FAQs

General

What exactly is Prosessed AI and who is it built for?

Prosessed AI is a full stack operating system built specifically for B2B businesses, importers, wholesalers, distributors, and manufacturers. Think of it as the connective tissue between every part of your business: the orders coming in, the stock sitting in your warehouse, your field sales team out visiting customers, and the decisions your team makes every day.

Most businesses in this space run on a patchwork of spreadsheets, WhatsApp groups, and disconnected ERPs. Prosessed AI replaces that fragmentation with a single unified platform that connects demand (customers, orders, sales reps), supply (inventory, procurement, warehouses), and intelligence (analytics and AI driven recommendations).

How is this different from a traditional ERP system?

Traditional ERPs are backend heavy systems built for record keeping and compliance. They are rigid, slow to configure, and not designed for the frontline reality of B2B commerce such as field reps placing orders, warehouse staff tracking batches, and customers wanting to reorder on mobile.

Prosessed AI takes an execution first approach. It makes daily operations faster first, then layers intelligence on top. The result is a system your team will actually use. It is also designed to work alongside your existing ERP rather than force a complete replacement.

What are the core modules that make up the platform?

Five tightly integrated modules share the same data:

Order Management System (OMS)

Order creation, tracking, credit enforcement, and pricing across every channel

Inventory and Warehouse Management

Real time multi location stock with batch and expiry tracking

Sales Representative App

Mobile first tool for field teams that works offline

B2B Commerce Platform

Self serve ordering portal for customers on web or mobile

Jerry AI

The intelligence layer for forecasting, recommendations, and automated communication

An order placed by a sales rep instantly updates inventory, triggers procurement recommendations if needed, and is visible to the customer in real time.

How does Prosessed AI improve profitability, not just efficiency?

Faster order processing means more orders fulfilled daily without adding headcount. Better inventory accuracy reduces capital tied up in excess stock and prevents stockouts. AI driven procurement planning reduces emergency purchases at poor prices. Self serve ordering captures revenue that would otherwise be lost outside business hours. Sales visibility lets managers catch performance gaps while there is still time to act on them in the current period.

How does the platform handle role based access across a business with multiple departments and locations?

Access controls in Prosessed AI are structured around roles, not just individual users. You define what each role can see and do. A warehouse operative at Location A should not have visibility into pricing negotiations happening at Location B, and a sales rep should not be able to override a credit block without manager approval.

Roles can be scoped by location, function, and data type. A regional sales manager sees their team's activity and their region's customers. A procurement head sees purchase recommendations and supplier data but not necessarily customer specific pricing. A finance user sees invoices, credit limits, and payment status across the business.

This granularity matters operationally because it prevents accidental data exposure, keeps each team's interface clean and relevant, and maintains an audit trail of who did what which becomes important when tracing a pricing error or a disputed order.

If we operate across multiple entities or legal companies, can a single Prosessed AI instance handle them, or do we need separate setups?

Multi entity operations such as a holding company with separate import, distribution, and manufacturing arms are a common configuration for the businesses Prosessed AI serves. The platform is designed to accommodate this with separate entity level data including invoicing, pricing, and stock that can still be viewed in consolidated form by group level management.

In practice this means each entity can have its own customer base, pricing rules, warehouse locations, and order workflows, while a group operations or finance head can see across all of them from a single dashboard. This avoids the scenario where each entity runs a completely separate system, making group level reporting a manual reconciliation exercise every month.

For specific configurations involving intercompany transactions or consolidated reporting requirements, the implementation team can scope what is achievable during onboarding.

What does a typical data migration look like when moving historical orders, customer records, and pricing from an existing system?

Migration typically covers three categories of data: master data such as customer records, product catalog, pricing tiers, warehouse locations; transactional history such as past orders, invoices, stock movements; and operational configuration such as workflows, approval rules, and credit limits.

Master data and configuration are migrated before go live. Historical transactional data is important for the AI layer to generate meaningful forecasts and recommendations, so it is prioritized during onboarding. The cleaner your source data, the faster this process goes. Businesses migrating from well maintained spreadsheets often move faster than those coming off legacy ERPs with years of inconsistent data entry.

The team provides structured import templates and validation tooling so data issues are caught before migration rather than discovered in production. Realistic timelines range from a few days for simpler setups to a few weeks for operations with complex pricing structures or large product catalogs.

Inventory and Warehousing

How does inventory tracking work across multiple warehouses?

Each warehouse location has its own stock record and movements inbound, outbound, transfers are tracked as they happen. When an order is allocated to a specific warehouse, available stock there updates instantly. This prevents selling stock committed elsewhere and gives a unified view across all locations from one dashboard.

