<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Lekshmi Rajagopal on Medium]]></title>
        <description><![CDATA[Stories by Lekshmi Rajagopal on Medium]]></description>
        <link>https://medium.com/@lekshmi.rajagopal?source=rss-318cf590e51e------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*dmbNkD5D-u45r44go_cf0g.png</url>
            <title>Stories by Lekshmi Rajagopal on Medium</title>
            <link>https://medium.com/@lekshmi.rajagopal?source=rss-318cf590e51e------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 20:37:02 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@lekshmi.rajagopal/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Responsive Sourcing: The Heart of Order Orchestration, Made Flexible]]></title>
            <link>https://medium.com/techunfiltered-ayata/responsive-sourcing-the-heart-of-order-orchestration-made-flexible-2085a44df337?source=rss-318cf590e51e------2</link>
            <guid isPermaLink="false">https://medium.com/p/2085a44df337</guid>
            <category><![CDATA[sourcing]]></category>
            <category><![CDATA[oms]]></category>
            <category><![CDATA[fluentcommerce]]></category>
            <dc:creator><![CDATA[Lekshmi Rajagopal]]></dc:creator>
            <pubDate>Sun, 27 Sep 2026 14:06:41 GMT</pubDate>
            <atom:updated>2026-10-02T08:12:46.531Z</atom:updated>
            <content:encoded><![CDATA[<h3>Why sourcing is the heart of orchestration</h3><p>Fluent Commerce’s Responsive Sourcing Framework moves much of fulfilment sourcing out of rule code and into versioned configuration. It changes where sourcing logic lives, who can change it, and what engineers still own. Business teams can now change how orders are routed without a development cycle.</p><p>Every order passes through one decision: which location, or set of locations, should fulfil it. That decision shapes delivery speed, shipping cost, split-shipment rates, store workload and whether stock sells before it goes stale. Get it wrong and every later step, from pick and pack to carrier and customer notification, inherits the problem.</p><p>Sourcing logic also changes more often than almost anything else in an order management system. Peak season, new stores, carrier changes, promotions and regional service levels all call for a change. The platform that makes this decision easy to change gives the business the most room to move.</p><h3>The old way: custom sourcing rules</h3><p>Before the framework, each implementation built its own sourcing logic as custom Rubix rules in Java. That approach worked, but every change to routing was a software change.</p><p>A typical custom sourcing change meant:</p><ol><li>Writing or editing Java rule classes to query locations, check stock, calculate distance and pick a winner.</li><li>Wiring those rules into the order workflow and passing props between rulesets.</li><li>Writing unit tests with mocked GraphQL responses for stock, locations and networks.</li><li>Building, versioning and deploying a new module, then updating the workflow to reference it.</li><li>Regression testing every order type, because the logic sat in one place and a change for one channel could break another.</li></ol><p>The costs added up over time:</p><ul><li><strong>Slow change.</strong> A request like “prefer stores within 25 km during peak” became a sprint item, not a same-day tweak.</li><li><strong>Logic hidden in code.</strong> Business users could not see why an order went to a location without asking a developer.</li><li><strong>Duplicated effort.</strong> Distance, stock checks and split logic were rebuilt in slightly different ways across projects.</li><li><strong>Upgrade risk.</strong> Custom code drifted from platform updates and had to be reviewed at every release.</li><li><strong>Hard to explain.</strong> Rules that grew over years of edge cases became difficult to test and trust.</li></ul><h3>What the Responsive Sourcing Framework is</h3><p>Responsive Sourcing moves the sourcing decision out of rule code and into a <strong>Sourcing Profile</strong>: versioned configuration the platform evaluates at runtime. The workflow calls one out-of-the-box rule, CreateFulfilmentWithSourcingProfile, and names the profile to use.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5M4kS4qaetwIToaPF2nVnw.png" /></figure><p>Everything else lives in the profile.The engine tries strategies in priority order, filters out locations that break hard rules, scores the rest, and builds a fulfilment plan. If nothing can fulfil the order, a rejected fulfilment is created instead.</p><p>Profiles have a lifecycle of DRAFT, ACTIVE and INACTIVE. Only one version of a profile is active at a time, so a new routing approach can be prepared as a draft and switched on when ready.</p><h3>The building blocks</h3><p>Four concepts cover almost every sourcing scenario. Together they replace what used to be hundreds of lines of rule code.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ygYf-NjtsMdqyKh6M4wyjw.png" /></figure><p><strong>Strategies</strong> are tried in priority order. A PRIMARY strategy must fulfil the whole order; if none can, a FALLBACK strategy may fulfil part of it. Each strategy has its own conditions and criteria, and can override the profile&#39;s network and maxSplit.