In Part 1, we focused on ZOE’s pivot to gut health. ZOE may be making the right strategic move by focusing on a specific benefit area, simplifying the test experience, and using Daily30+, a plant-based gut health supplement, as an accessible brand entry point. This gives ZOE a clearer consumer need, a lower-friction product, and a pathway into app engagement and testing.
However, on a retail shelf or Amazon page, Daily30+ will compete with many other digestive health products and supplements with similar (and sometimes bolder) claims.
Therefore, the strongest version of ZOE is probably not as a supplement company. It is a science-led gut-health system that helps consumers understand their biology, improve their diet, choose better products, and track progress over time. So, the question becomes, how might ZOE best use their science to sustain a point of difference?
Is ZOE's science enough?
The answer is likely yes at the brand level, but less clear for ZOE 2.0 and Daily30+ as specific offerings.
ZOE's science is a unique asset. Most supplement brands cannot point to the same depth of academic collaboration, microbiome data, and nutrition science.
But ZOE's science should be separated into three buckets:
- Foundational Platform Science: PREDICT, microbiome research, nutrition science, academic collaborations, data infrastructure, and the broader personalized nutrition model.
- Product-Specific Science: Does Daily30+ itself has enough evidence to support a durable marketplace point of difference?
- Model-Transition Science: Can AI replace some deep phenotyping in ZOE 2.0s without weakening recommendations or outcomes?
The PREDICT studies support ZOE's brand credibility, but Daily30+ needs its own evidence.
ZOE has initiated direct product research. The Daily30+ evidence page highlights a randomized control trial on the original formulation and provides links to an abstract and pre-print paper (ZOE Daily30+ evidence), but the paper has not (yet) been peer-reviewed.
There is another challenge with product-based research. Product-specific claims will be constrained by regulations. Thus, clinical outcomes may sound like very familiar claims (good source of fiber, supporting gut health, supports energy, etc.). If Daily30+ competes mainly on those claims, ZOE risks getting lost in the noise.
Daily30+ can still play a valuable role. It can become an accessible, evidence-based product for gut-health consumers that introduces them to the ZOE brand. It can support recurring revenue, and move some users into the app and testing to support engagement and additional value. For that to work, a Daily30+ buyer should understand why the science behind Daily30+ is unique and how the app and gut health test can result in better outcomes for them.
Can AI replace deep phenotyping?
ZOE appears to be moving from a more intensive phenotyping model toward a more streamlined model.
Its newer public FAQ says the test kit has been simplified and now includes a stool sample, health questions, and optional blood test inputs. ZOE states that, based on more than 300,000 microbiome profiles, its proprietary predictive algorithm can predict blood fat and blood glucose response without additional testing, meaning users no longer need test cookies, a blood sample, or a CGM (ZOE FAQ).
Removing CGMs, blood sampling, standardized test meals, and complex test logistics likely reduces cost, improves scalability, and lowers consumer friction.
But there is a tradeoff. AI prediction is only as valuable as the evidence showing that it can replace measured phenotyping with enough accuracy to support meaningful consumer outcomes.
ZOE's original research supports the original business proposition and program, but how well it translates to ZOE 2.0 is less clear. To maintain a similar scientific credibility, ZOE will need to demonstrate and publish findings of their new model.
Deep phenotyping still has a role.
The move away from deeper phenotyping may make sense for scale, but deep phenotyping may still be needed in several places:
- Validation: proving that AI predictions remain accurate as the consumer base expands.
- Premium personalization: serving consumers who want more detailed biological insight, especially those with stronger health motivations or complex needs.
- Product development: helping food, beverage, supplement, and health-tech partners design targeted interventions.
- Clinical credibility: supporting higher-value health use cases, practitioner confidence, and outcomes research.
- Model improvement: feeding AI models with high-quality measured data over time.
The better model may not be AI instead of deep phenotyping. It may be AI plus selective deep phenotyping. ZOE could use AI and app-based engagement for the broad market, while reserving deeper testing for validation cohorts, premium tiers, high-need consumers, research, and partner product development.
An alternate model: tiered testing plus partnerships.