Does the platform handle batch tracking and expiry dates?

Yes, built in not an add on. Each batch is tracked separately with arrival date, expiry date, cost, and location. The system alerts you as stock approaches its expiry window and supports FEFO first expiry first out fulfilment to reduce write offs and ensure customers receive products with acceptable remaining shelf life.

How does the platform help prevent stockouts and overstock?

Two layers exist. Configurable low stock alerts for operational control and AI driven purchase recommendations that factor in historical demand and lead times. This prevents both extremes stockouts that damage customer relationships and overstock that ties up cash.

How does the system handle stock adjustments such as cycle counts, write offs, goods returned from customers and how quickly do those changes reflect in committed stock calculations?

Stock adjustments whether from a cycle count variance, a write off of damaged or expired goods, or a customer return being reinstated into available stock are processed as real time transactions in the system. There is no batch overnight run or manual recalculation required. The moment an adjustment is posted, it flows through to available stock, committed stock, and any pending order allocations that depend on that inventory.

This matters most in the gap between ordering and receiving. If your warehouse team conducts a cycle count mid week and discovers a variance on a product that has 10 open orders allocated against it, the system immediately reflects the corrected position. Order fulfilment logic can then flag orders that are now at risk rather than your team discovering the shortfall on the day of dispatch.

Returns from customers follow a similar logic. The returned goods are inspected, a condition is assigned such as resaleable, damaged, or hold for supplier credit and the system updates stock accordingly. Only goods marked resaleable re enter available inventory. The rest are quarantined with a reason code until disposed of. Every adjustment carries an audit trail including who made it, when, and why which is important both for internal controls and supplier claims on damaged goods.

If we source the same product from multiple suppliers with different landed costs, how does the system track margin per order rather than averaging it across the whole SKU?

Batch level cost tracking is the key here. Because each batch of stock is recorded with its own cost including landed cost components such as freight, duties, and handling the system knows what each unit in a specific batch actually cost to bring in. When an order is fulfilled from a particular batch the margin calculation uses that batch's cost rather than a blended average across all stock of that SKU.

This is particularly important for importers and distributors who buy the same product across multiple shipments at different prices. A consignment bought at last quarter's exchange rate has a materially different cost than one bought after a currency movement and averaging them obscures which sales are actually profitable and which are not.

The practical output is that your team can see the actual margin on each order line at the time of fulfilment rather than a theoretical margin based on average cost. Over time this data also feeds into the AI layer's procurement recommendations which can factor in cost differences between supplier batches when suggesting reorder timing.

How does the platform manage stock that is in transit, goods that have left a supplier but not yet arrived at our warehouse?

In transit stock, goods on order from a supplier that have not yet been received, is tracked as a distinct inventory state. It does not count as available stock but it is visible in your procurement and planning views so that the team has an accurate picture of what is coming and when.

For businesses with longer supply chains such as importers bringing goods from overseas this distinction is critical. If your purchasing team can only see current on hand stock they will over order because they cannot see what is already inbound. With in transit visibility procurement recommendations account for expected arrivals preventing you from raising unnecessary purchase orders for stock that is already on its way.

Expected arrival dates update as they change. If a shipment is delayed that change flows through to the planning view and any orders that were being held pending that stock can be flagged for review. This keeps the gap between the physical reality of your supply chain and the data your team is working from as narrow as possible.

Jerry AI

What is Jerry AI and what does it actually do?

Jerry AI is the intelligence layer across Prosessed AI. It handles demand forecasting, smart inventory and procurement recommendations, automated customer follow ups via WhatsApp and voice, raw material planning for manufacturers, and conversational ordering interfaces. It learns from the operational data your business generates on the platform.

Is the AI replacing human decision making or supporting it?

Supporting it. The AI handles data crunching, pattern recognition, and routine follow up not the complex judgments that experienced people make. A procurement manager still decides whether to follow a purchase suggestion. A sales manager still sets priorities. Jerry AI is the well briefed analyst working in the background not the decision maker.

How does the AI know enough about my business to make useful recommendations?

It learns from your own operational data including order history, customer behaviour, stock movements, and procurement cycles. The execution first philosophy exists precisely for this reason because clean structured operational data is the foundation the AI needs to be useful. Businesses that try to implement AI before their data is clean get poor results. Prosessed AI is designed to avoid that by digitising operations first.

How does the demand forecast handle seasonality, one off promotions, and irregular spikes that would distort a standard average?

