to build, find out exactly what you're building",[49,53,57,61,65],{"slug":50,"navLabel":51,"heroHeadline":52},"manufacturing","Manufacturing","The works order reaches the floor as a printout from an email",{"slug":54,"navLabel":55,"heroHeadline":56},"automotive","Automotive","The customer exists in three places at once",{"slug":58,"navLabel":59,"heroHeadline":60},"industrial-automation","Industrial automation","You sell competence that doesn't photograph",{"slug":62,"navLabel":63,"heroHeadline":64},"e-commerce","E-commerce","The shop works. What sits behind it comes apart",{"slug":66,"navLabel":67,"heroHeadline":68},"logistics","Transport and logistics","The job arrives by email, the status comes back by phone",[70,77,82],{"slug":71,"client":72,"tag":73,"year":74,"summary":75,"image":76},"dss-by-caredo","Caredo","Product",2025,"We build the whole DSS product together with the Caredo team — from the CRM and planner to the software behind the key lockers and totems.","\u002Fimg\u002Fcase-studies\u002Fdss\u002Fkluczykomat\u002Fframe_0169.webp",{"slug":78,"client":72,"tag":79,"year":74,"summary":80,"image":81},"dss-by-caredo-website","Product website","We built the DSS product website end to end — from design to launch — on Caredo's visual identity.","\u002Fimg\u002Fscreenshots\u002Fdss-desktop.jpg",{"slug":83,"client":83,"tag":84,"year":74,"summary":85,"image":86},"2motion","Brand and website","A whole brand from scratch — logo, colours, business cards — a company website and blog posts that position 2motion in search.","\u002Fimg\u002Fscreenshots\u002F2motion-desktop.jpg",{"service":88,"relatedCase":192,"relatedPosts":268,"i18nSlugs":504,"subServices":506},{"id":89,"title":90,"audience":91,"body":95,"description":116,"extension":117,"faq":118,"heroHeadline":26,"heroSubline":131,"keyword":132,"meta":133,"navGroup":17,"navLabel":25,"navigation":134,"order":135,"path":136,"pricingModels":137,"process":144,"relatedCase":83,"relatedPosts":161,"scope":163,"seo":179,"sitemap":180,"slug":24,"stem":190,"__hash__":191},"servicesEn\u002Fservices\u002Fprocess-automation.md","Business process automation",[92,93,94],"Teams spending days on reports that get built the same way every month","Companies where approving a document means an email chain","Organisations hiring more people to do work a rule could do",{"type":96,"value":97,"toc":110},"minimark",[98,103,107],[99,100,102],"h2",{"id":101},"automation-starts-with-a-spreadsheet-of-hours","Automation starts with a spreadsheet of hours",[104,105,106],"p",{},"Before we write anything, we want to know how many times a month the process runs and how long it takes each time. That number tells us how much is reasonable to spend on automating it — and whether it's worth it at all.",[104,108,109],{},"Surprisingly often, the process that annoys everyone costs four hours a month — and the one nobody talks about, because \"it's always been done that way,\" eats two full-time roles.",{"title":111,"searchDepth":112,"depth":112,"links":113},"",3,[114],{"id":101,"depth":115,"text":102},2,"We remove repetitive manual work — document flows, reports, retyping data between systems. We start with the process that costs the most hours.","md",[119,122,125,128],{"q":120,"a":121},"Which process should we start with?","The one performed most often with the fewest exceptions — not the most annoying one. Annoyance and cost are two different things, and automating a process full of exceptions can end up more expensive than doing it by hand.",{"q":123,"a":124},"Does automation mean layoffs?","In practice, usually not. What disappears is work nobody wanted to do, and the team takes on things there was never time for before. If the goal is headcount reduction, we'll say honestly whether a given process can carry that.",{"q":126,"a":127},"How long until it pays for itself?","With a well-chosen process, usually 6–12 months. That's why we start by counting hours — without that, it's guesswork.",{"q":129,"a":130},"How is this different from off-the-shelf no-code tools?","It isn't, if your process fits what they can do — then we'll point you to a tool instead of writing code. The difference shows up with rules those tools can't express, and at volumes where they start getting expensive.","Every company has a few processes that eat up a full-time role and amount to moving information from one place to another.","business process automation",{},true,5,"\u002Fservices\u002Fprocess-automation",[138,141],{"name":139,"body":140},"Fixed price per process","One process, known scope, known price. The most common and safest model.",{"name":142,"body":143},"Automation retainer","A fixed monthly pool of hours for further automations, for companies that want to go process by process.",[145,149,153,157],{"step":146,"duration":147,"body":148},"Audit and process selection","1 week","We review the candidates and pick the one with the best savings-to-difficulty ratio.",{"step":150,"duration":151,"body":152},"Rule design","1–2 weeks","We write down the edge cases and exceptions. They decide success, not the happy path.",{"step":154,"duration":155,"body":156},"Rollout","2–6 weeks","The automation runs alongside the manual process, so results can be compared.",{"step":158,"duration":159,"body":160},"Switchover and expansion","2–4 weeks","We turn off the manual version and move on to the next process.",[162],"wdrozenie-ai-w-firmie-od-czego-zaczac",[164,167,170,173,176],{"title":165,"body":166},"Process map and hour count","We start by counting how much time the process costs today. Without that number, there's no way to judge whether automating it pays off.",{"title":168,"body":169},"Document and approval flows","Requests, orders and invoices with approval rules, reminders and a decision history.",{"title":171,"body":172},"Automatically generated reports","The summaries a person compiles today, built by themselves and delivered wherever they need to land.",{"title":174,"body":175},"Data synchronisation","No more retyping the same record into three systems.",{"title":177,"body":178},"Notifications and escalations","The system watches deadlines instead of a person who has to remember them.",{"title":90,"description":116},{"loc":181,"alternatives":182},"\u002Fen\u002Fservices\u002Fprocess-automation",[183,186,188],{"hreflang":184,"href":185},"x-default","\u002Fuslugi\u002Fautomatyzacja-procesow",{"hreflang":187,"href":185},"pl-PL",{"hreflang":189,"href":181},"en-US","services\u002Fprocess-automation","dWXLPn_QIMw3rILavszqeESlo1LijXwS70wlg6LmA0M",{"id":193,"title":194,"body":195,"challenge":229,"client":83,"description":230,"extension":117,"features":231,"meta":237,"navigation":134,"order":112,"path":238,"relatedServices":239,"results":240,"screenshotDesktop":86,"screenshotMobile":249,"seo":250,"sequence":251,"sitemap":252,"slug":83,"solution":259,"stack":260,"stem":264,"summary":85,"tag":84,"url":265,"usedBy":266,"year":74,"__hash__":267},"caseStudiesEn\u002Fcase-studies\u002F2motion.md","2motion — the whole brand, website and blog for industrial automation",{"type":96,"value":196,"toc":224},[197,201,204,207,211,214,217,221],[99,198,200],{"id":199},"starting-point","Starting point",[104,202,203],{},"An engineering company has a harder job than a shop: it sells expertise, not a product you can photograph. The site has to convince a technical visitor within seconds that there are people on the other end who know what they're doing.",[104,205,206],{},"On top of that: the client often opens the site on a phone, on the shop floor, on a poor connection.",[99,208,210],{"id":209},"what-we-did","What we did",[104,212,213],{},"The brand first: the logo, the colour palette and business cards — 2motion started without an identity, so the website had nothing to stand on. Then a statically generated site where the HTML is ready before anyone arrives. Every subpage is prerendered, images are served in modern formats, and JavaScript only ships where it's actually needed.",[104,215,216],{},"The content architecture was decided before the visual design: we first established which subpage answers which query, so they wouldn't compete against each other in search results. Each of the six services has its own page, and the blog posts we co-write with 2motion answer the questions their clients arrive with from search — from palletising to integrating robots into an existing line.",[99,218,220],{"id":219},"what-we-learned","What we learned",[104,222,223],{},"On a brand site, the decision with the biggest impact on the outcome isn't visible in the design — it's how much content actually gets written. A layout for eight paragraphs and a layout for two are two different projects, and that number is only known once someone sits down to write.",{"title":111,"searchDepth":112,"depth":112,"links":225},[226,227,228],{"id":199,"depth":115,"text":200},{"id":209,"depth":115,"text":210},{"id":219,"depth":115,"text":220},"A company in industrial automation sells expertise that can't be shown in a product photo. The site needed to build technical credibility, work flawlessly on a phone on a client's shop floor, and rank for industry-specific search terms.","We created 2motion's whole brand — logo, colours, business cards — built its website and co-write the blog posts that build its search visibility.",[232,233,234,235,236],"A visual identity from scratch — logo, colours, business cards","A company website with a page for each of six services","A blog with co-written posts aimed at industry search terms","Technical SEO — structured data, sitemap, canonical URLs","Static generation and image optimisation",{},"\u002Fcase-studies\u002F2motion",[28],[241,244,246],{"metric":242,"label":243},"6","Service pages, each for its own search term",{"metric":242,"label":245},"Blog posts co-written for SEO",{"metric":247,"label":248},"\u003C1.2 s","Time to first render","\u002Fimg\u002Fscreenshots\u002F2motion-mobile.jpg",{"title":194,"description":230},null,{"loc":253,"alternatives":254},"\u002Fen\u002Fcase-studies\u002F2motion",[255,257,258],{"hreflang":184,"href":256},"\u002Frealizacje\u002F2motion",{"hreflang":187,"href":256},{"hreflang":189,"href":253},"We created the brand from the ground up: the logo, the colour palette and business cards, and on top of them a statically generated website with a page for every service. Together with 2motion we co-write blog posts that answer their clients' questions and build visibility for industry terms.",[261,262,263],"TypeScript","Nuxt","Vercel","case-studies\u002F2motion","https:\u002F\u002F2motion.pl",[],"P9NX81zWImOQJrUEvem3RidKuJSGHm2UlrA5z02USqc",[269],{"id":270,"title":271,"author":272,"body":273,"category":488,"date":489,"description":490,"draft":491,"editorialNote":251,"extension":117,"image":492,"meta":493,"navigation":134,"path":494,"rawbody":495,"seo":496,"sitemap":497,"stem":498,"tags":499,"translationKey":502,"updated":251,"__hash__":503},"blogEn\u002Fblog\u002Fwdrozenie-ai-w-firmie-od-czego-zaczac.md","AI adoption in your company — where to start","zespol",{"type":96,"value":274,"toc":478},[275,291,295,298,305,309,312,318,324,330,333,337,344,347,351,354,360,366,372,378,382,385,407,410,414,440,444,450,456,462,468,472,475],[104,276,277,278,282,283,286,287,290],{},"Start with a process that meets three conditions at once: it's ",[279,280,281],"strong",{},"repetitive",", has a ",[279,284,285],{},"countable time cost",", and ",[279,288,289],{},"tolerates a mistake",". Classifying incoming documents usually meets all three. \"An assistant that knows the whole company\" meets none of them — which is why so many rollouts that start there never finish.",[99,292,294],{"id":293},"why-pilots-dont-reach-production","Why pilots don't reach production",[104,296,297],{},"A demo shows a model handling a dozen or so examples chosen by the person who prepared them. Production is tens of thousands of examples nobody chose: scans crooked on the glass, documents in a format outside the test set, invoices with a handwritten note in the margin.",[104,299,300,301,304],{},"The difference between the two isn't model quality. It's ",[279,302,303],{},"handling the cases where the model isn't sure"," — and that's exactly what a rollout is actually made of. A pilot that skips it isn't an earlier version of the rollout. It's a different thing entirely.",[99,306,308],{"id":307},"how-to-choose-your-first-use-case","How to choose your first use case",[104,310,311],{},"Make a list of candidates and score each on three dimensions.",[104,313,314,317],{},[279,315,316],{},"Repeatability."," How many times a month does someone do this? Below a hundred, the savings rarely justify a rollout.",[104,319,320,323],{},[279,321,322],{},"Time cost."," How many minutes does one instance take? Multiply by repeatability. That number sets the upper bound on a sensible budget.",[104,325,326,329],{},[279,327,328],{},"Tolerance for error."," What happens if the model gets it wrong? If the answer is \"someone catches it at review\" — good. If it's \"money leaves an account\" — you need a human approval step, which changes the maths.",[104,331,332],{},"The best candidates in a typical company: classifying and routing incoming documents, extracting data from invoices and orders, drafting first responses to repetitive support questions, searching internal documentation.",[99,334,336],{"id":335},"set-a-success-metric-before-you-start","Set a success metric before you start",[104,338,339,340],{},"Before a prototype exists, write down one sentence: ",[341,342,343],"em",{},"\"This rollout makes sense if the model handles at least X% of cases on its own, at accuracy of at least Y%.