Context Graph for CPG
David: What does “context” mean for data platforms?
Ophir: Ever since we started tackling the problem of messy, fractured product data, we found that domain knowledge is the key to making good decisions.
Context in AI goes beyond recording state data and decision traces. It must include the basis of the decision. For specific verticals — like CPG — the reasoning is usually embedded in industry-specific domain knowledge.
David: What does this mean in real life?
Ophir: A key aspect of our data platform is that industry knowledge is embedded in our logic. But the more expertise that we added, it became harder to manage, especially for performing QA.
This problem led us to add human-readable “context” right alongside the data results. This approach was powerful and efficient. It enabled agents, analysts, and developers to trace the chain of decisions for audit or validation.
Context is embedded vertically in our stack. Throughout our schema and processes, we can observe the entire "sausage-making" process, with natural language traces that lead clearly to the final outcomes.
David: Can you give some examples of domain knowledge about CPG products?
Ada: Sure, for example: How are units of measure correctly expressed in food versus non-food products? We see unexpected mutations all the time, such as made-up or foreign acronyms, even from “sources of truth” such as brand syndication and major retailers.
John: Another example: FDA regulations can be used to validate Nutrition Facts data. The FDA has strict requirements, which are codified in the Federal Register and have the force of law, but are often not applied on product description pages.
David: We see all the time how FDA regulations about food labeling, as expressed in the FD&C Act, are often violated on retail websites.
How do retailers get away with it? Because unlike food brands, grocers are not subject to federal oversight of product description pages.
Ophir: These are good examples of how we use domain knowledge to make decisions about normalizing, fixing, and enriching attribute values. While we don’t know about travel and automobiles, we do know CPG.
David: How does recording context help?
Ophir: Context is necessary to describe why decisions were made. It must be easily observable, auditable, and stored as first-class data.
David: Aren’t some data sources always reliable, so context about decision-making isn’t needed?
Ophir: Among the 47 sources we currently ingest, all have significant gaps and inaccuracies.
Ada: Some data sources have excellent quality, but only cover a subset of brands. Others, like syndicators, don’t have private label products. And long-tail SKUs are specific to each source, or not present at all.
John: Yes, and key attributes are often missing, such as images or net weight.
David: Sounds like a mess. How do we deal with it?
Ophir: We expect data chaos. Gaps and errors are the problems we solve for.
David: Aren't LLMs a reliable source of product data?
Ophir: Looking at generic AI results, they do a great job at synthesizing and generating data. But general LLMs don’t have vertical knowledge to classify with accuracy, correct errors, derive missing values, or adjudicate conflicts.
Ada: The answers from LLMs are only as good as the information they gather. We often see inaccurate results from general AI search and tools. Garbage in, garbage out still applies in the age of AI.
David: Is the answer to throw an army of human experts at the problems?
John: Human experts do have the knowledge, but an army of them is expensive and hard to manage. However, their knowledge can be embedded in vertical AI to create quality data at scale.
David: When did we start using context?
Ophir: We architected context from the beginning. We just didn’t call it a “Context Graph” because that term didn’t exist. Handling context as first-class data enables our systems and people to do their jobs better. We embed our expertise to improve quality and frequency, then capture how that expertise was used to create those better outcomes.
Ada: A simple example is knowing that cocktail mixers like tonic are non-alcoholic, even if retailers or brands list them in an alcohol category.
John: LLMs often get this classification wrong too. Humans like Ada never do 🙂.
David: These CPG nuances matter in lots of use cases, such as PDPs, digital analytics, nutrition and even price optimization.
Ophir: As our contexts grow, our learning compounds. The more we add to the graph, the more we are able to do better. Compare this context approach with hard-coded rules for addressing endless edge cases. That becomes unwieldy and doesn’t scale.
David: When does the Context Graph come into play?
Ophir: In our view, a Context Graph is systems of agents that enable autonomous decision-making across workflows. This full AI autonomy — when AI makes decisions across workflows — has not yet arrived. We see the benefits and the paths to get there, and we are working towards them.
David: Where are we on the journey to a Context Graph?
Ophir: The first requirement is having an AI Native system, with schemas and processes that are designed for context, not just state and observability. We re-architected our systems to be fully AI Native in 2024.
Second, we treat context data as “first class.” This means the "Why" data is of equal importance to all other data types.
Third, our agents already utilize context data to decide, store, observe, and audit decisions in discrete workflows.
We believe that the full concept of the Context Graph will be realized when systems of agents — not just specific services — can exercise autonomy across workflows with provable superiority.
David: Why not sacrifice quality to be more efficient?
Ada: Our value proposition is strongly rooted in quality data.
John: We aggressively lean into AI, but we won’t “ship it” if the results are not better or cause harm.
Ophir: We see a clear path to the Context Graph. We believe it describes how data businesses will scale in the age of AI, with quality and efficiency, grounded in domain knowledge.
David: Thank you, Ophir, John, & Ada. Glad to be on this journey with you!
For our first mention of the Context Graph, see our recent blog post The CPG Showdown: Love vs Strength.
