Customer spotlight: Once Upon a Farm — testing embedded in the product lifecycle

Please note that the following transcript was generated automatically and may contain typos or summarization.

Amanda: Welcome, everyone. We're really glad you're here. Today we're talking about something that comes up constantly in conversations with CPG teams: how do you actually build testing into your product process — not just as a one-off before launch, but as something that's an embedded part of the culture, woven into how you innovate from the start.

To dig into that, I'm joined by the wonderful Sydney Perotti from Once Upon a Farm. For those who don't know, Once Upon a Farm is an organic childhood nutrition brand focused on helping parents foster a lifelong love of healthy, real food for their kids. And I can say from experience, they genuinely walk the walk when it comes to consumer-informed innovation.

For our guests — please drop any questions using the Q&A functionality at the bottom of your screen, and we'll take a look at them at the end of this conversation.

Sydney, we're so glad to have you here. Would you share a little about yourself, what you do at Once Upon a Farm, and your background?

Sydney: Thanks so much for having me. My name's Sydney, I'm the Senior Manager of Consumer Insights here at Once Upon a Farm — quickly approaching my two years here, and super excited to be here. We have an amazing brand, and our organization really invests in consumer insights.

I've been working in insights almost my entire career. I started on the insights team at PepsiCo in 2018, and made my way into smaller, more emerging brands. I've fostered a real love of research and using it to make the most out of every launch, every reformulation, every time we change packaging or adjust positioning. Research touches all functions across the organization — not just marketing, not just R&D.

I get my fingers in everything everybody's working on. I support sales, I work closely with marketing, I'm in conversations with R&D and leadership. Everybody here loves insights, so it's a great place to be — I feel really supported in this role.

Amanda: We've seen that with Once Upon a Farm tenfold. Speaking of the different places you touch along the product lifecycle — I'd love to hear about your innovation philosophy. Once Upon a Farm is a brand that really leads with the consumer. Can you give us a quick picture of how your team approaches new product development, from ideation to prototype?

Sydney: At Once Upon a Farm, we do a good job balancing intuition with information. Our first step is already pretty in-depth, specific concept testing with a specific goal in mind. The team here is so creatively minded and able to identify what the next big bet is going to be — we're a founder-led brand, so Cassandra, who leads our R&D practices, is really at the forefront thinking about new categories and places we can make a big impact.

We take that rich intuition from leadership and go into: what's the best way to bring this to life and take it to market? We start with detailed concept testing. Coming out of that, we understand — how can we tweak the positioning, hone in on a specific functionality, adjust the messaging — in order to bring something to life that's really impactful and resonates with our parents and target audience.

That first step is already deeper than at a lot of organizations. At my last role, we'd start with a laundry list of potential categories and take a screener-type approach. At Once Upon a Farm, we lean on intuition as the first step, then get into the nuts and bolts of consumer research.

From there we cross-functionally dissect the results — R&D isn't the only one who looks at concept results, marketing isn't the only one, sales and finance look too. We want to make sure the proposition we're bringing to life checks off all our consumers' boxes, and that takes everybody cross-functionally.

Then it becomes a brief-writing process. We take all that information — a lot of times it's a concept test, but typically with concept tests we also do attitude and usage research, to understand the key occasions, current perceptions of the category, so we can optimize the launch — not just the physical product, but the competitive standpoint and how we embrace our brand equities.

Then we hand that off to R&D, and it depends on the weight of the launch — a huge bet versus a couple line extensions. If it's a big bet, sometimes we'll do packaging testing, occasionally messaging testing, but more often we take the next step into physical product testing.

What I always say is: if we don't have time to make changes to the product after we get the physical product test results back, we shouldn't do the test at all. The purpose of that final step before launch is to make sure we're bringing to life the concept that performed well in the initial round of testing. Getting to a winning concept is one thing; making sure the product lives up to the expectations the concept set is really important.

Amanda: So pretty much consistently, you start with a concept test no matter what?

Sydney: Yeah, we love a concept test here. We run a lot of them. I think the way we approach them is unique — we don't just take the typical template of "what's the purchase intent, how believable is this, how unique, how new and different." We like to ask more nuanced questions about category usage, parent perceptions, and connecting all of that to age of child and stage of development. We get really detailed with our concept testing, and I lean on the Highlight team a lot for that — the questions you ask, and making sure we're asking the right ones, is the most important part of handing off a brief that the rest of the team can actually act on.

