How technology supports accurate & compliant classification for food & drink

Tarifftel- How Technology Supports Accurate & Compliant Classification for Food and Drink

0:03
Good morning, everyone, and welcome to today's webinar, which is on how technology supports accurate and compliant classification for food and drink, hosted by our professional affiliate members, Tariff Tell. First of all, thank you to everyone for coming today.
0:17
We hope that you find this content useful.
0:19
There will be a Q &A at the end, so please put any questions that you have into the question box, and we'll do our best to go through them all at the end.
0:26
We'll also be doing a few polls, so please do engage and put in your answer when they come on the screen.
0:31
Following the webinar, we will be sending you an article email in the next couple of working days and this will contain a recording of the webinar, a copy of the slides and the contact details of today's presenter.
0:42
So without any further ado, I will hand you over to our host of this webinar, Elizabeth Davies.
0:49
Thank you.
0:50
Good morning.
0:51
My name's Elizabeth Davies.
0:53
I lead a team of classification experts at TariffTel.
0:57
We specialize in customs classification.
1:00
TariffTel has been around for over 15 years.
1:04
We're a UK company based in Molybdenum in central London.
1:09
And today I'm going to speak about how technology supports accurate and compliant classification, specifically in the food and drink industry.
1:17
I'm going to talk about the importance of detailed information in customs classification.
1:24
We have some very high profile clients that we've been working with within the food and drinks industry, like Martin Spencer and ASDA.
1:32
We've worked with these clients for over 10 years, providing a high level of customer service, classification, data accuracy and data storage.
1:42
So I've got a few slides to share.
1:44
Hopefully you might learn something or pick up some hints and tips to help you with your classification or show you how we can help you.
1:53
As we say, every day is a school day in food and drink classification.
1:56
you know we're constantly learning and picking up new hints and tips.
2:04
So just a little bit about myself, I've worked at TariffTel for four years now.
2:09
I've actually worked in logistics supply chain and customs for over 30 years and makes me feel quite old.
2:17
I've did 15 years of these at Siemens, so I've worked across many industries including automotive, aerospace, transportation, retail, including food.
2:27
We've been working with the FDF for the last six months.
2:31
This is our first webinar together.
2:34
So we're gonna share some information with you about Tariff Tell.
2:38
And some of you may have met us already at events.
2:40
We've done quite a few food and drink events this year.
2:47
So, I mean, we regularly conduct audits for potential new clients across various product ranges.
2:53
This can be across different industries.
2:56
we find on average that 40% of the tariff codes are inaccurate.
3:02
So even companies that are quite confident with their classification codes, we're finding this level of errors.
3:09
Clients can be underpaying or overpaying duty.
3:12
There is a severe lack of understanding in classification and the processes surrounding it due to the data that's available at the time or the data that people think they need to classify a product.
3:24
So we're going to run a first poll now.
3:28
You know, how accurate do you believe your product classifications are?
3:32
So we've given you a few options there from poor, you know, up to 40% to excellent above 90%.
3:39
So if you could sort of vote, then we can sort of get an idea of where you think you are with your classifications. So just wait for the results to come through.
4:03
Okay, so it's looking that most of you think you're good, sort of between 65 and 80 percent, a few very good and a few excellent and a few poor.
4:13
So the majority over 50 percent are you think that you're in the good range, okay.
4:18
So we're going to do that poll again at the end after you've listened to the webinar, just to see if you think you may have changed your mind based on the information that I'm going to give you. So classification, where do we start?
4:36
I mean, You can start with the tariff book.
4:38
It's a good place to start, but not the only place.
4:42
The tariff book is a crucial requirement for regulating international trade, insurance compliance with customs laws, and standardizing the classification of goods using the harmonized system.
4:54
It helps the government collect revenue through import duties and taxes, providing domestic industries by imposing tariffs, and enforcing trade agreements.
5:04
Additionally, controls restricted or prohibited goods, prevents fraud and enables accurate trade data for economic planning.
5:17
Who uses the tariff book?
