A day in the life of a…Product Data & Taxonomy Consultant

As part of our ‘A day in the life of a….’ we focus on the role of the product data and taxonomy business consultant – the person who makes sure the product data taxonomy is organised so well that accessing key product information is simply a question of a couple of clicks

Can I start by asking you role actually does?

Sure. Working on a PIM project, this person is responsible for creating the backbone taxonomy and product attribution as well as supporting the definition and delivery of a new PIM system. It’s a kind of split role between creation of the classification strategy and working with the system engineers, product managers and data management teams to drive this classification to a shared objective among stakeholders. I spend a fair amount of time doing in-depth research, presenting findings and proposals and consensus-building. So in a sense, it’s a leadership role, with a lot of people management involved.

Tell us a bit more about taxonomy.

Basically, it’s a structure used to organize all available products in a way that customers can find what they want in the least number of clicks. Generally, it works on a folder system basis as a kind of product hierarchy with different product categories. You ensure a complete set of attributes, like colour or size, for each product in each category. And this taxonomy schema can be optimised for customers, SEO and internal users.

What’s a typical day like?

I don’t really know, but I’m really busy! Actually, I’m spending a lot of time on calls and video, so I’m doing things like going through blueprints, requirements, the ‘as is’ and the ‘to be’s’. I think that last point is a key factor which we never need to lose sight of. After all, we do have an overarching objective. When a project is up and running, I’m in frequent touch with the data architect, solution architect and the business analyst, to give them a steer on what aspects of the project they might not be that close to. I also need to liaise with stakeholders, especially the data management and taxonomy teams. This is to make sure they’re fully aligned with what I am producing.

Data quality seems to be a key factor with all the consultants I’ve talked to…

Yes. It is very important. I guess you could look at it in terms of product master data. If the PIM system is supposed to be the master data used for product information and those data are being utilised by everyone – teams, business users, systems, applications, programmes. If there are wrong data, they’re going to be wrong everywhere.

So, how do you know how wrong they are?

I measure data quality in a couple of ways. Firstly, how populated the key data fields are. If your KPI is set at the high 90%s and in reality it’s 70%, you have a problem. Secondly, quality is about data consistency. You need clear product data standards. Otherwise, you have different people collecting data from different suppliers using different labelling. For example, someone will enter a measurement as ‘’2 millimetres’. The week after, it becomes ‘2.0 millimetres’ and the week after that, ‘2.00 millimetres’. Understandably, different suppliers may use different data sheets and capture the same data point differently. It’s not wrong information as such, but it’s just not standardised.

…and I guess that can cause issues for the customer?

It can. When a customer goes to a product filter and they want 2 millimetres, they’ll see three different variations. If you’re not careful, these variations may not even be grouped together. So then, the customers are missing products they potentially might need. