{"id":65636,"date":"2021-09-17T10:24:06","date_gmt":"2021-09-17T09:24:06","guid":{"rendered":"https:\/\/www.cashontime.com\/en\/?p=65636"},"modified":"2024-10-22T11:30:30","modified_gmt":"2024-10-22T10:30:30","slug":"artificial-intelligence-wcr-accounts-receivable","status":"publish","type":"post","link":"https:\/\/www.cashontime.com\/en\/news\/artificial-intelligence-wcr-accounts-receivable\/","title":{"rendered":"How can Artificial Intelligence optimize your WCR and receivables optimization processes?"},"content":{"rendered":"<p style=\"text-align: center;\"><strong>The topic was addressed during a 26-minute TV set at the last FORUMDIMO TV.<\/strong><\/p>\n<p style=\"text-align: center;\"><a target=\"_blank\" data-obflink-url=\"aHR0cHM6Ly9scC5jYXNob250aW1lLmNvbS9pbnRlbGxpZ2VuY2UtYXJ0aWZpY2llbGxlLWJmci1wb3N0ZS1jbGllbnRzLw==\" class=\"bouton vert\" tabindex=\"0\">SEE THE REPLAY<\/a><\/p>\n<p>Accounts receivable encompasses the processing of all invoices issued (accounting allocation, automatic lettering, collection). In this context, Artificial Intelligence (AI) is a set of processes that reproduce human reasoning and thought patterns. For CashOnTime, it is based on <strong>learning, via machine learning and deep learning<\/strong>. G\u00e9rard Anta, Manpower accounting and customer collections manager, talks about his use of CashOnTime Allocation since 2018 within the customer collection unit. A set hosted by Fanny Rondet, DIMO Software sales engineer, with Karine Gidemann, CashOnTime range manager, who manages processes linked to accounts receivable (CashOnTime Allocation for automatic <a href=\"\/en\/articles\/accounting-lettering\/\">lettering<\/a> of customer collections and CashOnTime Collection for <a href=\"\/en\/articles\/debt-collection\/\">debt collection<\/a>).<\/p>\n<h2>AI for optimizing accounts receivable<\/h2>\n<p>All daily settlements are subject to automatic lettering rules within a lettering strategy (invoice number, equal amount, invoice accumulation). This type of setting is entered in the platform.<br>\n<strong>80% of settlements are automatically lettered, with the remaining 20% handled by the AI engine.<\/strong> For each payment, there are two stages:<\/p>\n<ul>\n<li>Step 1: <strong>deep learning<\/strong> \u2013 learning in depth via a neural network \u2013 <strong>ingests patterns and deduces all combinations of invoice amounts corresponding to the settlement amount<\/strong> (a single amount, tens, thousands\u2026). The number of open invoices in a customer account, the height of the amount, the presence of credit notes, etc. are all factors that influence the figure.<\/li>\n<li>Step 2: <strong>machine learning<\/strong> will identify the most relevant combinations and propose them to the user. <strong>The machine will learn from a whole host of examples<\/strong>. It will make mistakes, but will integrate them into the process. Correlations are made between all paid invoices (measuring the proximity of invoice numbers, dates, etc.). The system will be able <strong>to score the various combinations to propose only the most relevant to the user<\/strong>, a bit like confidence indexes against all the memorized rules.<\/li>\n<\/ul>\n<p>The objective? <strong>Increase the rate of close liquidity by 99%!<\/strong><\/p>\n<h2>CashOnTime Allocation at Manpower<\/h2>\n<p>G\u00e9rard Anta explains: \u201cIn terms of accounts receivable, we separate customer accounting from debt collection. The accounts receivable department consisted of 7 people plus a group manager. Their main activity consisted in <strong>entering payments and performing lettering<\/strong>, i.e. reconciling payments with invoices. Centrally, a high-level collections team was responsible for <strong>resolving problems on complex accounts<\/strong> according to action plans drawn up in collaboration with the sales department. It was responsible for following up certain organizations with complex accounts and specific operating procedures, and ensuring that collection was carried out under the best possible conditions. Finally, offshore collection, outsourced, consists of <strong>dunning customers according to predefined workflows in line with company size, type and payment habits<\/strong>, with dunning done automatically or manually\u201d.<\/p>\n<p>Excel macros had been set up internally to perform part of the <strong>automatic lettering in the Information System<\/strong> (IS). Today, Manpower manages <strong>30,000 payments, 85% of which are transfers<\/strong>. The rate achieved was <strong>39% for semi-automatic entry and reconciliation<\/strong>. This prompted the organization to look for a <strong>more efficient solution<\/strong>. However, managing peaks was problematic, particularly at the beginning, middle and end of the month. G\u00e9rard Anta adds: \u201cOn average, we process 1,500 transfers, but they can reach 4,000 at the end of the month. We had to enter these transfers and reconcile the payments with a fairly small team. Data entry was done on the day, and required the use of related resources. Lettering took place over the following week. On a monthly average, we achieved an entry and<strong> lettering rate of 90% on D-day<\/strong>, with a fairly substantial human intervention: 7 people in accounts receivable, plus the collection team according to peaks\u201d.