Amanda (reading a Q&A): "To clarify — is the product already developed when you do concept testing, or are you testing thoughts before the product is developed?"

Sydney: Good question. We already have an idea of a territory we want to enter — a category, or a trend like protein — so we know we want to enter protein, but what concepts can we test within that territory to make sure we're bringing something new and unique, not just entering like everybody else? We don't have physical product yet at that stage — we have general ideas of territories, and go into concept testing to go deeper: if we're going to enter a category, we have three ideas of ways in, and we test each in detail before moving into product development.

Amanda: Companies in kids' food, baby food, or even pet products have a unique challenge — you're not just winning with the purchaser, you're winning with the person (or pet) actually using the product. Does that raise the stakes for you, and how do you mitigate that risk?

Sydney: It's so important to get it right in kids and pets. We think of our consumer base as two connected but distinct audiences — baby/toddler parents and "big kid" parents (early elementary school age). Getting it wrong looks very different for each.

For baby/toddler parents, there's a huge fear of getting things wrong with their baby — it's a new phase of life, a lot of parents don't have firsthand experience with feeding development and milestones. Brand trust, ingredient quality, sourcing, clean label claims, non-GMO, organic — all of that matters because these parents are fairly inelastic about food for their kids. They see their kids as pure little bundles of love and don't want to give them anything artificial or chemically processed. In one-on-one interviews, parents will say things like: their baby is six or seven months old, transitioning from milk to solids, and they want to set their child up for a healthy relationship with food into adulthood — they're already thinking that far ahead.

As kids get older, parents loosen the guardrails a bit — kids are trading snacks at school, having candy at friends' houses — so value becomes the center of decision-making, along with avoiding waste. Picky eating and the "brown food phase" (chicken nuggets, etc.) is a big concern. If we bring something to market that parents feel good about and kids ask for again and again, we're doing our job. There are also more snacking occasions in this stage — after-school activities, sports, birthday parties — so pack size, channel strategy (club), and shopping shifting from grocery to Costco/Walmart/Target all matter as kids get bigger.

Amanda: Can you walk us through a moment where testing early in the process saved you from going down the wrong path?

Sydney: I have an example I can't fully share details on yet since this line hasn't launched. We did physical product testing with enough lead time to decide whether to launch or fine-tune the recipe. This was a completely new line we were putting a lot of weight on. Bench-top testing with R&D seemed fine, but as the product sat in its packaging longer, we identified through the data that our hero flavor had textural issues emerging over time. We pushed the launch back significantly to fix it.

Texture is a huge trigger for kids trying a new food — the flavor can be great, but if the texture isn't right, it won't fly. If a parent buys a premium product and the kid takes one sip and doesn't like it, that's a wasted purchase, and the parents won't try us again. That's become something we're now much more hyper-focused on for big-kid launches.

Amanda: What are some wins that have helped you defend the ROI of testing to internal stakeholders?

Sydney: Teaching people something new. Our leadership team has so much experience in their field, so if research can teach them something new, the ROI speaks for itself. I get creative with questionnaires and go cross-functionally beforehand — letting sales, marketing, etc. know what I'm testing and asking if they have questions they want answered. That helps get total enterprise buy-in, not just from the one stakeholder most interested in a given piece of research, but everybody — so more people start asking to fund additional research when there's budget available.

I also share questionnaires broadly before running them, which shows good partnership. Consumer insights technically falls under marketing, but it delivers enterprise-level value — the more you connect people cross-functionally, the more success you'll have running research and getting budget.

Amanda: There's a time and place for different kinds of testing — employee testing, which Highlight supports, versus larger consumer panel testing. What's the time and place for each?

Sydney: Balancing that is important. Not every decision needs an entire research study. Sometimes we do messaging and claims research, sometimes we don't; sometimes packaging research, sometimes not. For smaller things — like exploring new shapes for our puffs — we can lean on our employee panel or our loyal subscriber base (keeping in mind they'll be a bit more biased). As an emerging brand, I don't have unlimited research budget, so for big launches or decisions heavily impacting future growth plans, I always invest in research. For smaller things, you can lean on intuition.

Amanda: How does it actually look to act on the data you get back — who gives it the go/no-go?

Sydney: It depends on the project. For physical product testing, I take the final report, comb through it, and within a day or two flip it to key stakeholders. We're a small enough team that getting everybody in a room is easy. I write a summary and attach the data — often we'll export the Highlight PowerPoint directly — but I always add my own summary, because quantitative data gives clear direction, but you have to add the business context. You can't expect R&D to look at results and know exactly what to do next.