\"",[104,345,346],{},"Without that number, there's no way to later decide whether the project succeeded. With it, the prototype ends in a clear decision: proceed, or stop. Stopping is also a good outcome — just a cheaper one than finding out a year in.",[99,348,350],{"id":349},"an-architecture-that-survives-production","An architecture that survives production",[104,352,353],{},"A rollout that lasts has four elements a demo doesn't.",[104,355,356,359],{},[279,357,358],{},"A confidence threshold."," The model returns a result along with its own confidence score. Above the threshold, automation runs; below it, the case goes to a human. Setting that threshold is a business decision, not a technical one.",[104,361,362,365],{},[279,363,364],{},"A queue for uncertain cases."," The place where a human resolves cases below the threshold — and its size is the real measure of how well the system is working.",[104,367,368,371],{},[279,369,370],{},"A decision log."," A record of what the model received, what it returned, and with what confidence. Without it you can't diagnose degradation or answer an auditor's question.",[104,373,374,377],{},[279,375,376],{},"Continuous measurement."," Data changes. A model that was 94% accurate in March can be at 80% by November — not because it broke, but because the input stream changed.",[99,379,381],{"id":380},"where-your-data-goes","Where your data goes",[104,383,384],{},"The question that comes up in every first meeting, rightly so. You have three options:",[386,387,388,395,401],"ul",{},[389,390,391,394],"li",{},[279,392,393],{},"A hosted model, no-training agreement."," Data leaves the company, but the provider commits not to use it for training. Cheapest and fastest, sufficient for most use cases.",[389,396,397,400],{},[279,398,399],{},"A hosted model in your own private cloud."," Data never leaves your infrastructure. More expensive, justified for sensitive data.",[389,402,403,406],{},[279,404,405],{},"A model on your own hardware."," Full control, the highest entry and running cost. Makes sense under regulatory requirements that rule out the other options.",[104,408,409],{},"The choice is a decision about risk, not technology. You make it once, and it affects cost for the entire lifetime of the solution.",[99,411,413],{"id":412},"what-it-costs","What it costs",[386,415,416,422,428,434],{},[389,417,418,421],{},[279,419,420],{},"Prototype on your data:"," PLN 15–30k net, 2–3 weeks. Ends with a number you can base a decision on.",[389,423,424,427],{},[279,425,426],{},"Production rollout:"," PLN 60–200k net, depending on the number of integrations and approval requirements.",[389,429,430,433],{},[279,431,432],{},"Running the models:"," from a few hundred to a few thousand zloty a month, roughly proportional to volume.",[389,435,436,439],{},[279,437,438],{},"Maintenance and tuning:"," usually 15% of the rollout's value per year.",[99,441,443],{"id":442},"four-mistakes-that-come-up-most-often","Four mistakes that come up most often",[104,445,446,449],{},[279,447,448],{},"Starting with the most impressive use case."," A conversational assistant impresses the board and is the hardest possible starting point — unbounded scope, no good success metric, high expectations.",[104,451,452,455],{},[279,453,454],{},"No human in the loop."," A system with no path for uncertain cases either performs badly or needs such a high confidence threshold that it only automates a small share of cases.",[104,457,458,461],{},[279,459,460],{},"Measuring accuracy alone."," 95% accuracy sounds great until you work out that at a thousand documents a day, that's fifty errors someone has to catch.",[104,463,464,467],{},[279,465,466],{},"Treating the rollout as a closed project."," A model needs watching, like any other piece of production infrastructure. A rollout with no measurement plan ages quietly.",[99,469,471],{"id":470},"the-first-step-you-can-take-this-week","The first step you can take this week",[104,473,474],{},"Pick one process. Count how many times a month it happens and how many minutes it takes. Write down what happens when someone gets it wrong.",[104,476,477],{},"Those three numbers are enough to judge whether a prototype is worth building — and they're worth more than a month of conversations about AI's possibilities.",{"title":111,"searchDepth":112,"depth":112,"links":479},[480,481,482,483,484,485,486,487],{"id":293,"depth":115,"text":294},{"id":307,"depth":115,"text":308},{"id":335,"depth":115,"text":336},{"id":349,"depth":115,"text":350},{"id":380,"depth":115,"text":381},{"id":412,"depth":115,"text":413},{"id":442,"depth":115,"text":443},{"id":470,"depth":115,"text":471},"ai","2026-09-02","How to choose your first use case, what it costs, and why most corporate AI projects end at the demo instead of reaching production.",false,"\u002Fimg\u002Fblog\u002Fwdrozenie-ai-w-firmie.jpg",{},"\u002Fblog\u002Fwdrozenie-ai-w-firmie-od-czego-zaczac","---\ntitle: AI adoption in your company — where to start\ndescription: How to choose your first use case, what it costs, and why most corporate AI projects end at the demo instead of reaching production.