<\/p>\n<h2>Before and after CashOnTime<\/h2>\n<p>Before CashOnTime Allocation, 39% of transfers were entered and lettered semi-automatically via Excel macros. Now, CashOnTime directly <strong>integrates the transfer account statement at 6 a.m<\/strong>. on a daily basis. CashOnTime runs algorithms, fetches invoice numbers from the enriched account statement, account balances and due dates, and achieves <strong>83% automatic lettering and entry<\/strong>. CashOnTime doesn\u2019t stop there: in many cases, the system has identified the customer issuing the transfer, but may be hesitant about which lettering to carry out. <strong>The system will issue lettering proposals<\/strong>, and it is the accountant in Accounts Receivable who will validate one or other of the proposals.<\/p>\n<h2>Benefits for other services, especially collections<\/h2>\n<p>G\u00e9rard Anta explains: \u201cAfter validating the proposals, we arrive at <strong>around 86% of lettering entries<\/strong>, with the remainder charged to the accounts. <strong>Collection is much easier<\/strong>. Before, dunning certain customers could prove problematic (accounts not up to date, etc.). It\u2019s always tricky to <strong>deal with a customer who claims to have already paid<\/strong>. The result: a certain <strong>climate of tension and doubt among the teams<\/strong>. Now, we can be sure that when a customer has paid by bank transfer at around 10 a.m., everything is in the accounts and collection can proceed without a hitch! He adds: \u201c<strong>We can now be sure of proof of payment\u2026 or the lack of it<\/strong>. We can also be sure that the money in our account will be charged to our customers on the day.<\/p>\n<h2>AI in CashOnTime software<\/h2>\n<p>In addition to lettering, AI is used to project <a href=\"\/en\/articles\/dso\/\">DSO<\/a>, the average time to payment. Karine Gidemann explains: \u201cWe use a linear regression algorithm to <strong>anticipate DSO trends<\/strong>. This is the number of days between the day the invoice is issued and the moment it is cashed. Based on historical data, AI analyzes the customer\u2019s payment behavior\u201d. There are other indicators: <strong>the CEI<\/strong> \u2013 Collection Efficiency Index \u2013 measures the performance of pure collection actions. It is the ratio between the amount collected over a given period and the amount to be recovered.<\/p>\n<p>The system stores all the data obtained either via user actions, by deduction of payer profiles or risk categories. AI enables <strong>correlations to be made between all this data<\/strong>, so that analyses can be deduced and the most appropriate decisions made by collection agents. We also have a <strong>collection forecast report<\/strong>, based on historical data: from the moment the invoice is entered in the accounts and the payment date is known, we are able to project and <strong>anticipate future payments<\/strong> on a weekly or monthly basis, with a view by account, collection segmentation, etc. The aim: to <strong>feed cash management<\/strong> and <strong>adjust the appropriate collection strategy<\/strong>.<\/p>\n<h2>AI expands the scope of tomorrow\u2019s possibilities<\/h2>\n<p>DIMO Software\u2019s R&amp;D team \u2013 data scientists, mathematicians, cross-functional staff\u2026 \u2013 is working on <strong>improving the exploitation of payment notices<\/strong>, on the Allocation part, the payment notices that its customers receive from their own customers. \u201cAI will enable us, without parameterization, to integrate any PDF format, whether native or a scanned image.<strong> We\u2019ll be able to locate the amount advertised, whether a string of characters corresponds to an invoice number<\/strong> advertised by the customer. We\u2019ve been exploiting payment information for several years, but we were limited to native PDFs and parameterization: in fact, each customer has his own PDF format, with a certain structure,\u201d explains Karine Gidemann. The AI engine was entirely designed and developed at the Limonest headquarters.<\/p>\n<h2>About debt collection<\/h2>\n<p>Soon, the <strong>system will analyze exchanges<\/strong> (e-mail, chat, etc.)<strong> between the debt collector and his contact at the customer\u2019s premises<\/strong> to detect key words (dispute or promise of payment, etc.). In automatic mode, the artificial intelligence will create an action so that the debt collector can automatically integrate this data into a <strong>follow-up workflow and reduce the time taken to contest invoice amounts<\/strong>. Ultimately, the aim is to <strong>provide visibility on all the levers for optimizing<\/strong> working capital, and to control the time lag caused by incoming payments.