Talk to an expert
Guide

Contract pricing online: real-time ERP calls or synced price tables?

The trade-offs in speed, accuracy and ERP load, and how to choose for your catalog size.

CommerceEdge team4 min read
Two ways the storefront gets a price. Real time asks the ERP on each view, so the price is fresh and the ERP is busy, and the price stops if the ERP is down. Synced copies prices on a schedule into a table, so the page is fast and a price still shows if the ERP is down, but it can lag.

Real time

  1. Storefront
  2. Live callSlowerEach view
  3. ERPFresh
  4. PriceStops if down

Synced

  1. Storefront
  2. Scheduled syncOn a schedule
  3. Price tableCan lag
  4. PriceStill shows

A distributor’s price is rarely a single number. It’s the result of a set of rules: the customer’s contract, their price class, quantity breaks, the unit of measure, promotions with start and end dates, and sometimes the branch the order ships from. Your ERP already knows how to evaluate those rules, because that’s where your team maintains them.

The question every B2B commerce project has to answer early is how the storefront gets the result. There are two basic patterns, and most mature builds end up using a mix of both.

Pattern 1: Ask the ERP in real time

When a logged-in buyer views a product, list or cart, the site calls the ERP (directly or through middleware) and asks, “What does this account pay for these items at these quantities?”

What you gain

  • One source of truth. The price shown online is the price the ERP will invoice. No second copy of the rules to keep in step.
  • Every rule is supported. If the ERP can calculate it, the site can show it, including the odd exceptions your sales team negotiated years ago.
  • Changes are immediate. A new contract entered at 9:00 is live at 9:01.

What it costs

  • Latency. Each call adds time to the page. A category page with 48 products and several units of measure can mean a lot of price lookups unless you batch them.
  • ERP load. Browsing traffic becomes ERP traffic. Bots, buyers comparing items and search pages all hit the system that also runs your warehouse and accounting. Check your ERP’s API limits and licensing before you assume capacity.
  • Coupled availability. If the ERP or its API is down for maintenance, the site either hides prices or shows something you can’t stand behind.

Pattern 2: Sync price data into the commerce platform

Prices, or the inputs to them, are exported from the ERP on a schedule or on change, and stored in the commerce platform’s own price structures (shared catalogs and price lists in Adobe Commerce, catalogs and price lists per company location in Shopify B2B).

What you gain

  • Speed. Prices are served from the platform and its cache, so category and search pages stay fast.
  • Resilience. The storefront keeps working while the ERP is busy or offline.
  • Predictable ERP load. The integration runs on your schedule, not your visitors’.

What it costs

  • Volume. Customer-specific pricing multiplies quickly. If you store one price per customer per SKU, the table grows with customers × SKUs × units of measure. That’s often where this pattern breaks down, not in the platform but in sync time and data hygiene.
  • Staleness. Between syncs, the online price can differ from the ERP. You need a rule for what happens at checkout when they disagree.
  • Rule translation. The platform’s price model is simpler than most ERPs. Rules such as “contract price, unless the promotional price is lower, but never below cost plus a margin floor” may not have a native equivalent, so someone has to flatten them, and flattening is where errors creep in.

The hybrid most distributors end up with

In practice the strongest pattern separates browsing from committing:

  1. Sync what’s stable and shared. Base lists, price classes and quantity-break tiers that apply to many accounts change rarely and fit platform price structures well.
  2. Calculate what’s specific at the moment it matters. Fetch the account’s exact price in real time on the product page, in the cart and at checkout, in batches, with a short cache per account.
  3. Re-price on order submission. The ERP prices the order when it’s created. If that differs from what the buyer saw, show the change before they confirm rather than after.
  4. Fail gracefully. If the live call times out, fall back to the synced price and label it (“Your contract price will be confirmed at checkout”) rather than showing nothing or showing list price as if it were final.

How to choose for your catalog

Work through these questions with whoever owns pricing in the ERP, not only the commerce team:

  • How many accounts have unique pricing, and how many SKUs do they buy? A few hundred contract accounts on a narrow catalog can sync comfortably. Thousands of accounts on a deep catalog usually can’t.
  • How complex are the rules? List them. If any rule depends on the full order (mixed-item quantity breaks, order-level discounts), the final price must come from a calculation, not a lookup.
  • How often do prices change? Daily cost updates or frequent promotions shorten the acceptable sync interval.
  • What can the ERP API handle? Ask for documented limits and test with realistic batch sizes. For Acumatica, confirm how the price your sales staff see on a sales order is exposed through the API you plan to use, and test it against real customer accounts.
  • Where do buyers need the exact price? Some distributors show contract prices on listing pages; others show them from the product page onward. That decision alone can change call volume dramatically.

What to decide before build starts

Write down three things and get them agreed: the source of truth for each price component, the maximum age of a price shown online, and what the site does when the ERP doesn’t answer. Those three decisions shape the integration design more than the choice of platform does.

If you’d like a second opinion on your pricing rules before a project starts, that’s what our architecture review covers.