eCommerce support share graphic with order and return workflow nodes.

Page guide

On this page

Most retail complaints are not support failures

They are fulfilment failures that support has to absorb. A parcel that did not arrive, an item that shipped wrong, a stock level that was inaccurate at checkout, a refund that has not appeared - none of those originate in the support queue, and all of them land there.

That shapes what an eCommerce support team actually needs to be. Answering the contact politely is the easy part. Resolving it means acting on the order: chasing the carrier, arranging the replacement, correcting the record, issuing the credit. A team that can only apologise generates a second contact.

RHI provides dedicated eCommerce operations teams working in your commerce platform and your help desk, handling the whole order lifecycle rather than only the conversation about it.

What the team handles across the lifecycle

  • Pre-purchase - availability, sizing, compatibility, delivery timescales, the question actually blocking the basket.
  • Order administration - placing, amending, cancelling, address corrections, and the payment problems that stall an order.
  • Fulfilment coordination - chasing warehouse and carrier, tracing missing parcels, arranging redelivery and replacements.
  • Delivery exceptions - the largest single contact driver in most retail operations, and the one most worth resourcing properly.
  • Returns and refunds - see returns and refunds support for how the reverse journey and refund authority work.
  • Product and catalogue data - correcting the listing errors that caused the contact in the first place.
  • Marketplace operations - see marketplace support, where platform performance metrics apply.
  • Fraud and verification checks - following your rules, flagging rather than deciding.

Peak trading is the whole design problem

Retail volume is not seasonal in a gentle sense. It is a multiple. A promotion, a sale period or a single successful campaign can put several times normal volume into your queue within hours, and the contacts arrive alongside the fulfilment problems the same volume created downstream.

Sizing a permanent team for peak wastes money for ten months. Sizing for the average makes peak visibly bad, exactly when new customers are forming their first impression of you. The workable answer is a core dedicated team plus a planned surge arrangement with a defined ramp - see after-hours and overflow support.

Surge staffing needs lead time for recruitment and training, so it is planned months ahead rather than requested in the week. Clients who tell us their trading calendar early get a trained surge team; clients who call in week one of the sale get people learning during it.

Authority over the order

This is what separates an eCommerce team that resolves from one that escalates. We agree explicitly what the agent may do without approval: cancel, amend, reship, issue a credit up to a threshold, waive a delivery charge, apply a goodwill gesture.

Set that too tight and every delivery failure becomes a two-contact case with a wait in the middle. Set it too loose and you have unmanaged margin exposure. We agree it per contact type, write it into training, and score both directions - escalating something the agent was empowered to resolve loses a point in the same way as overstepping.

The upstream feedback loop

Because most retail contacts originate upstream, the most valuable output of a support operation is telling you where. A carrier whose failure rate is concentrated in one region, a product whose return rate is driven by a misleading image, a size guide that is wrong for one range, a checkout step that fails on one browser.

We categorise contact reasons to a level that makes those patterns visible, and report them as findings with an owner rather than as volume. Fixing a listing removes a contact type permanently; handling it faster does not. See how root-cause reporting works.

Doing this properly requires the reason taxonomy to be finer than most retail operations run. "Delivery issue" is not a reason - it conflates a late parcel, a lost parcel, a wrong address, a failed delivery attempt and a carrier that marked something delivered when it was not, and each of those has a different owner and a different fix. We agree the taxonomy during scoping and revisit it, because an emerging problem hides inside a coarse category until it is large enough to show up as a spike.

Channels, systems and coverage

Retail customers use every channel and expect continuity across them - chat for pre-purchase, email for cases, phone when something has gone wrong, WhatsApp and social increasingly for everything. See omnichannel support.

Agents work in your commerce platform and help desk. Our teams are experienced across Shopify, WooCommerce, Adobe Commerce and Amazon Seller Central, alongside Zendesk, Freshdesk, Gorgias and Zoho Desk - platforms we are competent in, not partnerships. See technology.

Coverage is built from shift units. Retail contact skews to evenings and weekends because that is when people shop, so a weekday-office roster misses a large share of it.

Quality and reporting

Reporting covers contact rate per order - the single most useful retail support metric, because it normalises volume against trading - plus first contact resolution, resolution time by contact reason, refund and replacement turnaround, and satisfaction. Contact reasons are reported as a ranked list with movement, so an emerging problem is visible before it becomes a spike.

Interactions are sampled and scored against a scorecard agreed with you, with order-authority criteria scored pass-or-fail. See quality assurance. Figures published elsewhere on this site are historical or representative and campaign-dependent.

Getting started, and what it costs

Priced per dedicated agent per month with a five-agent minimum, with surge cover planned and priced separately - published rates and what moves them.

Bring contact volume alongside order volume so we can calculate contact rate per order, your ranked contact reasons, and your trading calendar. Related: eCommerce customer service, marketplace support. Talk to us.

Frequently asked questions

Because they originate upstream - a parcel that did not arrive, an item that shipped wrong, a stock level that was inaccurate at checkout. Support absorbs them. That is why an agent needs authority to act on the order rather than only to apologise; a team that can only apologise generates a second contact.

A core dedicated team plus planned surge cover with a defined ramp. Sizing permanently for peak wastes money for ten months; sizing for the average makes peak visibly bad exactly when new customers are forming their first impression. Surge needs months of lead time for recruitment and training, not weeks.

Whatever we agree per contact type - typically cancel, amend, reship, issue a credit up to a threshold, waive a delivery charge. Too tight and every delivery failure becomes a two-contact case; too loose and you have unmanaged margin exposure. The scorecard penalises both overstepping and escalating what the agent was empowered to resolve.

Contact rate per order. It normalises volume against trading, so it tells you whether service is improving or you simply sold less.

Yes, and it is the most valuable output. We categorise contact reasons finely enough to make patterns visible - a carrier failing in one region, a product whose returns are driven by a misleading image - and report them as findings with an owner. Fixing a listing removes a contact type permanently.

Shopify, WooCommerce, Adobe Commerce and Amazon Seller Central, alongside Zendesk, Freshdesk, Gorgias and Zoho Desk. Those are platforms our teams are competent in, not partnerships or certifications.