<\/p>\n<h2>For staff in charge of automatic lettering<\/h2>\n<p>\u201c<strong>Saving 2 full-time team<\/strong> members was the business case I put together to encourage my management to invest in <strong>CashOnTime Allocation<\/strong>,\u201d explains G\u00e9rard Anta. \u201cAfter deploying CashOnTime Allocation, we were able to redirect one person who wanted to do <strong>reporting and management control<\/strong>. The team was then able to <strong>concentrate on its core business and higher value-added tasks<\/strong>, such as responding to customer complaints. What\u2019s more, we no longer need to call on cross-functional teams, and everyone now remains focused on their core business. The health crisis has enabled us to focus on what\u2019s overdue, what companies are failing and what decisions need to be made, with all the cards in our hands. <strong>Having up-to-the-minute accounts, settlements and accounting records<\/strong> made certain work situations much easier!<\/p>\n<h2>Expectations exceeded thanks to accounting lettering solution<\/h2>\n<p>\u201cWe were <strong>won over by the lettering rate announced<\/strong> by CashOnTime Allocation, which was <strong>actually achieved and even exceeded<\/strong>. The solution was quickly implemented (1? months, 2 months on site). In the temping business, we issue a lot of invoices every month (200,000 to 250,000). When we\u2019re expecting a multi-million settlement, with thousands of invoices to be lettered, <strong>reconciliation will be automatic<\/strong> rather than manual, which <strong>saves us a considerable amount of time<\/strong>. Fanny Rondet notes that <strong>implementation time depends on the time and number of people the customer can devote to the recipe<\/strong>.<\/p>\n<h2 class=\"Chapo\">To remember<\/h2>\n<p class=\"Chapo\">Artificial intelligence and digitalization are formidable <strong>levers for unlocking working capital and reducing risk<\/strong>. These new technologies will continue to challenge the repetitive manual methods of the <a href=\"\/en\/articles\/order-to-cash\/\">Order-to-Cash<\/a> chain.<\/p>\n<p>\u00a0<\/p>\n<p>In the same vein: <a data-obflink-url=\"L2VuL25ld3MvY2ZvLWNyZWRpdC1tYW5hZ2VtZW50Lw==\" class=\"\" tabindex=\"0\">What role does finance and administration play in credit management?<\/a><\/p>\n<p><picture class=\"wp-image-25313  aligncenter\"><source srcset=\"https:\/\/www.cashontime.com\/en\/wp-content\/uploads\/2021\/04\/logo-cot-bleu-300x40.png.webp\" type=\"image\/webp\"><\/source><\/picture><\/p>\n\n\n","protected":false},"excerpt":{"rendered":"<p>The topic was addressed during a 26-minute TV set at the last FORUMDIMO TV. SEE THE REPLAY Accounts receivable encompasses the processing of all invoices issued (accounting allocation, automatic lettering, collection). In this context, Artificial Intelligence (AI) is a set of processes that reproduce human reasoning and thought patterns. For CashOnTime, it is based on [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":65637,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_seopress_robots_primary_cat":"none","_seopress_titles_title":"How does AI optimize your WCR and O2C processes?","_seopress_titles_desc":"The function of AI is to automate repetitive customer-related tasks by reproducing the thought processes of a human being.","_seopress_robots_index":"","inline_featured_image":false,"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","theme-transparent-header-meta":"default","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"set","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[123],"tags":[],"class_list":{"0":"post-65636","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-news"},"acf":{"url_francaise":"https:\/\/www.cashontime.com\/actualites\/intelligence-artificielle-bfr-poste-clients\/","url_espagnole":"https:\/\/www.cashontime.com\/es\/noticias\/inteligencia-artificial-nof-cuentas-por-cobrar\/"},"_links":{"self":[{"href":"https:\/\/www.cashontime.com\/en\/wp-json\/wp\/v2\/posts\/65636","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.cashontime.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.cashontime.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.cashontime.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.cashontime.com\/en\/wp-json\/wp\/v2\/comments?post=65636"}],"version-history":[{"count":2,"href":"https:\/\/www.cashontime.com\/en\/wp-json\/wp\/v2\/posts\/65636\/revisions"}],"predecessor-version":[{"id":72995,"href":"https:\/\/www.cashontime.com\/en\/wp-json\/wp\/v2\/posts\/65636\/revisions\/72995"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.cashontime.com\/en\/wp-json\/wp\/v2\/media\/65637"}],"wp:attachment":[{"href":"https:\/\/www.cashontime.com\/en\/wp-json\/wp\/v2\/media?parent=65636"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.cashontime.com\/en\/wp-json\/wp\/v2\/categories?post=65636"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.cashontime.com\/en\/wp-json\/wp\/v2\/tags?post=65636"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}