\ndate: 2026-09-02\ncategory: ai\ntags:\n  - ai\n  - automation\n  - rollouts\nauthor: zespol\nimage: \u002Fimg\u002Fblog\u002Fwdrozenie-ai-w-firmie.jpg\ntranslationKey: ai-adoption-where-to-start\ndraft: false\n---\n\nStart with a process that meets three conditions at once: it's **repetitive**, has a **countable time cost**, and **tolerates a mistake**. Classifying incoming documents usually meets all three. \"An assistant that knows the whole company\" meets none of them — which is why so many rollouts that start there never finish.\n\n## Why pilots don't reach production\n\nA demo shows a model handling a dozen or so examples chosen by the person who prepared them. Production is tens of thousands of examples nobody chose: scans crooked on the glass, documents in a format outside the test set, invoices with a handwritten note in the margin.\n\nThe difference between the two isn't model quality. It's **handling the cases where the model isn't sure** — and that's exactly what a rollout is actually made of. A pilot that skips it isn't an earlier version of the rollout. It's a different thing entirely.\n\n## How to choose your first use case\n\nMake a list of candidates and score each on three dimensions.\n\n**Repeatability.** How many times a month does someone do this? Below a hundred, the savings rarely justify a rollout.\n\n**Time cost.** How many minutes does one instance take? Multiply by repeatability. That number sets the upper bound on a sensible budget.\n\n**Tolerance for error.** What happens if the model gets it wrong? If the answer is \"someone catches it at review\" — good. If it's \"money leaves an account\" — you need a human approval step, which changes the maths.\n\nThe best candidates in a typical company: classifying and routing incoming documents, extracting data from invoices and orders, drafting first responses to repetitive support questions, searching internal documentation.\n\n## Set a success metric before you start\n\nBefore a prototype exists, write down one sentence: *\"This rollout makes sense if the model handles at least X% of cases on its own, at accuracy of at least Y%.\"*\n\nWithout that number, there's no way to later decide whether the project succeeded. With it, the prototype ends in a clear decision: proceed, or stop. Stopping is also a good outcome — just a cheaper one than finding out a year in.\n\n## An architecture that survives production\n\nA rollout that lasts has four elements a demo doesn't.\n\n**A confidence threshold.** The model returns a result along with its own confidence score. Above the threshold, automation runs; below it, the case goes to a human. Setting that threshold is a business decision, not a technical one.\n\n**A queue for uncertain cases.** The place where a human resolves cases below the threshold — and its size is the real measure of how well the system is working.\n\n**A decision log.** A record of what the model received, what it returned, and with what confidence. Without it you can't diagnose degradation or answer an auditor's question.\n\n**Continuous measurement.** Data changes. A model that was 94% accurate in March can be at 80% by November — not because it broke, but because the input stream changed.\n\n## Where your data goes\n\nThe question that comes up in every first meeting, rightly so. You have three options:\n\n- **A hosted model, no-training agreement.** Data leaves the company, but the provider commits not to use it for training. Cheapest and fastest, sufficient for most use cases.\n- **A hosted model in your own private cloud.** Data never leaves your infrastructure. More expensive, justified for sensitive data.\n- **A model on your own hardware.** Full control, the highest entry and running cost. Makes sense under regulatory requirements that rule out the other options.\n\nThe choice is a decision about risk, not technology. You make it once, and it affects cost for the entire lifetime of the solution.\n\n## What it costs\n\n- **Prototype on your data:** PLN 15–30k net, 2–3 weeks. Ends with a number you can base a decision on.\n- **Production rollout:** PLN 60–200k net, depending on the number of integrations and approval requirements.\n- **Running the models:** from a few hundred to a few thousand zloty a month, roughly proportional to volume.\n- **Maintenance and tuning:** usually 15% of the rollout's value per year.\n\n## Four mistakes that come up most often\n\n**Starting with the most impressive use case.