Sacher AI logo
Book a call

GLP-1 and obesity care

GLP-1 medicines work in trials. Real-world results depend on behaviour.

In a study of more than 125,000 US adults with overweight or obesity, just over half (54%) stopped within a year. Many of the reasons are behavioural: how people cope with side effects, unrealistic expectations, waning motivation, and the everyday barriers that make lasting change hard. That is the part we work on.

Source: Rodriguez PJ et al., JAMA Network Open, 2025.

Book a discovery callHow we work with you

Who we work with

Two kinds of team, three ways to start.

GLP-1 and obesity care providers

Telehealth platforms, clinics, pharmacies and weight management programmes with retention, support or patient-facing AI problems.

  • Design and deliver →

    From idea to a tested prototype or MVP: technical planning, product, design, behavioural logic and conversation design, alongside your own team. The production build stays with you.

  • Test and prove →

    PromptSafe testing, the FAST evaluation framework, independent evaluation, and research through to peer-reviewed publication.

Organisations building around GLP-1 treatment

Life sciences, medtech and digital health teams deciding what to invest in.

  • Decide →

    The Before You Build Method, delivered as a fixed fee Evidence Phase. Eight to twelve weeks, ending in a recommendation to go ahead, rework or stop.

Across the whole journey

Support that adapts from the first dose to long-term maintenance.

  1. 01

    Starting treatment

    Onboarding, expectations, and the first doses, when confidence is lowest.

  2. 02

    The first weeks

    Side effects and the questions they raise, handled early and calmly.

  3. 03

    Pauses and dips

    Missed doses, and the weeks when life gets in the way.

  4. 04

    Reaching a goal

    Recognising progress without letting the habits behind it slip.

  5. 05

    Long-term maintenance

    Whether someone stays on a maintenance dose, steps down, or stops.

Behavioural phenotyping

Support designed around how people actually think, feel, and act.

Two people on the same medication can need completely different support. One sets impossibly high standards and gives up after a single slip. Another eats to manage stress. Another is pulled off course by the people around them, or by a schedule with no room for self-care.

Behavioural phenotyping identifies the psychological and behavioural drivers behind each person's eating, activity, and adherence, and turns them into tailored, evidence-based support, delivered in the tone, framing, and intensity that fit the individual.

Perfectionism and all-or-nothing thinking

Cognitive restructuring and self-compassion, so one slip does not end the whole effort.

Emotional eating

Trigger and mood work, and mindful-eating prompts that separate physical hunger from emotional hunger.

Social and environmental pressure

Boundary-setting, assertiveness, and role-play for the real social situations that derail people.

A pull towards quick rewards over slow progress

Graded goals and immediate, non-food rewards that make steady progress feel worthwhile.

Unrealistic expectations

Honest expectation-setting and psychoeducation on realistic timelines, so disappointment does not drive early drop-off.

Time and life pressures

Quick-win actions, time-boxing, and stress management for busy and disrupted weeks.

Low confidence and self-efficacy

Curated psychoeducation, clear next steps, and affirmation that builds belief in change.

Scepticism about behaviour change

Motivational interviewing and credible, specific evidence that meets doubt with respect.

Fear of failure

Non-scale victories and graded exposure that lower the stakes of trying.

Where the line sits

Behavioural support, with clinical decisions left to clinicians.

  • Support, not clinical decisions

    Our behavioural AI supports people between appointments. Clinical decisions, such as whether to stop, pause or switch treatment, stay with the person's own clinician. We design that boundary in from the start.

Case studies

GLP-1 and obesity care work.

See all case studies →

Decide

Twenty AI ideas, narrowed to the six worth doing

An international GLP-1 care provider was facing heavy support demand, fragmented systems, and early churn driven by side effects, uncertainty, and inconsistent support.

Read the case study →

Design and deliver

Personalised AI health coaching deployed across obesity care

For Allurion, our team designed the behavioural architecture of personalised AI health coaches for obesity and GLP-1 care, deployed internationally and serving very large patient populations.

Read the case study →

Design and deliver

Safe, personalised patient-facing AI for people on weight-loss medication

Embedded within Numan's team from the very beginning, Sacher AI helped create an AI Health Assistant for patients on anti-obesity medications, alongside Aegis, a monitoring, safety, and escalation system.

Read the case study →

GLP-1 Digest podcast, episode 8

Can AI stop GLP-1 patients from churning?

Dr Paul Sacher with Dr Ashwin Sharma on how direct-to-consumer platforms keep patients without lowering prices.

Open the episode on GLP-1 Digest ↗

Next step

Working on GLP-1 or obesity care?

Book a discovery call. We will work out whether there is a fit and what a sensible first step looks like.

Book a discovery call