Hey folks, let’s cut to the chase: if you’re a formulation nerd, a product manager, or a small brand trying to level up to large-scale production, you’ve definitely hit that wall. You’ve got this lab formulation that works like a charm—smooth, stable, does exactly what it’s supposed to do for your 1L lab batches. But you ask your third-party manufacturer to crank it out to 1,000L, and suddenly it’s a hot mess: lumpy, uneven texture, shelf life half what it should be, maybe even a weird odor no one saw coming. As a formulation supplier who’s been deep in this muck for over a decade, I’ve seen it all—from skin serums that separated mid-batch to cleaning solutions that clogged 500L mixers. Scaling a formulation isn’t just “bigger bucket, same stuff”; it’s the part of the job that makes or breaks your launch, profit margins, and even your sanity. Today, I’m breaking down the real, no-BS steps I use to make scaling work, not just sound good on a spec sheet. Formulation
First off, let’s get one thing straight: your lab batch is a blueprint, not a finished playbook. Most people start scaling by just multiplying every ingredient by how much they want, right? That’s the fastest way to blow up a batch. Lab mixers spin at like 500 RPM max, but a 1,000L industrial mixer? It’s way slower—like 150 RPM for viscous stuff, sometimes even lower. If you pour a lab recipe straight into an industrial mixer, the surfactants or emulsifiers that keep your serum stable won’t disperse evenly—they’ll clump together at the bottom, and the top layer will be watered down. I learned this the hard way early on: a client had a CBD topical that worked perfect at 500mL, but when we went to 200L, half the batch was a thick goop that couldn’t be pumped. Turned out they just doubled all ingredients, no changes to mixing order or speed. I ended up reworking their process, and that’s when I realized: scaling starts before you touch any big equipment—it starts with the “mix matrix” for the target scale.
So step one, actually map the scale-up matrix before adjusting a single batch. What’s that? It’s basically a blueprint that matches your lab’s small-scale process to what the industrial mixer can handle, without guessing. First, define your “critical process parameters” (CPPs) — stuff that actually affects your product’s quality, not just vibes. For emulsions, that’s mixing speed, shear rate, temperature, and order of addition. For powders, it’s blending time, bin fill level, and impeller position. For our CBD topical, I broke down what the lab did: they added oil phase to water phase at 70C, mixed at 400 RPM for 10 mins, then cooled and added CBD at 40C. The industrial mixer here was a 200L double planetary mixer, which doesn’t have the same high-shear disperser the lab used. So instead of just cranking speed, I adjusted the order: first, I pre-dispersed the CBD in a tiny bit of oil at high shear in a small inline mixer (a trick we use all the time for actives that hate clumping), then added that pre-dispersed mix to the main batch when it was at 50C, and upped the mixing speed for the water phase to 250 RPM (not too fast, but enough to avoid clumps). That tiny adjustment fixed the goop problem in 2 test batches—no more wasted materials or downtime.
Next, test with “scale-down models” before you buy 500L of raw materials. Wait, what’s a scale-down model? It’s a small batch (like 10L to 50L) that mimics exactly what your industrial mixer does, not just what your lab does. Most manufacturers will have a small pilot mixer that matches the shear and temperature of their big equipment—use that. Don’t just make a 10x lab batch. For a client making a hair mask that was separating at scale, we tested 3 different 20L batches in the plant’s pilot mixer: one with the exact same mixing order as lab, one with a 5-minute hold after adding emulsifier, and one with a slower cool-down. The one with the hold worked, so we locked that in before touching the 500L batch. Here’s the thing: raw materials aren’t consistent either. A 100g bag of xanthan gum from your lab supplier might be slightly different from a 25kg drum from the bulk chemical house. Sometimes the particle size changes, or the moisture content, which can throw off your mix. I always tell clients to run 2 small tests with both a new drum and a leftover lab bag, just to catch that. Last quarter, a skincare brand had their preservative not work at scale, and it turned out the bulk parabens (yep, they still get used sometimes) had a higher moisture content, so the effective concentration was lower. We caught that in the 20L scale-down test, saved them a $40k batch that would’ve been ruined.
Then, get the mixer and process team on your side EARLY. Not after the lab is done, not when the drum is at the plant. When I take on a new formulation client, the first call I make after sending over the lab batch is to the manufacturing engineer at their chosen plant (or my in-house ops guy, for our direct clients). A lot of formulators get so caught up in the ingredients that they forget the equipment is a whole other beast. For example, a liquid laundry detergent client wanted a batch that had 15% actives, which was thin in the lab, but the plant’s piping system for filling 500L drums only works with liquid over 500 cP. So we had to bump the active blend to 17% to hit that viscosity, which was easy enough, but we would’ve wasted a month if we didn’t flag the filling constraint upfront. Another mistake I see: adding volatile actives (like essential oils or some CBD isolates) too early in the mix. If your mixer has a vent, those actives will evaporate before they mix in. We always tell clients to add those in the last 10% of the mixing time, when the batch is almost done, to avoid loss.
