PetSmart credits AI-powered decisioning with a 22% lift in salon bookings

How PetSmart's AI Decisioning Lifted Salon Bookings

PetSmart says AI-powered marketing decisioning helped drive a 22% lift in incremental bookings at its in-store grooming salons, according to Bradley Breuer, the company's SVP of Marketing, in an interview with Retail TouchPoints.

The system, built on Databricks for the first-party customer data foundation and Hightouch to activate that intelligence across CRM programs, continuously tests thousands of combinations of subject lines, offers, creative and timing to find the next best message for each customer. The immediate goal was straightforward: identify dog parents most likely to book grooming and reach them with an offer they would actually act on.

Breuer stressed that humans are not out of the loop. PetSmart's marketing team sets the objectives, guardrails, approved offers and brand guidelines, while AI evaluates customer signals in real time and recommends the next best action. The approach rests on two company assets — more than 80 million Treats Rewards members and grooming salons in nearly 1,700 stores — that give the predictive models meaningful data to work with.

The disclosure sits awkwardly with PetSmart's recent consumer campaign poking fun at AI hype, but Breuer frames the technology internally as a decisioning tool that makes interactions more relevant, timely and helpful rather than replacing strategy or creativity. The company says results exceeded expectations, though it has not disclosed the detailed methodology behind the 22% figure.

Reading PetSmart's 22% Salon Bookings Lift — and Its Limits

Interpreting PetSmart's 22% Salon Bookings Lift

The headline number is company-reported and shared via a trade-press interview, so it should be treated as directional rather than audited. 'Incremental' bookings typically means lift attributable to the campaign above a baseline, but the baseline period, test window and statistical approach are not disclosed. Even so, a percentage gain applied across roughly 1,700 salon locations represents a meaningful revenue effect if the measurement is sound.

Why First-Party Data Is the Real Foundation

The most durable part of this story is not the AI itself but the data feeding it. With more than 80 million Treats Rewards members, purchase history, pet types and life-stage signals, PetSmart can combine known behavior with predictive models in ways that smaller retailers cannot easily replicate. That same data also lets the company extend personalization across owned channels such as email, push notifications and SMS, which reduces reliance on increasingly expensive paid media.

The Human-AI Division of Labor

Breuer's description is a practical model for retail marketing: humans define the objective, set the guardrails and supply the creative, while AI handles the combinatorial testing of offers, timing and content that no person can evaluate manually. This is a narrower and more credible use of AI than fully autonomous marketing. It also frees marketers to spend more time on strategy and on analyzing why certain approaches outperform others.

What This Signals for Retail Marketing

PetSmart's account is one data point in a broader industry shift from calendar-based campaign planning to always-on decisioning. Personalization has been standard in paid media for years; the newer development is applying the same logic to owned channels where retailers control cost and data. PetSmart's result does not prove the approach works everywhere, but it does show how a retailer with strong first-party data can make the transition concrete.

What Retail Marketers Can Take From PetSmart's AI Stack

Retail marketers evaluating similar systems can draw several practical lessons from PetSmart's account:

• Build on a first-party data asset first: PetSmart's program depends on more than 80 million Treats Rewards members with purchase history and pet life-stage data; without that foundation, AI recommendations will be far weaker.
• Define tight guardrails before launch: PetSmart's team set approved offers, creative and brand guidelines before the engine started testing combinations; specify the same constraints so the system explores within brand limits.
• Pick a narrow, measurable objective: PetSmart directed the engine at one outcome — dog parents booking grooming appointments — and measured incremental bookings; broad objectives make lift difficult to isolate.
• Extend personalization to owned channels: the company says AI-driven decisioning now reaches email, push and SMS, transferring personalization that was already common in paid media to cheaper owned channels your business controls.
• Ask for methodology before trusting a headline result: the 22% figure is self-reported, so any comparable vendor pitch should include the baseline period, test duration and how incrementality was calculated.