ZOE may already be moving toward a tiered model, but the case study still points to an important strategic opportunity: the tiers must be explicit, affordable, and easy for consumers to navigate.
Tier 1 | Gut health starter: A low-cost digital assessment, food diversity score, symptom and habit screener, and simple gut-health recommendations. Daily30+ can be offered here as an optional first step.
Tier 2 | App-guided behavior change: AI food tracking, meal scoring, plant diversity tracking, mindful eating tools, and habit feedback.
Tier 3 | Microbiome test light: A lower-cost stool test focused on gut diversity, fiber intake, plant variety, and practical food and supplement recommendations.
Tier 4 | Full ZOE personalization: A premium program that combines microbiome testing, metabolic response insights, app-based guidance, and deeper personalization.
Tier 5 | Partner ecosystem: A network of food, beverage, supplement, retail, and health-tech partners using ZOE's science and data to develop targeted products and personalized solutions.
This architecture would preserve ZOE's core point of difference: people are different, so solutions should adapt.
Why partnerships may be the bigger opportunity.
ZOE does not need to manufacture every solution. Its most defensible asset may be the intelligence layer: the data, science, algorithms, behavioral insights, and consumer trust that help match people to better interventions.
A partnership-led ecosystem could be more powerful, scalable and profitable than a single supplement brand. For example:
- A supplement company could use ZOE insights to design targeted prebiotic, fiber, probiotic, or synbiotic products.
- A food company could develop microbiome-supportive meals or snacks aligned with ZOE's dietary recommendations.
- A beverage company could create gut-health products for specific occasions or consumer segments.
- A retailer could use ZOE's framework to guide shoppers through personalized gut-health journeys.
- A health-tech company could integrate ZOE-style nutrition feedback into metabolic, digestive, or longevity platforms.
This model would allow ZOE to scale its science without forcing the company to fight every SKU-level battle in the supplement aisle. It would also create a broader market for personalized nutrition innovation, where ZOE acts less like a product company and more like an ecosystem builder.
Did ZOE need to change direction?
ZOE did not make a mistake by investing in science. Their mistake may have been building the scientifically complete solution before proving the simplest scalable behavior and business model.
In personalized nutrition, companies often overbuild the data layer before validating the value layer. They assume that more biomarkers, more algorithms, and more precise recommendations will automatically create engagement. But consumers aren't buying into personalization for the most testing, they are looking for the best outcomes.
Nlumn has made a similar point in prior work: microbiome testing companies face persistent challenges around affordability, actionability, and behavior integration. Consumers may be interested in personalized microbiome insights, but cost, turnaround time, unpleasant sample collection, and unclear value can limit adoption (Nlumn microbiome newsletter).
ZOE's shift suggests the company may have learned that the science was strong, but the path to broad consumer adoption needed to be simpler, faster, more affordable, and more effective to deliver value equal to user investment.
What this means for personalized nutrition innovators.
1. Do not let the business model flatten the science. The goal is not to choose between personalization and scale. The goal is to design products that are accessible enough to scale and personalized enough to matter.
2. Build a ladder, not a leap. Consumers need steps: education, self-assessment, simple product trial, habit tracking, optional testing, personalized recommendations, targeted products, and measurable feedback.
3. Use supplements as interventions, not the whole strategy. Supplements can be valuable tools in personalized nutrition, but they should stay connected to the data, guidance, and outcomes that justify personalization. For supplement (and food) companies, the opportunity is to build products that fit into personalized journeys.
4. Partner to build ecosystems. Consumers eat across categories, shop across channels, use multiple apps, and respond to a mix of health, taste, convenience, price, culture, and habit.
5. Create solutions to solve problems. Companies need to show not only that a product is science-backed, but that it helps the right consumer take the right action and achieve a meaningful outcome.
Nlumn's Perspective
ZOE's science gave them a head start. Whether they win in the next phase depends on whether that science remains connected to a consumer experience that is meaningful.
Science can open the door, but the consumer experience has to carry people through it. The winners will make personalization easier to start, easier to sustain, and easier to connect to measurable progress.