5:19
You know, businesses, import, export traders, you know, using its classified goods, calculating your landed cost, again, insurance compliers, customs officials, border agencies, again, checking the tariff code, checking the valuation, checking the rules of origin have been followed, freight boarders and brokers often using the book and often using the other other other so you know not really focusing on the duties but more just getting it through customs and government policymakers trade authorities you know legal and tax professionals so you've got you know a lot of people using this tariff book so I would say it's a good place to start and but it's not the only place and very difficult for detailed food and drink, you know, it's down to interpretation, you get varied results.
6:13
If you use the search functionality, I've just used example here steak, I could have used meat, which could give me a little bit more information, but you know, could use beef.
6:23
So, you know, you'll get varied results depending on the complexity of your product.
6:28
But unless you, you know, you understand the full customs classification process, it's likely that you will be incorrect.
6:37
And it's very difficult to classify food and products easily with the tariff book.
6:47
So we're just asking now poll two before we go into the detail, you know, what is your current classification process?
6:54
Is it manual? Are you using Excel spreadsheets, for example? Do you outsource that to a third party?
7:01
Do you have a technology solution currently in place?
7:04
Are you relying on suppliers or is it not applicable for your business because you're not actually doing the classification?
7:12
We'll just wait for the results to come through so we can see.
7:31
Okay, so not unsurprising there that we've got the largest percentage using a manual process.
7:37
We see this in all sizes of organizations, even large enterprises using Excel spreadsheets, and then a small percentage, 20% using technology solutions, so that's good.
7:50
So yes, that's quite interesting for us to know.
7:55
So I'm going to talk about what data and knowledge sources you need.
7:59
So again, often people think, you know, we need the tariff book.
8:04
We just need a description. We'll use the tariff code we used last time.
8:08
It might have changed. You don't know where it originally came from.
8:11
There's no process of where it came from. There's no audit trail.
8:16
You know, companies of all sizes, classifications and afterthought, they don't want to employ people to do it specifically.
8:23
It ends up in logistics or purchasing or merchandise or a shipping clock.
8:28
These people are untrained or they're under extreme time pressure with their current role.
8:34
They did have time to look into classification.
8:37
They don't really understand how to do it or all the places that you need to look for the data.
8:42
Well, we said a lot of companies are using spreadsheets and these spreadsheets become corrupt.
8:47
You lose data.
8:48
They crash.
8:49
unmanageable because they're just so huge.
8:53
You know nobody is checking the tariff book, we say you know it's like checking your own homework, you know somebody's doing the homework and checking it off themselves, there's no verification of that code.
9:03
So I mean what we're showing you here is you know and some of the examples of products that I'll show you, you know you wouldn't just get that tariff code from the tariff book or from the description, you know you need the detailed product information, you need to be looking at the WCO explanatory notes which you have to subscribe to.
9:22
If you're reading the chapter and section notes, applying the general rules of interpretation, checking your rulings, you know, either in Europe, the UK or the US, reading the EU explanatory notes as well and keeping yourself informed and up to date with stop press, you know, reading classification guides, watching webinars, you know, all this detail needs to be looked at for you to get the right code.
9:46
It's really important that, you know, people understand that because your full responsibility if you're the importer of record, you can't rely on somebody else just to give you a code and you don't know where that code's come from.
10:03
So if we look at food and drink, again, you know, what people think they need, just look at an ingredients list or a product description and you'll see the examples of products that I'm going to show in a minute or where you'll fall over if that's all you had, you know, you need this detailed ingredients breakdown.
10:22
If you've got a fat percentage, you know, where is that fat coming from?
10:25
You might need to know if it's palm oil, if it's butter, you might need to know the percentage of milk fats.
10:31
If you've got a fish cake, you might need to know, you know, what fish is it, what species, is it minced in there, is it flaked, is it, you know, what's the state of the product, is it fresh, is it frozen, the weight of the product, you know, if you've got chocolate spread, how it's packaged, you know, different questions for different products and it's understanding what questions you need to ask for which products.
10:54
I mean, this can be very complex just for individual products, but when you start looking at composite goods, you know, such as ready meals, then it, you know, it gets even more complex, you know, where do you start?
11:05
You know, where do you know, how do you know what questions you need to ask and in the right order.