** A conversational assistant impresses the board and is the hardest possible starting point — unbounded scope, no good success metric, high expectations.\n\n**No human in the loop.** A system with no path for uncertain cases either performs badly or needs such a high confidence threshold that it only automates a small share of cases.\n\n**Measuring accuracy alone.** 95% accuracy sounds great until you work out that at a thousand documents a day, that's fifty errors someone has to catch.\n\n**Treating the rollout as a closed project.** A model needs watching, like any other piece of production infrastructure. A rollout with no measurement plan ages quietly.\n\n## The first step you can take this week\n\nPick one process. Count how many times a month it happens and how many minutes it takes. Write down what happens when someone gets it wrong.\n\nThose three numbers are enough to judge whether a prototype is worth building — and they're worth more than a month of conversations about AI's possibilities.\n",{"title":271,"description":490},{"loc":494},"blog\u002Fwdrozenie-ai-w-firmie-od-czego-zaczac",[488,500,501],"automation","rollouts","ai-adoption-where-to-start","3h3Uyc_r6v-R1oggm3sXVR3s8arJjyy0xQy78uryQSg",{"pl":505,"en":24},"automatyzacja-procesow",[507,575],{"id":508,"title":509,"audience":510,"body":514,"description":534,"extension":117,"faq":535,"heroHeadline":545,"heroSubline":546,"keyword":547,"meta":548,"navigation":134,"order":549,"parent":24,"path":550,"relatedCase":251,"scope":551,"seo":564,"sitemap":565,"slug":572,"stem":573,"__hash__":574},"subServicesEn\u002Fservices\u002Fprocess-automation\u002Fdocument-workflow.md","Electronic document workflow",[511,512,513],"Companies approving cost invoices by forwarding emails","Organisations where nobody knows whose desk a document is on","Finance teams hunting for proof of an approval from six months ago",{"type":96,"value":515,"toc":530},[516,520,523,527],[99,517,519],{"id":518},"where-the-resistance-usually-is","Where the resistance usually is",[104,521,522],{},"With the people who approve quickly today, because they do it with one tap in\nan email. If the new workflow needs a login and three screens, approvals slow\ndown and the system earns a reputation it won't recover from. That one detail\ndecides these rollouts more often than the rest of the scope combined.",[99,524,526],{"id":525},"where-we-start","Where we start",[104,528,529],{},"By writing down the rules that today exist only as custom: who may approve up\nto what amount, and what happens when they are away. That is often the hardest\npart of the project and it usually isn't technical.",{"title":111,"searchDepth":112,"depth":112,"links":531},[532,533],{"id":518,"depth":115,"text":519},{"id":525,"depth":115,"text":526},"Invoice and contract approvals in a system instead of an email chain. With a decision trail, deputies and deadlines that don't vanish into an absent person's inbox.",[536,539,542],{"q":537,"a":538},"Is this the same as a DMS?","Not quite. A DMS stores documents; a workflow handles what happens to them — who has to decide, and what happens when they don't. If you already have a DMS, we usually build the workflow layer beside it.",{"q":540,"a":541},"Can it be rolled out for cost invoices only?","Yes, and that is usually where we start. It is the most countable process in a company: the number of documents and the approval time are known, so the effect can be measured without argument.",{"q":543,"a":544},"What about electronic signatures?","For internal approval an authenticated trail in the system is usually enough. A qualified signature comes in where the law or the other party requires it — and it is then a separate integration, quoted separately.","Approval stops being a chain of emails","An email-based workflow doesn't have one flaw but three: you cannot see where a document is stuck, there is no trace of the decision, and the approver's holiday stops everything.","electronic document workflow",{},12,"\u002Fservices\u002Fprocess-automation\u002Fdocument-workflow",[552,555,558,561],{"title":553,"body":554},"Approval paths with rules","Who approves, up to what amount and in what order — written into the system instead of into custom. The rule also covers the cases everyone knows about and nobody has written down.",{"title":556,"body":557},"Deputies and deadlines","An absence doesn't stop the flow, and a document sitting too long chases itself. This is usually the first thing people notice after launch.",{"title":559,"body":560},"A full decision trail","Who approved what and when, together with the version of the document