A straight average of historical demand is one of the least useful inputs for wholesale planning and Jerry AI does not rely on it. The forecasting model separates baseline demand from seasonal patterns and one off events. A spike in December for a product that sells steadily the rest of the year is treated as a seasonal signal. The model increases the forecast for the equivalent period next year rather than raising the annual average.

Promotions are handled as tagged events in the system. When you run a promotional period you mark it as such which allows the model to exclude that spike from the baseline forecast and instead treat it as an event driven uplift. If you plan to run the same promotion again you can instruct the forecast to include the promotional uplift for that future window.

Outliers such as a one off bulk order from a customer that does not reflect recurring demand can be excluded or downweighted so they do not skew future recommendations. The model also gets more accurate over time as it accumulates more of your business specific history which is why forecast quality in month 12 is meaningfully better than in month 1.

When the AI recommends a purchase order what specific inputs is it using and can we see the reasoning not just the output number?

Transparency in AI recommendations is important because a procurement manager who cannot see why they are being told to order 400 units instead of 200 is unlikely to trust or act on the recommendation. Jerry AI surfaces the inputs behind each purchase suggestion not just the output.

The recommendation view shows current on hand stock, quantity already on order and in transit, forecasted demand over the coverage period typically your supplier lead time plus a buffer, the reorder point that triggered the recommendation, and the suggested quantity with the logic behind it. If a spike in recent orders is driving a higher than usual recommendation that is visible. If a long lead time from a specific supplier is pushing the trigger earlier that is shown too.

Your procurement team can accept, adjust, or reject the recommendation and adjustments they make consistently such as always ordering 15 percent more than suggested for a specific supplier are factored into future recommendations over time. The system learns from the pattern rather than repeatedly suggesting something the user always overrides.

How does Jerry AI handle automated customer outreach without it coming across as generic or poorly timed the kind of message that damages the relationship instead of recovering the order?

Automated outreach done poorly such as a generic we miss you message sent to a customer who placed an order three days ago erodes trust faster than silence. Jerry AI outreach logic is built with this in mind.

Messages are triggered by specific behavioural signals such as a customer being outside their normal reorder window, a customer's basket shrinking compared to their typical order, or a customer opening the B2B portal but not completing an order. Each trigger is calibrated per customer based on their individual pattern rather than a blanket rule applied to everyone.

The content is personalised to that customer's history referencing the specific products they usually order, their last order date, and any relevant context rather than a template message. Timing rules prevent outreach from going out at midnight or on known public holidays in the customer's region. If a customer has been contacted and has not responded escalation logic hands the account back to the assigned sales rep rather than continuing to send automated messages to someone who is actively choosing not to engage.

Orders And OMS

What is the Order Management System and what does it handle?

The OMS handles the full lifecycle of a sales order from placement through processing and fulfillment to tracking and completion. It supports multi channel order capture including sales reps, B2B portal, WhatsApp, and API, real time tracking, automated workflows, credit limit enforcement, and flexible pricing. It is designed to scale from 50 orders a day to 5000 while keeping workflows consistent and auditable.

Can customers place orders themselves, or does everything go through our sales team?

Both run in parallel. Sales reps place orders through the app. Customers place orders through the B2B portal. All of them flow into the same OMS, giving a single consistent view of everything. The self serve channel is particularly valuable for capturing repeat orders outside business hours and reducing admin load on the sales team.

Does the platform support different pricing for different customers?

Yes. You can define multiple price tiers and assign customers to them, with individual overrides for specific customer product combinations. When a rep places an order or a customer logs into the portal, they see only the prices that apply to them pulled automatically from their profile with no risk of quoting the wrong rate.

Can orders be placed via WhatsApp? How does that work?

Yes. Jerry AI interprets the WhatsApp message, confirms the details with the customer, and logs it as a structured order in the OMS. Your operations team sees a clean order instead of a screenshot to type manually. Many B2B customers already use WhatsApp with their suppliers and will not change that habit, so the platform works with it.

How configurable are the order approval and fulfilment workflows, can we define our own stages, conditions, and escalation rules?

Order workflows in Prosessed AI are configurable rather than hardcoded. You can define the stages an order moves through from placement to credit check, to picking, to dispatch, to invoice and set the conditions that trigger each transition. For instance, orders above a certain value might require a secondary approval. Orders to customers with overdue balances might be automatically held pending finance review.

Escalation rules can be layered on top. If an order has been sitting at a stage for more than a defined period without action, it can be flagged to a supervisor automatically rather than silently ageing in a queue. This matters in high volume operations where manual monitoring of every order's progress is not realistic.