11:16
So again, we're just reiterating here, you know what you need to do is build your knowledge base you know analyze the tariff book read all those rules and regulations check the rulings know your product like we just said detailed product information and keep up to date you know keep up to date with the tariff book changes subscribe to stop press and information make sure you interpret those changes sometimes it's not just a code change it could be a wording change which changes where your product goes You know the major updates in the tariff book usually every five years.
11:48
I think the next one is 2028 But the tariff book does change every year, you know, and it has changed quite a lot this first quarter with small changes So we're going to look at some products now as examples really of what I've been saying If we look at a yogurt here You know, you've got a natural yogurt here not flavored no added sugar, fat content 5%, that contains two types of bacteria, I'm not going to attempt to say them.
12:21
And then you've got another product, which is a probiotic yoghurt, again, not flavoured, no added sugar, same fat content.
12:28
But because it contains this additional bacteria, it's not a true yoghurt, according to the tariff book, so it falls down into other yogurts and this information is only in the EU explanatory notes so if you haven't read those then you might not understand that you would be non-compliant on this product.
12:52
Another example, Easter eggs, if you've just gone through this, this is an example where you can't always apply the same GRI or the same rule for a similar product.
13:01
If you look at the top example we've got a milk chocolate Easter egg and it's got an assortment of milk and dark truffles inside it.
13:10
So for this one, you know, the chocolate Easter egg is the star of the show. It's the predominant white product. So we would be looking at the egg.
13:18
So it would be looking at the non-filled chocolate product.
13:23
But if you replace that chocolate egg with a white chocolate egg, but you've still got those same assortment of milk and dark truffles inside, you kind of play the same role because straight away white chocolate is not chocolate.
13:36
So you would be in 1704, chapter note 1a would tell you you can't be in 1704, so you're back in 1806.
13:45
So now you'd be looking at the chocolates inside the product, so you'd be going to bite-sized filled chocolates.
13:52
So very similar products, different rules, different ways of applying the GRIs that need to be understood.
14:02
And we look at a drink product here, we've got Sangria. Sangria can often be mistaken for red wine.
14:08
There's a few different headings that this could go under to begin with, you know, it's made predominantly using red wine, should it be in 2204, it's a flavored wine with plants or aromatic substances, should it be in 2205, or it's this fermented beverage that's got other additions, do we look at 2206?
14:30
So there's three different areas there that we would look at to start.
14:35
So the fact is that Sangria is not red wine and if you read the WCA explanatory notes it excludes beverages with the basis of wine which is Sangria so that automatically would make me fall into 2205 and then we need to look at 2205 tells us that we can only be here if we've got an ABV percentage of 7% or more and this one is 5% so we can't be so we're now going to drop into 2206.
15:05
So you can see again it's going through this process of questions and different information you know WC explanatory notes, EU explanatory notes, the tariff book, you know you need to be abreast of all this information.
15:23
Last but not least the stake and this is showing how important the customs description is.
15:30
So if you look at these two products, you know, they are exactly the same description and they are exactly the same ingredients, but they are classified differently.
15:41
And this is due to the packaging.
15:44
So if we look at the top product, you know, the sirloin steak is packaged separately from the garlic butter.
15:50
So it would be classed as a non-prepared meat product.
15:53
And if you look at the product below the meat, it's got the garlic butter on top of the steak, that would be classified as a further prepared 1602.
16:03
And there is a difference in duty there.
16:05
So same product here, same product description.
16:08
It's only because we know to ask that question and we know to look for an image to know that there could be a difference in classification.
16:15
If you're importing a lot of these stakes, you could be paying a lot of duty there that you don't need to pay.
16:26
So we'll just look at tariff tail briefly.
16:30
What is the solution?
16:31
How can you ensure You've got all the data you need at the right time.
16:36
How can this technology help you?
16:38
How can it make sure you're following all the classification rules or guidance, using all these right data sources and having the customer's expertise?
16:47
Solution like TariffTel can help that.
16:50
We believe that the answer is to collect that data at source.
16:53
Your suppliers know their product, they manufacture it, they understand it.
16:57
Are you utilizing your suppliers to collect the data that you need, As long as you can identify what data needs to be provided, then they can provide it.
17:07
They're a valuable source of knowledge.
