The reverse-engineering approach to extruded dog food formulation works backwards from target kibble properties—density, texture, digestibility and palatability—to determine the ingredients and extrusion conditions needed to achieve them. In the pet food extrusion process, moisture, specific thermal energy (STE) and specific mechanical energy (SME) interact with starch, protein, fat and fibre to influence final kibble properties. This makes formulation and extrusion connected decisions within pet food product development.
Dr Radhika Ganesan, R.D., PhD, Head of Regulatory Affairs & Nutrition Science, Food Research Lab
15+ years of experience in functional food and nutraceutical formulation, including strain selection, stability engineering, and FSSAI/FDA claim substantiation.
The reverse-engineering approach to extruded dog food formulation works backwards from target kibble properties—density, texture, digestibility and palatability—to determine the ingredients and extrusion conditions needed to achieve them. In the pet food extrusion process, moisture, specific thermal energy (STE) and specific mechanical energy (SME) interact with starch, protein, fat and fibre to influence final kibble properties. This makes formulation and extrusion connected decisions within pet food product development.
Reverse-engineered extruded dog food formulation takes the target kibble into account and moves back to the process of formulation and extrusion that can create this target kibble. As opposed to ingredient selection followed by extruder adjustments, the development process begins with the measurable targets, including density, expansion, hardness, porosity, nutritional performance and palatability. These targets then guide pet food formulation, ingredient selection and processing conditions. [1]
This methodology is particularly relevant to dry pet food manufacturing and more specifically to the pet food extrusion process, where any changes in moisture, starch, protein, fat and/or fibre can affect formulation response to temperature, shear, pressure and mechanical energy.
The goal is not to replicate a kibble’s appearance but to find the formulation and process that deliver its required properties — a structured pet food product development cycle.
Figure 1 — Reverse-engineering workflow for extruded dog food
Target kibble specifications guide formulation and extrusion decisions, followed by final-kibble measurement and iterative refinement.
The pet food extrusion process combines hydration, heating, mixing, shearing, compression and pressure before the die and drying. Starch gelatinizes, protein changes structure, fat affects lubrication, and fibre affects water binding — so the same settings give different results when the formula changes. [2]
Important extrusion cooking parameters include feed moisture, temperature, STE, SME, screw speed, residence time, pressure and die geometry, and must be considered together, since a formulation change reshapes the ingredient-process interaction pet food systems meet during thermal and mechanical processing.
Moisture control extrusion conditions influence viscosity, pressure, mechanical-energy demand and starch cooking, while temperature sets the environment for ingredient transformation. Sá et al. (2026) processed canine and feline diets at SME levels of 8, 20 and 30 kWh/t, adjusting STE to hold total specific energy; different STE: SME ratios still produced well-formed kibbles and supported starch gelatinization extrusion. The point is to manage the thermal-mechanical energy balance, not any single variable. [3]
8 kWh/t SME — low mechanical energy | 20 kWh/t SME — mid mechanical energy | 30 kWh/t SME — high mechanical energy |
Figure 2 — Ingredient-process interactions during extrusion
Moisture, thermal energy (STE) and mechanical energy (SME) interact during extrusion, influencing starch gelatinization, protein transformation and final kibble properties.
During starch gelatinization extrusion, water and energy promote matrix formation — but starches differ. Kaelle et al. (2024) compared seven sources at matched starch levels: tuber sources gave lower kibble density and more expansion, while pea gave denser, less-porous kibbles. [4] For kibble texture optimization, let target kibble density and expansion steer starch choice.
Protein denaturation occurs during extrusion, but the nutritional outcome depends on formulation and conditions. Hsu et al. (2024) reported higher amino-acid digestibility and protein quality in most grain-free diets after thermal treatment, with methionine and tryptophan limiting. Protein selection should consider amino-acid quality, moisture demand and functional behaviour, not crude protein alone.
Match the Right Starch and Protein to Your Target Kibble
Food Research Lab can support ingredient screening, extrusion-response evaluation and final-kibble validation against defined product specifications.