Custom stages are important for businesses with unique fulfilment models such as a business that does a quality check before dispatch or one that splits orders across multiple warehouses before consolidating for delivery. The workflow engine accommodates this rather than forcing a generic four step process. For highly specific requirements, the implementation team can scope what is configurable versus what would need a custom discussion.

When a customer disputes an invoice, wrong quantity, wrong price, short delivery, how does the platform handle credit notes and order corrections without creating a reconciliation mess?

Invoice disputes and corrections are a routine reality in wholesale trade and handling them poorly creates reconciliation problems that compound over time. Credit notes that never get applied, stock adjustments that do not match the accounting, outstanding balances that nobody can explain.

In Prosessed AI, corrections are handled as structured transactions linked to the original order. A short delivery generates a quantity adjustment that flows back to inventory, stock not delivered is reinstated, updates the invoice value, and adjusts the customer's outstanding balance all as a connected chain of records rather than separate manual entries that have to be reconciled later.

Credit notes are raised against a specific order line with a reason code, giving finance a clear audit trail of what was corrected, why, and by whom. This makes month end reconciliation significantly cleaner and gives customer facing teams accurate balance information when a customer queries their account.

Can we run promotional pricing or volume based discounts for specific periods without manually updating every customer's price list?

Yes. Promotional and volume based pricing rules can be defined at the platform level and applied automatically based on conditions such as customer tier, product category, order quantity, or date range without manually editing individual customer price lists each time.

A promotion running for a specific SKU across all Tier 2 customers during a defined window is configured once and applies automatically to every qualifying order placed during that period. When the window closes, standard pricing resumes without any manual rollback. Volume breaks where a customer ordering 50 units gets a different rate than one ordering 10 are similarly handled as rules rather than manual overrides, so they apply consistently across channels whether the order comes through the B2B portal, the sales rep app, or WhatsApp.

This removes one of the most common sources of pricing errors in wholesale operations such as a promotion that someone forgot to remove or a volume discount that was applied inconsistently depending on which rep took the order.

Pricing and Implementation

How is Prosessed AI priced?

Pricing is structured around the modules you use and the scale of your operation. You can view base plans on the pricing page or speak with the team for a proposal tailored to your specific configuration particularly relevant for larger or multi entity deployments.

Is there a free trial available?

Yes, you can get started for free and explore the platform before committing. For businesses evaluating a larger rollout a guided pilot with implementation support is typically the better path. Reach out to the team to arrange it.

How long does implementation typically take?

Core OMS and Sales Rep App deployments can go live within days to a couple of weeks. More complex configurations such as multiple warehouses, ERP integrations, or custom pricing take longer. The platform is configured not coded which keeps timelines significantly shorter than traditional ERP projects. Most businesses are processing live orders within weeks not months.

What does a realistic integration with an existing ERP like SAP or Infor actually involve and who owns that technical work?

ERP integration is one of the most commonly underestimated parts of any software deployment and it is worth being concrete about what it involves. Prosessed AI connects to existing ERPs via API and has built integrations with common systems including SAP and Infor. The depth of integration such as what data flows in which direction and how often is scoped during the onboarding process based on what each side of the system owns.

A typical configuration has Prosessed AI handling the frontline operational layer including order capture, inventory tracking, sales rep workflow, and customer communication while the ERP remains the system of record for finance including general ledger, accounts payable, and statutory reporting. Data flows between the two on a defined schedule. Orders confirmed in Prosessed AI are posted to the ERP. Stock receipts recorded in the ERP are reflected in Prosessed AI.

The Prosessed AI implementation team handles the configuration on their side. Your IT team or ERP partner handles any configuration required on the ERP side. The scope of that work depends on how standardised your ERP configuration is and whether your ERP vendor API is accessible. For systems with limited API access or heavily customised ERP environments the integration scope and timeline should be discussed explicitly before committing to a go live date.

How do you measure whether the platform is actually delivering ROI and what metrics should we be tracking in the first six months?

ROI from a platform like this shows up across several dimensions and the right metrics depend on which problems you were trying to solve. For most B2B operations the six month metrics worth tracking fall into a few buckets.

On the order side you should track order processing time from placement to confirmation, order error rate such as wrong products wrong quantities wrong pricing, and the share of orders coming through self serve channels versus phone or email. The first two should drop and the third should climb.

On inventory you should track stock accuracy comparing physical versus system, frequency of stockouts on high velocity SKUs, and inventory holding days. Better accuracy fewer stockouts and lower holding days together is the target combination.

On sales productivity you should track orders per rep per day, customer visit frequency against target, and the percentage of at risk accounts that were re engaged before they churned. On working capital you should track days inventory outstanding and the reduction in emergency procurement events.