17:09
You don't need to have any classification expertise when using TariffTel.
17:14
All you need to do is the data of your product.
17:18
So TariffTel is an online classification system.
17:22
We offer a unique and comprehensive approach to tariff classification to ensure businesses to remain compliant and efficient in today's complex trade environment.
17:32
We provide seamless integration options tailored to different business needs, from document upload with Excel to fully integrated systems that deliver valuable data across organizations.
17:45
So we take data from customer systems, either from their master data or the purchase orders.
17:51
We can then display that information to suppliers and then ask all the right questions we need to ask on those products or we can take the information from the master data and work on that missing data with the client to enrich that data.
18:07
We also store images and associated data, any documentation, maybe export health certificates, nursing codes, so it's all that product data is all held in one place in a centralized system, good preparation for when the digital product passports introduced, and then we feed all that data back to wherever our customers need it fed back into maybe their ERP systems, maybe they want it fed to their brokers, you know, we can feed all that information back.
18:36
So the key is they don't need any classification knowledge, they don't need to look up any data, you know, we take the data, we build these decision trees in the background, and we have this two-step verification, so it collects the data, it assigns the code, and then it's signed off by somebody else. Again, somebody else is checking the work.
18:57
It's a fully auditable system.
18:59
You can ask questions and answers and communicate with your suppliers or clients or with ourselves through the system and it records all that data.
19:08
So who provided what data and when, who answered what question.
19:12
So, you know, if you get an audit from HMRC, you've got all that data to hand at your fingertips.
19:17
You haven't got to go and search for any information anywhere of how you obtain this commodity code, you know, and who classified it.
19:23
And this record keeping is really important.
19:26
It's a documented process of how you reach to classification.
19:31
So it's simple, reliable and efficient.
19:34
And we make sure the tariff codes are kept up to date constantly.
19:37
And we inform our customers in advance of tariff codes changing and ask additional information of those tariff codes which I'll show in a minute with the tomatoes.
19:47
So our customers, they don't just purchase the system, we support the system with our classification advice and concerns, you know, when tariff codes get challenged, you know, we can provide all the data of why that tariff code has, you know, been calculated.
20:01
You can see on our website, a case study with Mark's Lens of Food, because we do all their food and drinks, so if you wanted to have a look at that.
20:10
So if we just look at the steak again, you know, the way that we would classify it, obviously, the data comes in the system, the supplier would log in, we would ask the data on that product, so we would ask if the garlic butter was separate on the stake or not, because we know to ask that question, and then the code would be produced, and then the code would be approved, and then the classification is complete.
20:36
So very, very simple, methodical process, minimizes errors, ensures compliance and enhances operational efficiency and this is how effective our system is.
20:53
So just talking about the tariff code changes tomatoes had a fat shaming at Christmas.
20:59
If you look at the top of half of this slide you know tomatoes just used to be cherry tomatoes or not.
21:05
Now they want to know the diameter of the tomatoes and if the tomatoes are on the vine or So obviously we knew this was coming at Christmas.
21:14
So we work with our clients to go through all their tomatoes and make sure that we've got all the information we needed in advance.
21:22
So if we needed to then go and find out the diameter of the tomatoes or go to the supplier and find this detail, we could get all that information in advance and then get everything reclassified so that when our customer starts work on the 1st of January, all their commodity codes are up to date and nothing gets stuck in transit anywhere, you know.
21:41
The misclassification can result in shipments being stopped or delayed, financial penalties.
21:47
You know, if you've got fresh produce, you don't want to be losing that because of a code.
21:56
So AI, we'll talk a little bit about this.
22:00
You know, how good is AI with classification?
22:03
A lot of people are using it as a bit of a buzzword at the moment.
22:08
I think for food and drink, it's extremely difficult, as you can see from the examples that I've provided, you know, if you're buying a lipstick, it might be fine, but if you're buying lemon tea, you might get lemons.
22:19
If you think about the steak, which is the same description, what commodity code would you get?
22:25
You know, the data is the backbone of accurate product duty classification.
22:31
It enables us to make informed decisions about the tariff code.
22:34
You know, we need this detailed product information.
