How Médident prioritised its AI use cases across 22 dental centres
Fifteen days of operational and technical audit, run in the field alongside the teams. A patient journey mapped end to end, and use cases prioritised by volume and ROI rather than by marginal gain.
What a dental care network faces at this scale
Dental care networks face a common set of constraints. A volume of patient calls that saturates at peak hours, in a business where the first contact determines whether an appointment happens. Practices that build up centre by centre, shaped by local hiring and habits, with no shared foundation to consolidate them. And reporting that often rests on decision-support tools whose future is not guaranteed. Médident faced all three across its 22 centres. The question was therefore not to find AI ideas, there are always plenty, but to know where to apply them first and for what return.
centres with divergent practices
Processes built site by site, with no shared foundation to consolidate them.
of patient calls to absorb
Inbound volume that saturates at peak hours, in a business where the first contact determines the appointment.
for the decision-support tool
Visibility on quotes and insurance coverage was about to disappear, with no replacement in place.
Médident, FranceMédident
Médident is a network of 22 dental centres in France, mostly in the Paris region, active for around fifteen years.
The group employs between 500 and 600 people, a large share of them in front office roles across the centres. At that scale, every point of friction in the patient journey is paid for in volume: what happens on one call happens 22 times over.
Go and look in the field before costing anything
The audit did not produce a catalogue of use cases. It started from direct observation of the real work, then kept only what could be costed, deliberately setting aside marginal gains.
On-site immersion in the centres
Tandem went out to work alongside the teams and watch how they actually operate: how a day unfolds, the friction points, the bottlenecks, what gets lost between two tools. Nothing was inferred from a questionnaire.
Mapping the patient journey
From the first call to invoicing, identifying the breakpoints that cost appointments and quotes.
Checking against the existing systems
Conversations with management, field teams, the quote management software vendor and the telephony provider, to compare declared processes against what was actually observed.
Prioritisation by volume and ROI
The selected cases cover activating features already present in existing tools but never switched on (SMS reminders, electronic signature), inbound and outbound voice agents to recover unanswered calls and automate appointment reminders, a conversational agent on the group's websites, and an internal agent querying quote and coverage data in natural language. Each is modelled as conservative, median and optimistic.
Costed use cases, and one immediate win
Several use cases identified and costed across three scenarios, each with an estimated implementation cost. Prioritisation was driven by volume and expected return, not by ease of implementation. The audit also identified an immediate win requiring no development: features already available in the group's tools but not activated, usable straight away. Going into the field proved decisive: several of the costliest friction points appeared in no documented process, they were only visible by watching a typical day inside a centre.
Want us to map your requests?
Free audit of your processes, we identify high-ROI use cases together, and quantify the gains before deploying anything.
More case studies
All case studiesSantiane
↗Ops
Part of the support desk automated, without health data leaving the perimeter
Moose Knuckles
↗Audit
Three AI opportunities, quantified on the brand's own data
Doctolib
↗Adoption
From a handful of pioneers to 90 % building their own agents in three months
InfraVia Capital Partners
↗Internal GPT
How InfraVia equipped 100+ staff with AI using Copilot and Perplexity Enterprise