A reverse-engineering approach first sets measurable targets:
These turn a product concept into criteria that trace back to formulation and processing, supporting kibble texture optimization.
Once a target is set, each attribute links to the formulation and process that drive it. Across the macro-nutrient framework:
Heat- and moisture-sensitive nutrients or functional ingredients need to be evaluated for their tolerance to the process as well as how much can be used.
Retention cannot be determined by the extrusion temperatures alone since the results vary with the nutrient, formulation and complete processing sequence.
For dog food nutrient retention, the evaluation must include:
Extrusion → Drying → Storage
Nutrient stability is nutrient- and process-specific. While extrusion can enhance certain nutrient attributes, while reducing retention of other sensitive components. Testing of finished products should be done to determine their performance.
Brief: Kim et al. (2025) compared fat from inside whole soybeans, from outside (soybean oil) and the absence of fat on the extrusion behaviour of soy-based dry expanded dog kibbles.
Objective: Investigate effects of fat distribution on extrusion behaviour.
Study design: A 2 × 3 factorial design evaluated two fat levels across three inclusion methods: internal fat, external fat and no added fat.
Key findings: To obtain the same target bulk density, the external soybean oil needed around 404 rpm, while internal fat needed 351 rpm and 309 rpm without addition of fat. Soybean oil was found to cause extruder surging and more variability in kibble size.
| 404 rpm External fat — to reach target bulk density | 351 rpm Internal fat | 309 rpm No added fat |
Why it matters: Fat placement can change the operating conditions and process stability required to achieve the same target product. It should therefore be treated as both a formulation and process variable.
Fat placement should be set alongside screw speed, target bulk density and process-stability requirements rather than being treated as a final recipe adjustment.
H3 Practical Takeaways for Pet Food R&D Teams
Connect the product specification to ingredient selection, extrusion conditions and final-product testing — the discipline behind reliable pet food formulation.
Workflow: define the target → identify ingredient drivers → set the extrusion window → measure the kibble → refine the formulation-process combination.
Factor | Interaction | Potential kibble outcome |
Starch source | Gelatinization and matrix formation | Density, expansion, porosity |
Protein source | Structural transformation and matrix interaction | Texture and nutritional quality |
Fat localisation | Lubrication and energy transfer | Expansion and process stability |
Fibre | Water binding and viscosity | Density and hardness |
Moisture | Viscosity and energy response | Expansion and structure |
STE/SME balance | Thermal and mechanical treatment | Gelatinization and product quality |
Take Your Kibble from Target Spec to Stable Production
From starch and protein screening to STE/SME optimisation, stability testing and nutrient-retention validation, Food Research Lab supports evidence-based pet food product development.
Modelling links formulation and extrusion cooking parameters to the ingredient-process interaction that pet food products show during extrusion. Nielsen et al. (2025) map moisture, temperature, screw speed, formulation and product properties; Cheng and Feyissa (2026) review hybrid and machine-learning prediction methods. [6] These pinpoint influential variables before physical trials.
The reverse engineering links target kibble to formulation and extrusion factors that should be used to obtain it. Through the combination of the management of starch, protein, fat, moisture and energy, pet food R&D teams can provide more consistent results and evaluate better evaluate dog food nutrient retention.
Turn your formulation concept into a validated pet food product with Food Research Lab. Get expert support and end-to-end pet food product development services across ingredient selection, formulation, extrusion optimisation, process development and final-product validation.
Start Your Pet Food Product Development Project →
Reverse engineering begins with specified kibble, and nutrition needs and goes backwards from there to determine proper ingredients and extrusion settings, rather than starting with ingredients and then changing the process.
While starch impacts gelatinization, expansion, density and porosity, protein affects structure and nutritional quality. Sources may vary in their performance under the same extrusion conditions.
The impact varies depending on the nutrient, formulation, and the severity of the extrusion process. Nutrient retention thus needs to be considered after extrusion and in cases where needed, during subsequent drying and storage.
Get a free packaging compliance review within 24 hours.
Food Research Lab strives for excellence in new Food, Beverage and Nutraceutical Product Research and Development by offering cutting edge scientific analysis and expertise.