at that moment. Reconstructing a case from a year ago stops being archaeology in an inbox.",{"title":562,"body":563},"Wired into the accounting system","An approved document moves on automatically. Without that the workflow is only a prettier inbox and somebody is still retyping.",{"title":509,"description":534},{"loc":566,"alternatives":567},"\u002Fen\u002Fservices\u002Fprocess-automation\u002Fdocument-workflow",[568,570,571],{"hreflang":184,"href":569},"\u002Fuslugi\u002Fautomatyzacja-procesow\u002Fobieg-dokumentow",{"hreflang":187,"href":569},{"hreflang":189,"href":566},"document-workflow","services\u002Fprocess-automation\u002Fdocument-workflow","sIQKzoyRMJ8_l3TL6XzpxTHHsv9qhPT1s917h1W1d2Y",{"id":576,"title":577,"audience":578,"body":582,"description":600,"extension":117,"faq":601,"heroHeadline":611,"heroSubline":612,"keyword":613,"meta":614,"navigation":134,"order":615,"parent":24,"path":616,"relatedCase":251,"scope":617,"seo":630,"sitemap":631,"slug":638,"stem":639,"__hash__":640},"subServicesEn\u002Fservices\u002Fprocess-automation\u002Freporting-automation.md","Reporting automation",[579,580,581],"Teams spending days preparing reports that recur on a cycle","Companies where the same metric has two different values in two departments","Boards making decisions on month-old data",{"type":96,"value":583,"toc":596},[584,588,591,593],[99,585,587],{"id":586},"the-biggest-surprise-in-these-projects","The biggest surprise in these projects",[104,589,590],{},"Not technical but definitional. On the first report it almost always emerges\nthat two departments calculate the same metric differently and both definitions\nhave a rationale. Settling that is part of the project, even though it doesn't\nlook like programming work.",[99,592,526],{"id":525},[104,594,595],{},"With the one report that takes the most time. Not with the full set — the first\none mostly serves to check whether the source data is what everyone assumes.",{"title":111,"searchDepth":112,"depth":112,"links":597},[598,599],{"id":586,"depth":115,"text":587},{"id":525,"depth":115,"text":526},"Reports that produce themselves instead of being assembled monthly from three exports. With figures calculated live and one definition per metric.",[602,605,608],{"q":603,"a":604},"Do we need a data warehouse?","Usually not to start. With a handful of sources and a sensible volume, an intermediate layer is enough. A warehouse answers scale and history, not the mere fact that a report is assembled by hand.",{"q":606,"a":607},"Will you build it in Power BI or similar?","If you have it and it meets the need — yes, and our work is then preparing the data rather than building another tool. We build a custom view where the report has to live inside an application the team already uses.",{"q":609,"a":610},"How long does this take?","The first report usually within a few weeks. The time depends almost entirely on the state of the source data, not on the number of charts.","A report that doesn't take three days","A monthly summary built by exporting, copying and stitching costs several days of work and is out of date on the day it lands. It repeats every month, identically.","reporting automation",{},13,"\u002Fservices\u002Fprocess-automation\u002Freporting-automation",[618,621,624,627],{"title":619,"body":620},"One definition per metric","Margin, turnover, on-time delivery — calculated once, in one place. The biggest value of these projects is usually not the time saved but the end of arguments about which number is real.",{"title":622,"body":623},"Pulling from the sources","ERP, CRM, spreadsheets, external systems — read automatically on a set rhythm. Manual export survives only where a system genuinely offers no other route.",{"title":625,"body":626},"Live reports and views","Summaries calculated when the view is opened, not overnight on the fifth. Available to the people who need them without asking for an export.",{"title":628,"body":629},"Scheduled delivery where it earns its place","Not every report is one somebody will open. Some have to arrive by email at a set time, and that is part of the automation too.",{"title":577,"description":600},{"loc":632,"alternatives":633},"\u002Fen\u002Fservices\u002Fprocess-automation\u002Freporting-automation",[634,636,637],{"hreflang":184,"href":635},"\u002Fuslugi\u002Fautomatyzacja-procesow\u002Fautomatyzacja-raportowania",{"hreflang":187,"href":635},{"hreflang":189,"href":632},"reporting-automation","services\u002Fprocess-automation\u002Freporting-automation","jbH32T6z7-nGNcX7g8F4LOxw2sE1MIULA8AqBazxhPQ",[642],{"code":643,"label":644,"name":645,"to":185},"pl","PL","Polish",1790522544623]