There is no technical argument for removing modified starch from a sauce. It is inexpensive, it survives heat and shear, it tolerates acid, and it behaves predictably. Every clean-label reformulation that removes it is doing so because the words "modified starch" are unwelcome on the pack — which means the replacement must be judged on label acceptability first and everything else second. Low acyl gellan can do this, in tomato sauces and similar products at low percentages of the formulation, and it does so with far less material than starch. What it does not do is cost less, and pretending otherwise wastes everyone's time.
What starch does that everything else does badly
| Property | Modified starch | Native starch | Hydrocolloid route |
|---|---|---|---|
| Heat stability | Excellent | Poor without modification | Good |
| Acid tolerance | Very good | Poor | Depends on pH |
| Freeze–thaw | Good when modified | Poor | Varies |
| Cost to thicken | Lowest | Lowest | Higher |
| Label position | The problem | Fine | Usually fine |
| Shear tolerance | Good | Poor | Generally good |
Note that native starch — the label-friendly version — is worse on almost every performance axis. This is why most clean-label sauces do not simply swap modified for native starch and call it done.
Substitution candidates
| System | What it delivers | Working dose | Trade-off |
|---|---|---|---|
| Low acyl gellan | Clean thick body, good suspension of spices and solids | low percentage of the formulation | Needs cations, less cheap to thicken |
| Xanthan | Reliable viscosity at low dose | low | Perceived as highly processed by some consumers |
| Native starch plus hydrocolloid | Closest to the original texture | moderate | Still needs the second component |
| Fibre-based systems | Body and a fibre claim | higher | Strong flavour and colour impact |
The honest summary: for most sauces the realistic answer is a native starch plus a low dose of gum, rather than any single replacement. Claims about one-material solutions tend not to survive a pilot plant.
The dose arithmetic
| Line | Modified starch system | Gellan-supported system |
|---|---|---|
| Thickening component | high single-digit percentage | low acyl gellan at a low percentage |
| Powder per 1,000 kg | substantially higher | substantially lower |
| Weight ratio | baseline | a small fraction for the thickening contribution |
| Additional components | none | often native starch alongside |
A gum route gets you out of modified starch at a small fraction of the dose, which matters because shipping, storing and weighing kilograms of starch is real cost even where the material itself is cheap. It will not usually get you to a lower ingredient figure on its own, because gum is priced higher per kilo. Multiply these weights by your own prices — that is the only part you need to supply.
Three cases where you should not switch
- Products with no label pressure. If nobody objects to modified starch, there is no reason to pay more for a worse-performing system.
- Very high-solids, very acidic products. Both extremes stress hydrocolloid systems in ways starch shrugs off.
- Cost-led lines where the label is already accepted. This substitution is bought, not saved into. Budget for it accordingly.
Migration in four steps
- Write down why starch has to go. "Clean label" is a reason; the specific words matter, because they determine which replacements qualify.
- Match viscosity across temperature, not at one point. Sauces are made hot and eaten warm. Systems that match at 20 °C often diverge badly at 60 °C.
- Test after retorting or hot fill. Whatever heat step you run changes the system. Validate afterwards.
- Check acid tolerance at your actual pH. Tomato products sit low enough to matter.
Failure modes
| Symptom | Likely cause | First check |
|---|---|---|
| Thin after retort | Heat or acid breaking the system | Test after the real heat step |
| Grainy texture | Native starch behaving as native starch | Add or increase the gum component |
| Weeping on standing | Poor water binding | Solids and gum dose together |
| Viscosity differs hot versus cold | Wrong material for the temperature profile | Measure across temperature, not at one point |
| Flavour changes | For fibre-based systems, expect it | Sensory panel before committing |
| Separation of spices or solids | Suspension not handled by the thickener | Add a suspending component |
Tools
- Dosage calculator — includes ketchup and tomato sauces, and chilli and hot sauces.
- Formula pre-check — flags pH and ion combinations that fail first.
- Troubleshooting tree — work back from the defect.
Frequently asked
Is this worth doing on cost alone?
Almost never. Modified starch is the cheapest way to thicken a sauce. This substitution is justified by the label, not by the costing sheet.
Can I just use native starch instead?
Usually not by itself. Native starch performs worse than modified on heat, acid and freeze–thaw, so most clean-label sauces pair it with a low dose of gum.
How much less material goes in?
Dramatically less — the gum route works at a small fraction of the starch weight for the thickening contribution. Whether that translates to a lower ingredient line depends on your own two prices.
Will it survive retorting?
Usually, but validate after the actual thermal process rather than assuming. Both heat and acid stress these systems in ways that only show up at scale.