Lleverage vs Tekst: AI agents past the inbox
Tekst works the mailbox in front of SAP. Our AI agents do the job inside the ERP itself.
“We had four and a half people · about 3.8 full-time equivalents · sitting there all day long, manually entering every single order into our Business Central ERP.”
Bryan van IngenTopa Bathroom Products
Tekst vs Lleverage, side by side
We read Tekst's own pages and its support documentation on 20 October 2026: the homepage, how it works, the SAP order automation page, the order intake and shared inbox use cases, and the integration guides for SAP S/4HANA and Infor M3. Where Tekst does not describe something, the table says so. What Lleverage does comes from our order management and governance pages.
| Capability | ||
|---|---|---|
| Where the work happensThe screen your team keeps open | The agent fills the same fields your order desk fills in the ERP, so the ERP checks the work the same way it checks a person's | In the mailbox, with the Tekst Sidepanel described as the last checkpoint before SAP, right inside Outlook |
| Orders in any formatEmail, PDF, spreadsheets, portal exports | Read in each customer's own layout, and every correction teaches the agent that layout | PDFs, Excel files, web forms and screenshots, with missing or inconsistent details flagged, such as prices, SKUs and supplier terms |
| Creating the sales orderWhat actually reaches the ERP | The sales order created in your ERP, and the confirmation sent back once the ERP has accepted it | A Create Sales Order block creates a sales order in SAP S/4HANA directly from an automation flow, and the Infor M3 integration builds one from a company number, a customer number and order lines with item number and quantity |
| Mapping how the work runs todayProcess mining | Task and process mining are coming soon | Process mining with a Context Graph Builder and patent-pending Universal Tracing across email, ERP and CRM. Its homepage says mapping a process used to take six months and now takes an afternoon |
| Sorting and routing a shared mailboxTriage, assignment, volume reporting | Order and delivery questions answered from the live record in the channel the customer used, without sorting a department's whole mailbox | Every incoming email tagged and assigned by topic, urgency and language, routed across sales, finance and legal, with dashboards for volumes, response speed and bottlenecks |
| Supplier invoicesWhat happens to an incoming invoice | Three-way matching within the tolerances finance sets, payments applied and debtors reminded, with disputes handed to credit control and no payment ever released by the agent | Invoice emails and attachments read, key metadata extracted and a matching purchase order or reference checked, with lookup, validation and audit logging in SAP, Oracle and MS Dynamics. Three-way matching is not described |
| Purchasing, planning and master dataThe jobs on the same desk | Purchase orders raised and every supplier confirmation read back against them, the plan redone when reality changes, and article records kept in step across systems | Stock levels tracked and re-orders triggered as validated purchase orders synced with SAP, Odoo and NetSuite. Production planning and master data are not described |
| ERPs with a published connection | SAP, Business Central, Dynamics 365 F&O, Navision, Infor, Exact, AFAS, Epicor, Ridder iQ, Oracle, NetSuite and the IBM AS/400 | SAP S/4HANA and Infor M3 in the support guides, beside SAP Service Cloud, Salesforce, Zendesk, Outlook, Gmail, SharePoint, FTP and S3. The marketing pages also name Oracle, NetSuite, Odoo and MS Dynamics |
| How it gets into SAP | Our team sets the connection up with your SAP team, and the SAP account it uses decides what the agent may see and change. Lleverage is an SAP partner | A Tekst iFlow package deployed on your own SAP BTP Cloud Integration tenant, called with a service key that carries a dedicated Tekst.ESBMessaging.send role on its own Process Integration Runtime instance |
| Older systems and no API | On-premise systems reached without opening them to the internet, and where there is no API the agent works the screen with the access a person has | Not described: the published integrations are cloud services, and the SAP route runs through BTP Cloud Integration |
| Asking from a chat assistant | Dashboards can be read and written through our MCP tools that AI assistants connect with, and the agent brings the work to your team in Teams, Slack or the inbox | A remote MCP server that connects a Tekst workspace to Claude, Google Gemini and Microsoft Copilot Studio |
| Deciding what runs on its own | A confidence line per process. Below it a person decides, above it the work goes through, and only you move the line | The Sidepanel is the checkpoint before SAP, and follow-up questions go out when details are missing. A threshold you set per process is not described |
| Who gets it running | Forward deployed engineers build the first agent on your own documents and stay on as your customers change | Its homepage promises integration within 1 hour, process intelligence within 1 day to 1 week and agentic automation within 1 month |
When each one fits
Tekst stays with the mail and shows you how the work really moves. Lleverage carries the work into the ERP and keeps going into the jobs behind it.
When Tekst fits
- The first question is where the time goes, and you want the process mapped before anything is automated.
- One mailbox takes thousands of messages a week in several languages, and the job is to tag, assign and report on them.