I slice the data multiple ways (I love how easily you can slice and filter segments in the Highlight tool), then write a short summary highlighting the top three or four main findings, the direction we feel strongly about, and next steps and implications — often not just "go with concept A over B" but "share this finding with sales," "share this with R&D," "share this with marketing" because it affects messaging and campaign development.

I also send weekly summaries to our CMO on key findings and trends — household penetration, buy rate, repeat rate from panel data — so the most important things make their way up to the C-suite. I'll sometimes put myself out there and say the data points in a direction that's a bit different from where we've been discussing, and even if the answer is "we hear you, but we're moving this way for now," it's worth saying.

Amanda: You mentioned sharing findings with sales — a lot of times we think of research as being for R&D or marketing, but giving sales tools around what people care about right now is pretty astute.

Sydney: Marketing and sales work really closely at Once Upon a Farm. This goes a long way with customers too — if we're in a meeting with Walmart and say "we ran a survey with people who primarily purchase this category at Walmart," that's valuable to them, because buyers don't always have the resources to run their own research. Anything incremental we can give them on top of point-of-sale data goes a long way. Sales is a great audience for research — it fosters great relationships with customers, and account managers are constantly in conversations with buyers who benefit from that. We'll even include a clip from Highlight's video responses in a sales deck — the Highlighter community gives us amazing quotes.

Amanda: Shifting to your testing roadmap — you've talked about innovation and concept testing. What about what we'd call "renovation" — maintaining hero SKUs, and reacting if a product's sales suddenly plummet?

Sydney: That's hard to plan for at the start of the year — timelines change, submission dates move, and a big business question can come out of nowhere needing research. The way we think about roadmapping is: what products are we launching, how many resources go to new launches, and how much do we save for core research? I always over-save credits and resources for core research, because maintaining a healthy core is so important.

We partnered with Highlight last year on physical product testing for a core line — a phased project with in-home usage testing followed by one-on-one interviews with people from that original test, where we could sit down for an hour and hear, for example, what their child said about a flavor or their immediate reaction to packaging. Surveys are capped around 20 minutes, so one-on-ones really supplement that. We learned a lot about how to optimize a core piece of our business.

Successful innovation launches are few and far between. If you're choosing between investing in an innovation launch or maintaining a healthy core, I'd choose core every time — that's the foundation of the business. Emerging brands especially can't afford to let their core slip. We've changed packaging and updated our logo multiple times to stay modern with new generations of parents — millennial and Gen Z parents look very different from Gen X parents. Constant evolution of your core, supplemented by innovation research, gives you a strong foundation to grow from.

Amanda (reading a Q&A): Are you mining trends and data to decide what areas to go into, versus running a study specific to a concept you're already testing — i.e., are you monitoring large-scale consumer trends?

Sydney: Yes, absolutely. We monitor trends across protein, gut health, immunity, brain development, and more. No single trend completely dictates a direction — it acts more as a prompt: does this make sense for us to participate in, and if so, how? We use intuition to figure out where OFARM should show up. We have a strong, unique brand identity, and we like to "OFARM-ify" things — we're averse to launching something just because it's trending for someone else. We want to stay true to who we are, and that's the basis for concept testing as a first step.

Amanda (reading a Q&A): What key skills do people need to grow a career in consumer insights, especially with AI taking up so much space?

Sydney: Probably a controversial answer — I think AI is really helpful in some ways. Some research platforms can be limiting in how you slice and export data, and AI is a great way to take raw survey responses, upload them, and go back and forth — a great way to practice asking good questions and dig into whether a segment is perceiving something differently.

What I don't lean into as much is AI persona testing. Call me old-fashioned, but I love leaning on real people, real parents, real consumers. Qualitative testing, even though it takes longer — 13 hours of moderating interviews — gives insights I find far more valuable. I'm a huge proponent of continuing to test with real people, but using AI to make the research process more creative — helping sequence really good questions to answer a complex business question. I'll test with real people every time.

Amanda: Thank you so much, Sydney — this has been a wonderful conversation, and we've loved learning about the research culture at Once Upon a Farm. If you'd like to learn more about how Highlight can help your team do what Once Upon a Farm does, we'll drop a link in the chat. See you all next time.

Sydney: Thank you so much, Amanda. This was so much fun.