22:38
You know, is AI going to ask you where your facts come from, because you're going to know it needs to ask about milk fats or palm oil.
22:45
AI, it does play a significant role in enhancing our capabilities by gathering and analyzing these large data sets.
22:53
But it's not a standalone solution.
22:56
You can see from the complexity and the nuances of the customs regulations that you need that human expertise and judgment.
23:05
If you search for baby food in AI, are you going to know you've got to look for homogenized food.
23:10
You know, AI is only using the data that it's fed.
23:15
You know, so we need to be careful and be cautious.
23:17
You know, if you, it's like driving up a one way street the wrong way with your sat nav and the police officer stopping you and giving you a ticket and you say, well, my sat nav told me to do it, told me to drive up here.
23:28
You know, HMRC will have the same response as that police officer with your AI, Christmas classification.
23:33
If you've just typed in something and got a number than that. It's not good enough, you know.
23:40
So it's really important, you know, like I said, bad data in, bad data out. How much data, bad data, have these AI solutions been fed?
23:49
There's a Harvard Business Review that was conducted in 2018 to say bad data can rear its ugly head twice, first in the historical data used to train the machine learning, and second in the new data used by that model to make these further decisions.
24:05
So I would be It can be used, but think about your type of products that you've got.
24:11
Are they simple enough to be able to get an accurate code from AI?
24:18
So we're going to go back to the first poll.
24:21
Just to be interesting, do people still feel that they're in the right place with classification?
24:28
Just from the information I've provided.
24:47
Just waiting for those results to come through.
24:59
So yes, we've still got a fair chunk in good, but probably a few more in average than we have before.
25:06
So maybe some of you are not feeling as confident as you were at the beginning.
25:16
And that's it really.
25:17
That is, you know, a little bit of classification hints and tips there and a little bit about our system and what we do.
25:25
I'm happy to answer any questions now if anybody's got any specific questions for me.
25:32
I'll just check the chat.
25:46
So I've got a question here about data.
25:49
How can I ensure my data is safe and secure?
25:53
A solution like TariffTariff, it's a good question.
25:56
We have a broad and multifactor approach to security to keep our clients' data safe.
26:02
We've all heard the news lately.
26:05
All our data is partitioned, backed up, and access controlled.
26:09
So we're CyberEssentials Plus accredited, the UK government standard.
26:14
So when we onboard a new client to our system, we ensure they're onboarded properly and we maintain that high level of security.
26:25
Looking at another one.
26:27
How easy is TariffTel to implement?
26:31
It's very easy.
26:32
We have standard APIs to and from our customer systems, whether you're Oracle SAP or any other ERP system.
26:40
but we can also upload or extract data with Excel if people don't want to go down the API route.
26:48
We can engage with suppliers or we can not do that.
26:51
We can just work with your master data system.
26:53
There's different options available depending on the business requirement.
27:03
Then somebody's asked here about what features are important.
27:06
We're looking at technology solutions.
27:08
Any solution that you're looking at, whether it's TeraFTA, whether it's AI, I would look at the percentage of accuracy that solution offers you.
27:19
Is it showing you the clear process and the method of getting the classification?
27:24
Is it giving you a clear audit trail to show you how that classification is achieved?
27:29
How is it gonna maintain that classification and ensure you're getting the right code every time, making you compliant and making sure you've got a clean bill of health?
27:41
I think that's all the questions that have come through at the moment.
27:52
Okay, and that's it.
27:53
I mean, obviously, you can get in touch with us through our website, or you can message myself directly.
27:58
The slides will be sent out in links where you can get in touch with us if you want to talk to us about classification.
28:09
Cool, thank you very much, Liz.
28:12
There are a couple of questions that are in there, but if there are any others that we aren't able to cover, then please do get in touch with Liz and the team afterwards.
28:21
Again, we'll be sharing the recording with all of you guys after the webinar, along with a copy of the slides, and Elizabeth's contact details for any further questions.
28:30
So yeah, Enes, there's anything else from you, Liz? No, that's it. Thank you.
28:35
Cool. Okay. Well, thank you very much, everyone.
28:37
I hope you have a great rest of the week, and we'll see you in another webinar again soon. Goodbye.
28:42
Bye.