For more on how context graphs are being discussed across the tech and venture communities, see What are context graphs by Simple.AI and Context graphs one month in by Foundation Capital.
For the original Context Graph idea, see AI's trillion-dollar opportunity: Context graphs, by Foundation Capital.
Thank you for following our progress. Send me a note with thoughts or questions. I’d love to hear how we can support your work.
Warm Regards,
David Goodtree
Founder and CEO, Foodgraph
More to explore
Discover additional articles, updates, and perspectives.
GLP-1 Rewrites CPG Strategy
GLP-1 Rewrites CPG Strategy
The best-selling weight-loss drug GLP-1 is rewriting grocery across the aisles.
CPG is feeling the heat:
- 31m US adults already take GLP-1 drugs, representing 23% of household shoppers.
- Consumers are dramatically shifting baskets to more protein and fiber, and less carbs and sugar.
- Big CPG is being forced to adapt. General Mills cut its annual forecast due in part to GLP-1 causing “a lasting influence in the food and nutrition landscape.”
And the pace is accelerating:
- “The GLP-1 surge is still in its infancy,” reports The Wall Street Journal.
- Search for “GLP-1 foods” just hit an all-time high in Google Trends.
10x Growth in GLP-1 Support Products
The number of CPG products positioned with “GLP-1” soared 10x in just 12 months.
- Unique GTINs with GLP-1 positioning zoomed from 38 to 388 products in our catalog over the past year.
- “GLP-1 Support” or similar phrasing appears on both new products and repositioned products, using on-pack icons and eCommerce titles & descriptions.
- Products span the aisles from fajitas to yogurt — not just supplements.
- GLP-1 users prioritize protein per bite and favor high fiber, while shunning sugar and seeking small serving sizes, healthy snacks, and new beverages.
- Claim words are all over the road — using phrases such as “GLP-1 Activator,” “Booster,” “Friendly”, “Nutrition”, “and Support” — and all terms are unregulated by the FDA.
Some Brands Promote GLP-1 Support. Retailers Don’t.
To capture demand for “GLP-1 Support” products, CPG approaches vary widely, while retailers are lagging.
Nutrition brands launch new product lines.
- Nestlé targeted the opportunity by launching Vital Pursuit, creating bowls, fajitas, and pizzas with “GLP-1 Support” badges, gaining wide retail distribution.
- Abbott introduced Protality nutrition shakes — “specially designed for GLP-1 users”, selling multi-packs at Walmart.
Food brands extend existing product lines & deepen commercial ties.
- General Mills’ Fiber One and Conagra’s Healthy Choice display GLP-1 icons and messaging on the front of packages and via syndicated data attributes.
- Danone created Oikos Fusion yogurt as a “one-of-a-kind opportunity” to serve GLP-1 users, but skips any reference to an explicit “GLP-1 Support” claim.
- PepsiCo originally saw GLP-1 as having a negligible impact on its business, but has since aggressively embraced the trend, adding high-protein and fiber variants to Doritos, SmartFood, and SunChips, while avoiding explicit GLP-1 Support positioning.
- Mars’ CEO even joined the board of a top GLP-1 drug company.
Private label brands pump up the protein & fiber, but avoid GLP-1 positioning.
- Dozens of new protein- and fiber-rich items are rolling out, such as Kroger’s “Simple Truth Protein” product line.
- But store brands haven’t used GLP-1 positioning in titles, descriptions, or on-pack.
- While 3x consumers take GLP-1 drugs versus those who follow a keto diet, retail brands haven’t updated their product positioning beyond that older weight loss approach.
Supplement brands take a novel approach of using disclaimers.
- Wary of appearing to make a medical claim — which is a regulatory hazard for supplements especially — many brands only reference “GLP-1” in the fine-print.
- Centrum gummies state “Consult with your health care practitioner on GLP-1 and Centrum use.”
- This approach gives products the needed GLP-1 association without overt marketing.
Retailers tip-toe into eCommerce tools or miss making PDP updates.
- A few retailers have posted GLP-1 Support landing pages, like ShopRite and Walmart.
- Most retailers’ eCommerce listings only include supplements, ignoring relevant foods like Progresso’s new high-protein soups or even their own private label protein and fiber offerings.
- Titles and descriptions keep legacy references to “Keto, Paleo, Mediterranean and Weight Watchers Diet Friendly”, like Kroger’s PDP for Star Kist Tuna, without referencing GLP-1.
Get the Data: 50 Brands’ Strategies
For a summary of top strategies, click to see 50 Brands’ Strategies for GLP Support, as excerpted in this table:
Foodgraph Takeaways
GLP-1 Support wins the prize for “Food As Medicine” in the mass market.
- Traditional weight-loss approaches — like precision nutrition, meal kits, and diet apps — require high engagement or expense.
- The GLP-1 approach requires less consumer commitment and often gets better results.
- With GLP-1, dietary adjustments are necessary — but not extreme — to handle the muscle and intestinal effects of the drug.
Millions of U.S. adults have already benefited from the combination of GLP-1 drugs and Food As Medicine. This grocery phenomenon began in 2021 when the FDA approved the current generation of drugs, and all signs point to continued acceleration.
