Sales reps at a wholesale distributor I worked with were spending three, sometimes four days a week on the phone taking reorders from retailers who had ordered the same twenty SKUs the month before. This b2b ordering portal case study is about what happened when we moved those repeat orders to self-service and let the reps go do what reps are actually good at: opening new accounts.
The distributor supplies packaged goods to roughly 300 small retailers, minimarkets, and warungs across a mid-size Indonesian city. Reorder volume was predictable, almost boringly so, but every single one of those predictable orders still required a human to answer a phone, write it down, and key it into the order system by hand.
That is not a sales process. That is a very expensive data entry job wearing a sales title.
Where the time was actually going
We shadowed three reps for a week before writing a line of code. The split surprised the owner more than it surprised me:
- About 65 percent of rep time went to taking and confirming repeat orders from existing retailers.
- About 20 percent went to chasing payment confirmations before releasing an order.
- Only 15 percent went to anything you could call actual selling: new accounts, upselling new product lines, negotiating shelf space.
The owner had hired good salespeople and then buried them under logistics work a form could do. That is the pattern behind most B2B ordering portal case studies worth telling: the problem is rarely the sales team, it is what the sales team has been forced to spend its day on.
What we built
Nothing exotic. A web portal, mobile-friendly because most retailers place orders from a phone, with:
- A catalog scoped to that retailer's usual price tier and product mix, not a generic price list
- One-tap "reorder last basket" for the recurring SKUs, editable before submitting
- Real-time stock visibility per warehouse, so retailers stop ordering things that are out
- Order status and delivery tracking, so "where's my truck" calls stopped
The reorder button did more work than anything else in the build. Most retailers order the same core basket every cycle with small variations. Giving them a one-tap starting point instead of a blank catalog cut order-placement time from a phone call of several minutes to under a minute of tapping.
The fear that turned out to be unfounded
The biggest internal objection before launch was pricing transparency. Sales management worried that once retailers could see prices online, side by side, they would start comparing notes with each other, or worse, use it as leverage to negotiate discounts they had no basis for.
It did not happen, for a reason that is obvious once you see it: retailers already knew their prices. They had been paying them for months or years. What the portal exposed was not new information to the buyer, it was new visibility to the seller, because now every price shown, every order placed, and every basket size sat in one system instead of scattered across 300 relationships living in reps' heads and notebooks.
If anything, transparency helped close gaps. When two retailers on similar volume tiers noticed a pricing inconsistency, it surfaced a manual pricing error the distributor didn't know it had, rather than a negotiation.
What changed in the numbers
Measured over the two quarters after rollout:
| Metric | Before | After |
|---|---|---|
| Rep time on order-taking | ~65% | ~20% |
| Average basket size | Baseline | +18% |
| Order errors (wrong SKU/qty) | Frequent, uncounted | Rare, tracked |
| New accounts opened per rep per month | 1-2 | 4-5 |
Average basket size going up 18 percent was not something we designed for directly. It came from retailers browsing the full catalog on their own time instead of rattling off a memorized list to a rep on a call. A retailer scrolling a catalog at 9pm notices a product they'd normally forget to ask about. That browsing behavior, the same one that makes consumer e-commerce work, applies to B2B buyers too. They are still people shopping, just for their shop instead of themselves.
Reps opening 4-5 new accounts a month instead of 1-2 is the number that actually mattered to the owner. That is the entire economic case for the portal in one line: the same headcount, doing the job it was hired for.
Rollout lessons that mattered more than the tech
- We didn't force adoption. Reps kept taking phone orders for retailers who wanted that. Within three months, most retailers had switched on their own because the portal was faster for them too. Mandates create resentment; a better option creates adoption.
- Reps got credit for portal orders in their territory. This killed the fear that self-service would make reps redundant. Reps became the onboarding force, training retailers to use the portal, because their commission structure rewarded it.
- We launched with the busiest retailers first, the same approach that worked when a printing business automated its quote-to-order flow: find the accounts feeling the most pain, prove it there, let the rest follow the social proof.
The takeaway
If your sales team spends most of its week re-entering orders it already knows are coming, you don't have a sales capacity problem, you have an automation gap dressed up as a headcount problem. A b2b ordering portal built around the reorder pattern, not a generic e-commerce clone, pays for itself in freed-up selling time within a couple of quarters. Map where your reps' hours actually go before you hire another one, and if the map points at repetitive order-taking, that's a conversation worth having.