- Your team wants to stay in Outlook, with a checkpoint in the side panel before anything reaches SAP.
- Your ERP is SAP S/4HANA or Infor M3, and your SAP team is happy to run a vendor iFlow package on your own BTP tenant.
- You want your Tekst workspace available to Claude, Gemini or Copilot Studio through an MCP server.
When Lleverage fits
- The people reading the mail also book the supplier invoice, chase the purchase order and replan the week.
- You run Exact, AFAS, Ridder iQ, Business Central or an AS/400, or a system with no API where the agent has to work the screen.
- You want each process to sit at a confidence line your team sets, and one setting to put it back under full supervision.
- The order has to be created and confirmed in the ERP itself.
- You want the engineers who built the first agent still there when your customers change their layouts.
Governance & security, built in.
AI that acts in your operations has to be accountable. Approvals, permissions and a full audit trail are part of the platform: hosted in Europe, compliant by design, and your data always yours.
Book a demoAgents ask before they act. You approve what matters.
Every step logged end to end. Trace any decision back to its source.
Your data stays in Europe. And stays yours.
Role-based access for people and agents, down to a single action.
Audited controls, EU data protection, zero retention.
What the agent does after the mail
These are the jobs the order management agent picks up after the mail, in the systems you already run.

- 01
Creates the order and confirms it
Customer, item, price, availability and credit checked line by line, the sales order created in your ERP and the confirmation sent back to the customer once the ERP has accepted it.
- 02
Matches the supplier invoice
Price, quantity and cost centre held against the purchase order and the goods receipt, inside the tolerances finance sets. A dispute reaches credit control with the history gathered, and releasing a payment stays with a person.
- 03
Reads every supplier confirmation
Purchase orders raised, and each confirmation checked back against what was ordered. Buyers see the lines where a date, a price or a quantity moved, and nothing else.
- 04
Rebuilds the plan when the week changes
A forecast drawn from the order book and a production plan made again when a rush order lands or a machine stops.
- 05
Holds the records in agreement
Customer and article data kept in step across every system that holds it, so the next order matches a real item at a real price the first time it is read.
- 06
Answers with the order open
Where the goods are, what the lead time is and whether a date can move, each answered from the live record in the channel the customer used, with the shipping papers drawn from the same order.
- 07
Works to a line your team sets
Every process has its own confidence line. Below it a person decides, above it the agent carries on, and one setting returns the whole process to supervision with nothing to rebuild.
From the teams running it
“Customer support gets a lot of questions every day, but 80% of the questions are just the same.”
Jeremy ParsserJ. Kisch & Zonen
“We’re constantly looking for ways to ensure we can last another 100 years and further improve our service and quality.”
MartKoninklijke Dekker
“Nobody can memorize every manual from every manufacturer. When a customer calls with a machine problem, the quality of the answer depends on who picks up the phone and what they happen to remember.”
See it on your own documents.
A 30-minute demo, on real data from your back office.
Tekst vs Lleverage: questions
What is the difference between Tekst and Lleverage?
Tekst sits on the communication around the work. It mines how a process runs across email, ERP and CRM, sorts and routes the shared mailbox, and puts a review step in Outlook ahead of SAP. Lleverage puts AI agents in the ERP itself, where they create the sales order and go on to invoices, purchasing, planning and master data.
Does Tekst create the order in SAP, or only prepare it?
Its support guides describe a Create Sales Order block that creates a sales order in SAP S/4HANA directly from an automation flow, and an Infor M3 integration that creates orders from a company number and order lines. Its Outlook side panel is described as the last checkpoint before SAP. How a person reviews the order there is not described.
Which ERPs does each one connect to?
Tekst documents SAP S/4HANA, reached through an iFlow package on your own BTP Cloud Integration tenant, and Infor M3. Lleverage is an SAP partner and works in SAP, Business Central, Dynamics 365 F&O, Navision, Infor, Exact, AFAS, Epicor, Ridder iQ, Oracle and NetSuite, and on an AS/400, where the agent works the screen with the access a person has.
When is Tekst the better choice?
When the mailbox itself is the problem. Tekst mines the process before it automates anything, tags and assigns every incoming email by topic, urgency and language, and reports on email load, response speed and bottlenecks. Lleverage does none of that: our task and process mining is coming soon.
Can we keep working in Outlook with Lleverage?
Yes. Orders, confirmations and supplier replies keep arriving in the mailbox your customers already write to, and nothing moves to a new inbox. The difference is where a person gets involved: only on the lines the agent stops on, in Teams, Slack or that same inbox, with the reason attached.
Do we have to map our processes before we start?
No. The agent is built on the documents your desk already handles, your own orders and invoices, and every correction your team makes shapes it further. Our forward deployed engineers start from those documents with your team. Tekst takes the other route and mines the process first, automating afterwards.