Rollout Framework

How to roll out customer self-service without disrupting sales reps

Self-service projects rarely fail on technology. They fail when the sales team concludes the portal costs them money, because a rep who loses commission when their account orders online will stop recommending it, and they'll do it quietly rather than in a meeting. Fix the compensation before you fix the user interface, and the same reps become the fastest adoption channel you have.

There's a version of this project that goes badly and it's remarkably consistent. The portal launches, adoption is thin, leadership blames the user experience, and a second round of design work changes nothing. Meanwhile the reps have done the arithmetic. If an order placed online doesn't show up in their number, the portal is a pay cut, and they'll route their accounts around it without ever saying so in a meeting.

Sales teams aren't the obstacle here. Compensation design is. Get that right and reps become the fastest adoption channel you have, because they're already talking to every account you want to move.

Fix compensation first

This is the whole game. McKinsey's 2025 guidance on B2B growth is explicit that top performers compensate reps for sales across all channels, and some go further by making digital penetration a metric in the incentive plan itself. The language that stuck from McKinsey's earlier work is double-counting credit: when a customer interacts both digitally and with a rep, credit both.

A named example helps more than a principle. Adrienne Hartman, EVP of Marketing at J.J. Keller, described their approach on Master B2B: for an assigned account, the sales rep gets credit no matter where the order is placed. She calls the philosophy channel agnostic, and it covers web, e-procurement, and punchout orders. Her internal pitch to the sales team was that you can make sales while you're on vacation.

The cost of getting this wrong is well documented. Modern Distribution Management's account of rep resistance names the top objection plainly, that they won't get credit for sales processed through the system, and warns that if the perception takes hold that people aren't paid for digital orders it can leave you dead in the water. It only takes a few instances for the belief to stick, and it includes an example of a single top rep who found one order error, used it as proof, and blocked adoption in their territory outright.

Channel conflict is measurable at the strategy level too. In McKinsey research covering more than 3,800 decision-makers, 38% named channel conflict as the biggest reason they avoid selling online. Treating it as a change-management problem rather than a technology problem is what separates the manufacturers who get adoption.

Start where buyers least want a human

Sequencing should follow buyer preference, and there's research that segments it precisely. McKinsey found that only 15% of B2B buyers want to speak to a salesperson when repurchasing the exact same product, against about half for a repeat with new specifications, and 76% who find a salesperson helpful when researching something new. That study is from 2017 and should be dated honestly, but it remains the only named research that splits channel preference by purchase type, and it points somewhere useful.

So begin with exact repeats. Reorder from purchase history, order status, tracking, and invoice access. Nobody is defending the phone call that asks where a shipment is, buyers don't want it, and reps don't enjoy taking it. That's the work with the least political friction and the fastest visible payoff.

Leave configure-to-order quoting, contract renegotiation, and anything requiring judgment for later, and expect reps to stay in those loops. Sequencing this way also matches where the data is cleanest, so the first release is less likely to show a customer something wrong.

Give reps something, not just something to lose

The strongest adoption lever is making the portal a tool reps use rather than a channel that bypasses them. Order-on-behalf, where a rep signs in as the customer, sees their catalogue and pricing, and places or edits the order with them, changes the politics entirely. Modern Distribution Management's account describes exactly this: a special sales login let reps experience the site as the customer, and reps appreciated it.

This is well supported in packaged platforms, which is worth checking before you choose one. Adobe Commerce has Login as Customer, gated on the customer's consent. BigCommerce B2B Edition has masquerade for assigned companies. Sana has sales agents who can act for an account, writing back to the ERP. Shopify handles it through draft orders in the admin rather than storefront impersonation, so front-end impersonation needs an app.

1. Rep dashboards – Give reps a view of their accounts' portal activity: open quotes, abandoned carts, unusual order gaps. That turns the portal into a lead source for them.
2. Order-on-behalf – Let reps place and edit orders as the customer, so a phone call still ends in a clean digital order rather than a re-keyed one.
3. Onboarding credit – Recognize reps for getting accounts onto the portal, not only for orders. The behaviour you want early is enrolment.

Pilot small, then open it

Don't launch to the whole customer base. SAP's Susanne Adam describes distributors who test-market their online store with a small customer group, gather feedback, adjust, and only then open it up. Jason Hein of Acumental B2B frames the same discipline more bluntly: the magic of digital is that you can try it for 5% of your traffic, and roll it back if it doesn't work.

Choose the pilot group with the rep, not around them. Accounts that order predictably, have clean pricing, and phone in often are ideal, because the portal will visibly save everyone time. Let those reps report the result to their peers, which is considerably more persuasive than a leadership announcement.

Expect the migration itself to be the hard part rather than the launch. Master B2B's 2026 research found 56% of organizations cite moving the business from offline-centric to digital-centric as their top challenge, for the second consecutive year.

The evidence that digital deepens relationships

When reps push back, the most useful counter-argument is public company disclosure rather than vendor marketing. Watsco told investors that customers who become active users of its technology platforms produce higher growth rates and exhibit approximately 50% less attrition. In its FY2025 results the company reported roughly 73,000 authenticated digital users and ecommerce at about 35% of total sales, with the attrition improvement consistent year over year.

Fastenal's leadership makes the same argument in plainer language, saying digital integration makes them stickier with customers and more indispensable. Neither company reduced its sales force to get there. Digital adoption raised switching costs, which is the opposite of cannibalization.

Once the portal is live, measure outcomes rather than logins: the share of routine orders arriving untouched, quote turnaround, and how much re-keying has left your customer service team. What a portal needs to include covers what to build, and pre-built versus custom covers the platform decision.

Frequently Asked Questions

Next Step

Get the foundation right before you build.

For readers scoping a platform decision or wanting a full architecture recommendation.