</p><p><strong>Conditions</strong> read any value from the order, fulfilment choice, items or locations using a path, and compare it with operators such as equals, in, greater_than, between and exists. Custom order attributes work the same way, so a new business flag needs no code.</p><p><strong>Criteria</strong> come ready to use:</p><ul><li>Exclusions: distance limit, location type, network, and stock availability.</li><li>Sorters: distance, distance bands, daily capacity, network priority, stock level, stock bands and order value.</li></ul><p><strong>Scoring</strong> and ranking are handled by the framework. Individual criteria can normalise their ratings to a 0–1 range, but criteria are applied in sequence rather than added together. The first ranking criterion sets the primary order, and later criteria break ties. For distance-based criteria, the normalisation is inverted so shorter distances get higher ratings.</p><p><strong>Split shipments</strong> are built in. With maxSplit above 1, the engine first tries one location, then two, and so on. Fewer splits always win, so customers get fewer parcels by default.</p><h3>Custom rules vs Responsive Sourcing</h3><p>The biggest shift is who can make a change, and how fast.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ACJs6ITo3cC5eiG1vVN66w.png" /></figure><h3>Everyday changes, now configuration only</h3><p>Each of these used to be a development request. With Responsive Sourcing, each is a profile edit.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/962/1*79hh6uUFIldRIbGn9bJWBQ.png" /></figure><p>For example, the express rule is a few lines of JSON rather than a new Java class:</p><pre>{<br>  &quot;ref&quot;: &quot;HD_EXPRESS&quot;,<br>  &quot;priority&quot;: 10,<br>  &quot;type&quot;: &quot;PRIMARY&quot;,<br>  &quot;sourcingConditions&quot;: [<br>    { &quot;name&quot;: &quot;IsExpress&quot;, &quot;type&quot;: &quot;fc.sourcing.condition.path&quot;,<br>      &quot;params&quot;: { &quot;path&quot;: &quot;fulfilmentChoice.type&quot;, &quot;operator&quot;: &quot;equals&quot;, &quot;value&quot;: &quot;HD_EXPRESS&quot; } }<br>  ],<br>  &quot;sourcingCriteria&quot;: [<br>    { &quot;name&quot;: &quot;NearbyOnly&quot;, &quot;type&quot;: &quot;fc.sourcing.criterion.locationDistanceExclusion&quot;,<br>      &quot;params&quot;: { &quot;value&quot;: 25.0, &quot;valueUnit&quot;: &quot;KILOMETRES&quot; } },<br>    { &quot;name&quot;: &quot;ClosestFirst&quot;, &quot;type&quot;: &quot;fc.sourcing.criterion.locationDistance&quot; }<br>  ]<br>}</pre><h3>Business impact, and where code still fits</h3><p>The framework shortens the path from a routing idea to live orders, and makes the result easy to see and explain.</p><ul><li><strong>Faster change.</strong> Routing tweaks move from a sprint cycle to a profile edit and activation.</li><li><strong>Lower cost to run.</strong> Less custom code to build, test, upgrade and support.</li><li><strong>Safer rollout.</strong> Draft a new version, test it, activate it, and fall back to the previous version if needed.</li><li><strong>Clear decisions.</strong> Anyone can read a profile to see how routing is set up. From Order Module 2.3.0, each sourcing decision can also leave a Sourcing Audit: the strategies tried, which conditions passed, and the location scores and fulfilments for the winning strategy. So you can explain not just the rules, but why this order went where it did.</li><li><strong>Consistent behaviour.</strong> Distance, stock and split logic behave the same across every channel and project.</li></ul><p>Custom code has not gone away, but its role is narrower. When a need goes beyond the built-in options, such as preferred carrier or geofenced zones, developers write a single custom condition or criterion. Developers implement and register it in their module, then expose its schema through the matching custom setting (fc.rubix.order.sourcing.criteria.custom or fc.rubix.order.sourcing.conditions.custom). It then appears alongside the reference building blocks. The framework still handles strategy order, scoring, splits and fulfilment creation. Developers write one small, testable piece instead of the whole engine.</p><p>A few things are worth knowing when configuring:</p><ul><li>Operators are lowercase (equals, not EQUALS).</li><li>Place exclusions before sorters to shrink the candidate pool.</li><li>Keep maxSplit modest on large networks; split search grows quickly with more locations.</li><li>Sourcing checks on-hand stock (inventory positions), not the available-to-sell figure shown on the storefront.</li></ul><h3>Conclusion</h3><p>Sourcing is where order orchestration makes its most valuable decision, and Responsive Sourcing makes that decision configurable. Strategies, conditions and criteria replace bespoke rule code, so routing can keep pace with the business.</p><p>Teams spend less time rebuilding distance checks and split logic, and more time tuning routing for speed, cost and customer experience. Custom code is still there when it is truly needed, as a small plug-in rather than the whole engine.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2085a44df337" width="1" height="1" alt=""><hr><p><a href="https://medium.com/techunfiltered-ayata/responsive-sourcing-the-heart-of-order-orchestration-made-flexible-2085a44df337">Responsive Sourcing: The Heart of Order Orchestration, Made Flexible</a> was originally published in <a href="https://medium.com/techunfiltered-ayata">TechUnfiltered@Ayata</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>