CPG is capitalizing. But retailers haven’t yet.
- CPG is pivoting across the aisles from chicken bowls to snack chips to supplements.
- Strategies vary from launching new brands and products to just repositioning.
- Private Label has embraced protein and fiber, but not yet GLP-1 positioning.
- Retailers lag in eCommerce, despite their easier digital path versus new product formulation.
Further Reading
Foodgraph, Harvesting Trends x 1M UPCs
Our original coverage of GLP-1 support products
ADM, Insights from Anti-Obesity Medication Users
80% of consumers are willing to pay more for food and beverage products.
Circana, GLP-1 Users to Represent 35% of U.S. Food and Beverage Sales
GLP-1 users already represent 23% of household shoppers.
Hartman Group, The impact of GLP-1s in an era of disruption
31m US adults currently take GLP-1 drugs to lose weight and manage diabetes.
Foodgraph Next
Our next major catalog update will arrive in time for Mother’s Day, Memorial Day, and Graduations.
How can we support your 2026 goals? Let's talk!
Warm Regards,
David Goodtree
Founder and CEO, Foodgraph
GLP-1 Support: 50 Brands. 4 Strategies.
GLP-1 Support: Much More Than Supplements
“GLP-1 Support” products are designed or marketed to make the consumer’s weight loss efforts sustainable and effective while taking these drugs.
The fast growth of GLP-1 drugs is rewriting CPG strategy:
- 31m US adults already take GLP-1 drugs, representing 23% of household shoppers.
- Consumers are dramatically shifting baskets to more protein and fiber, and less carbs and sugar.
- The number of CPG products positioned with “GLP-1” soared 10x in just 12 months.
- Brands across categories are feeling the heat to satisfy the demand and differentiate.
For background on the GLP-1 phenomenon — and its effects on CPG — read our companion blog post GLP-1 Rewrites CPG Strategy
The 4 Strategies: How Brands Approach GLP-1 Support
To capture demand for “GLP-1 Support” products, CPG takes one of four main approaches:
- Acquire a brand, such as Lactalis’ purchase of :ratio.
- Launch a new brand, such as Abbott’s introduction of Protality.
- Launch new products, such as Bonduelle’s lunch bowls.
- Update messaging, such as Haleon’s Centrum and Tums.
Get the Data: 50 Brands’ Strategies
Use the table below to explore the 4 strategies by 50 specific brand names:
- The strategy column lists the approach used by the specific brand name.
- The keywords column shows whether each brand uses the terms “GLP-1, “protein, and/or “fiber” on packaging and digital content.
- The “GLP-1” usage column lists whether the term “GLP-1” is used on-pack, in eCommerce text, using disclaimer language, or none (no explicit "GLP-1" reference).
Instructions:
- Click the “columns” button to select the fields displayed.
- Click the “XLSX” button to download the data in an Excel spreadsheet.
- Use the column header controls to sort and filter the rows.
How Can We Help?
For more information about GLP-1 and CPG, read our blog post GLP-1 Rewrites CPG Strategy.
Foodgraph offers the largest US catalog of CPG product data.
How can we support your needs? Let's talk!
Warm Regards,
David Goodtree
Founder and CEO, Foodgraph
The Great AI Debate
Our executive team is having a heated debate about the proper use of AI.
The debate is passionate, respectful — and I believe — highly productive.
AI is deceptively good. Initial AI results are high-quality. The LLM explains its “thinking” robustly in natural language. Humans are convinced.
The user may even say “My friend Claude got it right”.
Our CTO is worried that AI tools are becoming agents of human assimilation. Users unwittingly abdicate their responsibility to a convincing bot, because the results appear credible. He calls this phenomenon the “Borgification” of software engineering, referring to the famous Star Trek story line.
Our VP PM is more sanguine, saying “Claude is my fast dumb friend” who needs oversight. She agrees with Anthropic’s President Daniela Amodei, who believes “Claude is really a tool to help promote your ability to think more deeply, to solve problems, and to be ambitious about the types of projects you are taking on.”
Our AI Architect shared how Steve Jobs described a rock tumbler as a metaphor for how great ideas come to life. This machine smashes together crude, unfinished inputs to create beautiful polished stones. Jobs explained that 90% of the work in creating great products is not in the big idea or the raw materials, but through the combination of friction + time + teams to turn the inputs into amazing products.
Job’s metaphor rings true to me:
Friction + time + teams create great products, not AI or other tools.
The SaaS Apocalypse may be coming for some, but only when humans abdicate their role.
AI raises the bar of what humans can do — and are now expected to do.
To paraphrase Steve Jobs:
We — the humans — are the rock tumblers, not AI.
Watch: Daniela Amodei, Responsible AI and Steve Jobs’ Rock Tumbler Metaphor.
How Can We Help Your Work?
Send me a note with thoughts or questions. I’d love to hear how we can support your work.
Warm Regards,
David Goodtree
Founder and CEO, Foodgraph
Get better CPG product data today.
Book a demo today to learn how Foodgraph can help.