AI copilots in sales apps: which features do reps actually keep using?

KEY TAKEAWAYS Summary by the editors

1. Published research on long-term copilot use by fashion wholesale reps is scarce; the best available evidence comes from cross-industry sales surveys by Gartner, Salesforce and BCG.
2. Gartner's survey of 210 sales leaders, published in May 2026, found that AI tools save sellers an average of 4.8 hours per week, but 72 percent of sales organisations report low reinvestment of that time into high-value activities.
3. Features that remove administrative work, such as meeting summaries, order recaps and account briefings, tend to be used more consistently than open-ended chat assistants.
4. Gartner predicts that by 2028 AI agents will outnumber sellers ten to one, yet fewer than 40 percent of sellers will report that agents improved their productivity.
5. BCG found that employees with at least five hours of AI training are much more likely to become regular users, and frontline regular use had stalled at 51 percent in 2025.

Six months after launch, wholesale reps tend to keep using the AI copilot features that save them administrative time on every account, such as account briefings, meeting summaries and order recaps, while open-ended chat and generic suggestions fade. There is little fashion-specific research on this yet, so brands should measure use feature by feature rather than rely on vendor adoption claims or launch enthusiasm.

What is an AI copilot in a wholesale sales app?

A copilot is an AI assistant built into the app reps already use to present collections and write orders. Typical features include a pre-meeting summary of the account, suggestions for styles to add, size curve proposals, answers to product questions, transcription of appointments and drafted follow-up emails. Some vendors now describe these features as agents that can act on their own, for example preparing a draft re-order.

What does the research say about sales copilot use?

Fashion-specific evidence is thin, so the most useful data comes from broader B2B sales studies. They point in a consistent direction: AI saves time, but the gains are uneven and depend on data, training and how the saved time is used.

Read also

How can AI help fashion sales reps prepare, sell and follow up?

Which copilot features do reps keep using?

Based on the pattern in this research and on how wholesale selling works, the features that last are those tied to a recurring task that reps dislike and that the AI does reliably with data the brand already has.

• Account briefing before an appointment: last orders, open deliveries, returns and notes in one view. It saves preparation time on every visit.
• Appointment summary and order recap: a draft email to the buyer with what was agreed, checked by the rep before sending.
• Gap suggestions: styles or categories the account usually buys but has not yet ordered this season, with the reason shown.
• Product answers: fabric composition, care, delivery dates and minimums, drawn from PIM and ERP rather than generated from general knowledge.

What these features have in common is that the rep can check the output in seconds. An account briefing can be compared with what the rep already knows; a drafted recap can be corrected before it is sent. Features whose output is hard to verify, such as a predicted order value for the season, need more trust and therefore more evidence before reps rely on them in front of a buyer.

Which features tend to fade?

Open-ended chat assistants are often tried at launch and then used less, because reps do not know what to ask and answers are not specific enough to an account. Recommendations without an explanation also lose trust quickly: if a rep presents a suggested style and the buyer asks why, the rep needs an answer. Gartner's buyer survey supports this. Buyers rated sales reps well ahead of generative AI for understanding their needs and instilling purchase confidence, so a copilot that tries to replace the rep's judgement adds little.

Why do copilot results differ so much between brands?

1. Data quality: Salesforce found 51 percent of sales leaders say disconnected systems hamper AI initiatives. A copilot without sell-in, sell-through and stock data can only produce generic advice.
2. Training: BCG found that five or more hours of training markedly increases regular use, and that only 36 percent of employees felt adequately trained.
3. Workflow fit: features inside the order screen get used; features in a separate tab often do not.
4. Management use of saved time: Gartner's finding that 72 percent of organisations do not reinvest saved time well means the hours may disappear into more administration.

How should a brand evaluate its copilot after six months?

Measure use per feature, not overall logins: the share of appointments with a briefing opened, the share of order recaps sent from the drafted text, and the acceptance rate of style suggestions. Ask reps which features they would miss if removed. Then link use to outcomes such as time from appointment to confirmed order and styles per order. Gartner advises measuring capacity and effectiveness, not only time saved.

Six months is also a good point to retire features. Removing a feature that few reps use reduces clutter on the screen, lowers support effort and makes the remaining features easier to find. Gartner's 2026 research on AI-enabled next best actions found that organisations providing them were 2.6 times more likely to achieve commercial growth, which suggests that focused, actionable guidance matters more than the number of AI features on offer.

Read also

How to run remote buying sessions with key accounts in a digital showroom

What should reps and buyers be told about AI?

Reps should know which data each feature uses and that they remain responsible for what they send to customers. Buyers should be told if appointments are transcribed or if suggestions are AI-generated. Transparent use protects the relationship, which, as Gartner's buyer research shows, is still where buyers find the confidence to commit.

Frequently asked questions

Do sales reps actually use AI copilots?

Many do, but use is uneven. Salesforce found 87 percent of sales organisations use some AI, while BCG found frontline regular use had stalled at around half of employees in 2025. Use tends to concentrate on features that save administrative time.

How much time does an AI copilot save a sales rep?

Gartner's survey of sales leaders found an average of 4.8 hours per week. That figure is cross-industry and self-reported by leaders, so brands should measure their own savings per feature.

Which AI features are most useful for wholesale reps?

Account briefings before appointments, appointment summaries and order recaps, gap suggestions with reasons, and accurate product answers from PIM and ERP. These support recurring tasks and rely on data the brand already holds.

Why do AI copilot projects in sales disappoint?

Common reasons are disconnected data, too little training, features outside the reps' workflow and no plan for the time saved. Gartner predicts fewer than 40 percent of sellers will say AI agents improved their productivity by 2028.

GuideThe complete guide to AI in fashion wholesale and B2BRead the complete guide