6 replies to this topic

[mm.stf]

Hi everyone!

 

I’d say we’re a very hyped plant-based protein powder and supplements brand, which also means we have a very particular picky type of customer.

 

Every now and then, we receive customer complaints about the flavour or texture of a product. For these types of complaints, our current procedure is to run a sensory evaluation on a sample from the batch in question. The thing is, 10 out of 10 times, there’s absolutely nothing wrong with the product — it ends up being a bit of a waste of time. We already run sensory evaluations during production, before authorising packaging of the batch, so if there’s any non-conformance, we catch it then.

 

Recently, this topic came up because, in theory, something could go wrong during packaging that affects the sensory profile — although this has never actually happened before.

 

What would you do?
Do you think we should implement a trigger before testing a batch — for example, “if we receive at least two complaints on the same batch, then we test it” — or should we keep things as they are?

 

Thank you!

Edited by mm.stf, Yesterday, 09:20 PM.

[GMO]

Hi everyone!

 

I’d say we’re a very hyped plant-based protein powder and supplements brand, which also means we have a very particular picky type of customer.

 

Every now and then, we receive customer complaints about the flavour or texture of a product. For these types of complaints, our current procedure is to run a sensory evaluation on a sample from the batch in question. The thing is, 10 out of 10 times, there’s absolutely nothing wrong with the product — it ends up being a bit of a waste of time. We already run sensory evaluations during production, before authorising packaging of the batch, so if there’s any non-conformance, we catch it then.

 

Recently, this topic came up because, in theory, something could go wrong during packaging that affects the sensory profile — although this has never actually happened before.

 

What would you do?
Do you think we should implement a trigger before testing a batch — for example, “if we receive at least two complaints on the same batch, then we test it” — or should we keep things as they are?

 

Thank you!

 

I think there are two things I'd look at here.

 

For quality complaints which have a risk of being a perception issue (i.e. they don't like the product but it's as intended) then I do tend to focus on adverse trends.  So rather than investigate every one, I'd investigate if there was a specific spike and write that into the procedure.  I'd also look to how you measure things like texture (which ideally you should not just organoleptically if possible) and look to whether the extremes of "in spec" are actually organoleptically acceptable.  Lastly I would make sure there were regular checks to ensure things are really being made to spec.

 

BUT...

 

If you have a particular trend on texture or flavour, sorry to be blunt but perhaps your standard isn't great?  Perhaps people don't actually like it?  How do you know that whatever texture or flavour you're targeting is the best one?  Have you benchmarked with professional taste testers vs. the competition?  Have you tried recipe tweaks?  Perhaps it is worth passing that back to your NPD team (once you're 100% sure no mess ups are happening in production) to see if they could optimise the recipe?

[mm.stf]

I think there are two things I'd look at here.

 

For quality complaints which have a risk of being a perception issue (i.e. they don't like the product but it's as intended) then I do tend to focus on adverse trends.  So rather than investigate every one, I'd investigate if there was a specific spike and write that into the procedure.  I'd also look to how you measure things like texture (which ideally you should not just organoleptically if possible) and look to whether the extremes of "in spec" are actually organoleptically acceptable.  Lastly I would make sure there were regular checks to ensure things are really being made to spec.

 

BUT...

 

If you have a particular trend on texture or flavour, sorry to be blunt but perhaps your standard isn't great?  Perhaps people don't actually like it?  How do you know that whatever texture or flavour you're targeting is the best one?  Have you benchmarked with professional taste testers vs. the competition?  Have you tried recipe tweaks?  Perhaps it is worth passing that back to your NPD team (once you're 100% sure no mess ups are happening in production) to see if they could optimise the recipe?

 

Unfortunately we do measure texture organoleptically only, we don't have an in-house lab but hopefully one day!

 

 

Our customer complaints in relation to our number of units sold is really low: around 0.03% for this year.

We have around 110 different flavours/SKUs and it doesn't seem to have a pattern of a specific flavour people don't like to explore recipes tweaks. 

 

I have been asking to add a second sensory evaluation in the middle of packaging but been shut down as it could impact on production time 

Edited by mm.stf, Yesterday, 10:43 PM.

[jfrey123]

Consumer complaints that don't pertain to food safety get logged and categorized for trend analysis, and like GMO inferred, spikes or adverse trends could be something we look at or relay to product development or procurement teams.  When someone complains about a food safety hazard (alleged illness, fm findings, etc), those get taken seriously and we investigate.

 

I'll say the absolute bane of my existence is when our retail customers decide to share their consumer "complaints" or reviews with us, and they share them in a way where they expect us to do something about them.  I hate them.  From bots interjecting fake reviews to picky consumers who actually take the time to write they don't like the color of a package...  Gah, drives me batty when a customer sends us an excel file with 1,000 lines of "complaints".

[Tony-C]

Hi mm.stf,

 

Normally you would investigate food safety complaints and an abnormal number of other complaints. Your trend analysis should assist you in identifying adverse trends by product and complaint type.

 

0.03% isn’t particularly low but that depends on the nature of the product, value of the product and how much reward the customer gets for complaining 

 

If you don’t actually get that many complaints to investigate then I would continue the sensory analysis as I prefer to satisfy myself that there is nothing wrong with the product. Persistent texture and flavour complaints where you find the product to be acceptable would mean to me that NPD should probably have a look at making some improvements.

 

Kind regards,

 

Tony

Edited by Tony-C, Today, 04:27 AM.

[GMO]

Complaints are normally expressed as CPMU (complaints per million) which would put you, I think at 300.  That's really high even for British consumers (who complain a lot.)  I'd expect for something that should be ultra consistent like your product probably <10 CPMU would not be unreasonable.

 

Do you not even use one of these for assessing viscosity?  They're super cheap but obviously need careful set up and instruction (how long to leave before taking the value, ensuring it's flat). 

 

BBY Bostwick consisometer for Test Consistency of Viscous Liquids Using a Consistometer : Amazon.co.uk: Grocery

[TimG]

I do as others mentioned, log a quick taste/texture complaint for those and look deeper into the ones that merit it. I have completely stopped logging complaints where people can't tell the difference between oz by weight and fluid oz. I had a home ec. teacher argue with me about how I ruined her bake fair because she didn't have enough honey to make her baked goods all because we don't 'sell our product right.' Bottle clearly states it's ounces by weight.

 

Oh, also in case anyone didn't know, if you heat up your maple a few times and then toss it back in the fridge, it will precipitate out what looks like shards of glass. They are not shards of glass, don't throw your maple away