Now, here’s the part that’s not in any textbook: human error and batch consistency. I’ve had a client have 3 perfect scale batches, then a dud that was totally off, and it turned out the new operator didn’t set the mixer timer for the full duration. At small lab scales, you’re one person mixing for 5 mins, but at 1,000L, you have 2-3 steps, 10-minute holds, temp checks every minute. I always build in a “batch record review” step, where we go through every parameter with the operator and their supervisor before starting. Also, we do a “mini QC check” after the first 10% of any scale batch—just test pH, viscosity, or active content to make sure nothing’s going sideways. For a toner client, that catch was crucial: the first 50L of a 500L batch had a pH of 6.2, which is the target, but the next 50L was 7.8, and it turned out the operator added the wrong buffer. We stopped the batch before adding the rest of the ingredients, saved half the raw materials, and adjusted the process so buffers get added slowly while mixing, not dumped in.
Another big one: shelf life validation at scale, not just lab level. A lot of formulators do a 2-week stability test in the lab, which is fine for a 1L bottle, but at scale, the batch has more “product mass” in the center, so it takes longer to reach temperature extremes. I always tell clients to run a 4-week stability test on the first 500L or 1,000L scale batch, even if the lab test was good. I once had a sun care brand that passed lab stability (no separation, no SPF loss after 4 weeks), but their 2,000L batch had 10% SPF loss after 2 months, because the center of the big storage tank stayed warm for longer, breaking down the avobenzone. We fixed that by adding a tiny bit of octocrylene (a common SPF stabilizer) that only adds 0.5% to the formulation, and it brought the SPF loss back to zero. The lab never would’ve caught that, because small batches cool too fast.
Wait, let’s talk about what NOT to do, because that’s the stuff I repeat to clients like a broken record. First, don’t “rescale” without adjusting shear. Small mixers have high shear; big mixers have low shear for viscous materials. If you have a cream that’s 100,000 cP in the lab, a 1,000L mixer will struggle to blend it evenly if you just multiply. Second, don’t skip the scale-down tests for high-risk ingredients—actives, emulsifiers, preservatives, all that stuff that can make your product unsafe or ineffective. Third, don’t forget about transportation and storage. A batch that’s fine in a 1L bottle might separate in a 200L storage tote because the tote is stored sitting still for weeks, instead of being rotated like a lab bottle. We add a tiny amount of anti-settling agent (no more than 0.1%, so it doesn’t change the feel) for storage over 30 days, and that fixed a problem for a fruit puree puree bar client whose puree was settling into a slush at the bottom of their 1,000L totes.
Now, let’s wrap this up with a real example, to make it concrete. Last year, a small clean beauty brand came to me with a hyaluronic acid (HA) serum that worked amazing at 1L: 0.2% HA, 5% glycerin, 1% panthenol, the rest water, stable for 6 months. They needed 5,000L for a Sephora launch, and their first manufacturer ran a 500L batch that was lumpy, had clumps of HA, and separated after 1 week. I got on site with them, first did a scale-down test in their plant’s 50L pilot mixer. The original process was add HA to water at 60C, mix at 300 RPM for 15 mins. The pilot mixer’s shear was way lower, so HA couldn’t disperse. So we adjusted the process: pre-mix the HA in 10% of the water with a small high-shear lab mixer (inline type) first, so the HA particles are already broken up, then add that mix to the main water phase at 50C, mix at 200 RPM for 10 mins. We also added a tiny amount (0.05%) of a non-ionic surfactant to help HA stay dispersed at scale. We ran the 50L test, it was smooth, no clumps, stable for 2 months. Then we ran the 500L batch, same parameters, and it worked perfectly. That batch is now on Sephora shelves, selling like hotcakes. The total time was 2 weeks from when they came to me, no wasted raw materials, no delays.
Here’s the key takeaway: scaling a formulation isn’t about making a big version of your lab batch. It’s about adapting every step—mixing, temperature, ingredient order, even storage—to fit the equipment and scale you’re working with. If you skip that, you’re looking at wasted money, delayed launches, angry clients, and maybe even a product that doesn’t work. As a formulation supplier, my job isn’t just to give you a lab recipe that sounds good—it’s to make sure that recipe works when it’s 1,000L, 5,000L, whatever you need.
If you’re gearing up for your first scale batch, or you’ve been burned by a bad scale-up before, hit us up to chat. We’ve got the hands-on experience, the pilot equipment, and the process tricks to make your scale run smooth, no chaos, no wasted materials.
0.5-2.5m2 Stainless Steel Holder References:
- Paul, R. G., & Fema, C. S. (2010). Formulation Scale-Up: Principles and Practices. Journal of Pharmaceutical Sciences, 99(11), 4425-4447.
- Aulton, M. E. (2017). Pharmaceutics: The Science of Dosage Form Design (4th ed.). Churchill Livingstone.
- Rieger, M. M. (2018). Emulsion Scale-Up for Cosmetics and Personal Care Products. International Journal of Cosmetic Science, 40(3), 289-298.
- Food and Drug Administration. (2022). Guidance for Industry: Processing Parameters and Controls for Pharmaceutical Dosage Form Scale-Up and Postapproval Changes. U.S. Department of Health and Human Services.
Hangzhou Guidling Technology Co., Ltd.
Address: No.795, 18th Street, Qiantang New District, Hangzhou City, Zhejiang Province, China
E-mail: export1@guidling.net
WebSite: https://www.guidlingfiltration.com/
