Three-Thirds Demand Model
Forecast demand by separating eager, observant, and certainty-seeking customers
- Difficulty
- Easy
- Time to result
- ~days to results
- Steps
- 6
- Confidence
- 94%
The model divides a recovering market into three behavioral groups. The first is an eager or “invincible” third that resumes activity immediately despite uncertainty. The second is a reserved third that observes early outcomes, safety practices, and social proof before returning. The final certain third waits for a definitive risk-reduction event, such as a vaccine or other strong guarantee. Businesses first estimate the timing, needs, and economic value of each group. They then serve early demand while deliberately generating visible evidence that reassures observers. Because irresponsible early customers can frighten the reserved group, customer conduct becomes part of the recovery strategy. Forecasts must also account for the possibility that the certainty-seeking segment contains older, wealthier customers whose delayed return disproportionately affects premium offerings.
Origin
Extracted from Masters in Business, where Jon Taffer used the model to forecast how restaurant and bar customers would return during the COVID-19 reopening period.
Core principles
- 01Customers return at different speeds based on perceived risk.
- 02Early customers create visible evidence for more cautious customers.
- 03Responsible behavior can increase demand by signaling safety.
- 04High-income customers may be concentrated in the slowest-returning segment.
How to run it
- 1
Define the uncertainty
Specify the perceived risk preventing normal customer behavior. Identify what is known, unknown, and publicly visible.
Pro tip Describe the risk from the customer’s perspective rather than the operator’s.
Watch out Operational readiness does not guarantee that customers feel safe.
- 2
Segment by certainty threshold
Separate customers into those ready now, those waiting for evidence, and those waiting for near certainty. Treat the thirds as a mental model rather than exact measured proportions.
Pro tip Use customer interviews or behavioral data to refine the approximate split.
Watch out Do not present one-third shares as precise forecasts without evidence.
- 3
Serve the eager group safely
Prepare an experience for early customers without allowing their risk tolerance to weaken safeguards. Their behavior will become evidence observed by others.
Pro tip Make compliance easy and establish expectations before arrival.
Watch out Visible crowding or unsafe behavior can delay the next segment’s return.
- 4
Signal safety to observers
Show the reserved group that procedures are followed and early participation has not produced adverse outcomes. Use visible spacing, protective measures, and credible updates.
Pro tip Design safeguards that can be recognized from outside the venue or in customer communications.
Watch out Claims of safety are weaker than observable behavior.
- 5
Plan for certainty seekers
Assume some customers will not return until a definitive external milestone occurs. Preserve relationships and offer alternatives where possible.
Pro tip Maintain communication without pressuring risk-sensitive customers.
Watch out Premium businesses may underestimate this segment’s share of revenue.
- 6
Update the forecast
Track movement between segments as evidence and conditions change. Recalculate capacity, cash requirements, and marketing timing accordingly.
Pro tip Monitor behavior rather than relying only on stated intentions.
Watch out A negative incident can reverse movement and push customers back into caution.
In the wild
A restaurant expects younger, risk-tolerant guests first, followed by customers who want to observe safety practices for several weeks. It enforces masks and spacing among early visitors, publishes clear procedures, and keeps delivery available for older customers waiting for stronger certainty.
→ The business serves immediate demand while improving the evidence needed to attract cautious customers later.
An event organizer identifies attendees willing to return immediately, those waiting for proof that new controls work, and those unwilling to attend until an independent audit is complete. It runs a small pilot, reports verified results, and retains a remote option.
→ Demand returns in phases instead of being treated as a single launch-day event.
Common mistakes
Treating thirds as exact statistics
The model is a practical segmentation heuristic; actual segment sizes must be validated for the specific market.
Ignoring customer-created signals
Early customers can undermine an operator’s precautions if visible behavior appears crowded, careless, or unsafe.
Valuing every segment equally
The last group may return slowly while representing an outsized share of discretionary spending or premium demand.
Is it for you?
Best for
It is best for businesses forecasting a phased return after a health, safety, trust, or market-confidence shock.
Not ideal for
It is not ideal when customers face no meaningful perceived risk or when reliable behavioral data supports more precise segmentation.
From the transcript
“I believe the third of the population, the younger third, the invincible third, they're gonna go out right away.”
“The second third I called the reserve third. They want to watch and see for a couple of weeks what's happening.”
“The third third I called a certain third. They're not going out to there's a vaccine and they can be absolutely certain.”
From the episode
Jon Taffer on Bars, Restaurants and Hotels on Lockdown (Podcast)
Jon Taffer