These are measurable from the data the platform generates. Within the first few months most businesses have a clearer picture of where the gains are coming from than they had with any previous system simply because the data is being captured consistently for the first time.

What happens to our data if we ever decide to leave the platform and can we export everything cleanly?

Data portability is a reasonable thing to ask about before committing to any platform and the answer should be clear rather than vague. All data you generate on Prosessed AI including orders, customer records, inventory history, pricing configurations, and invoices is yours and it is exportable in structured formats.

Full data exports can be requested at any time not just on exit. This is useful for businesses that want to run their own analysis in external tools, share data with auditors, or maintain a local archive alongside the platform. On exit the team provides a complete data export in standard formats so you are not locked out of your own operational history.

For specific data residency requirements such as which jurisdiction your data is stored in and how long it is retained after contract end these details should be confirmed with the team as requirements vary and the specifics matter for businesses in regulated industries or operating across multiple countries.

Sales Teams

What does the Sales Rep App do for field teams?

It replaces notebooks, WhatsApp messages, and printed price lists with a single mobile tool. Sales reps can place orders with correct customer pricing loaded automatically, check real time stock before committing, log visits and notes, plan routes, and access invoice history all in the field. Works offline too, syncing automatically when connection is restored.

How does the platform give managers visibility into what their sales team is doing?

Every action is recorded centrally including orders placed, customers visited, routes traveled, and notes added. Managers get a live view of field activity without relying on manual reports. They can see which customers have not been visited recently, which reps are hitting targets, and where activity is dropping off while there is still time to act on it.

How does the B2B Commerce Platform reduce the burden on the sales team?

A large part of what most B2B sales teams do is order taking for established customers who want the same things they ordered last month. The B2B portal lets those customers self serve by browsing their catalog, seeing their pricing, checking credit limits, and reordering in a few taps. This shifts routine reorders off the sales team entirely and lets them focus on growing accounts and handling situations that genuinely need human engagement.

How do you prevent a sales rep from offering a discount or overriding a price in the field without eroding margin across the whole business?

Pricing controls in the Sales Rep App work at the role level. A standard sales rep can only sell at the pre configured price for that customer. There is no field to type in a discount. If a rep wants to offer a price override they must raise it for manager approval which is handled as a workflow step rather than a verbal conversation that never gets recorded.

Managers can be granted a discretionary discount ceiling such as up to 5 percent below the customer's standard price without needing to escalate further. Any override above that ceiling requires a second approval from a commercial director or pricing team. Every approved override is logged with the reason code, the approver, and the order it applies to.

This creates a structure where the field team has some flexibility to close deals but within guardrails that protect overall margin. It also gives the commercial team visibility into how often overrides are being requested, by which reps, for which customers, and whether the discount pattern suggests a pricing tier needs to be renegotiated rather than constantly overridden on a case by case basis.

Can the system tell us which customers are at risk of churning based on their ordering behaviour before they actually stop buying?

Yes. Jerry AI tracks order frequency and basket composition for each customer and flags deviations from their established pattern. A customer who usually orders weekly and has now gone three weeks without an order is surfaced as a risk rather than discovered after they have quietly moved their business to a competitor.

The signals the system monitors go beyond just frequency. A customer who is ordering the same frequency but with smaller basket sizes or who has dropped specific product categories they used to buy regularly is also flagged. These are often early indicators of a customer trialling a competing supplier rather than a clean break and they are much easier to address at that stage than after the relationship has fully eroded.

Sales managers can see these at risk accounts in their dashboard and assign follow up actions to specific reps. Jerry AI can also trigger automated outreach such as a WhatsApp message or call for accounts that have gone quiet ensuring no customer silently disappears simply because no one had the bandwidth to notice.

How does the platform handle sales territories ensuring reps only see and act on their own accounts without creating blind spots for management?

Territory management works through account ownership assignments at the customer level. Each customer is assigned to one or more sales reps and a rep's app view is scoped to their assigned accounts. They do not see customers belonging to another territory which prevents accidental double calling and protects customer relationships built by a specific rep.

At the manager level this scoping is lifted. A regional sales manager sees all accounts within their region across all reps. A national sales director sees everything. This layered visibility means territory boundaries enforce clean accountability for individual reps while giving leadership the unobstructed view they need to spot patterns, reassign underserved accounts, or identify customers who are slipping through the gaps between territories.

Territory reassignment when a rep leaves or a region is reorganised is handled at the account level and takes effect immediately so there is no period where customers fall into a grey zone with no active owner.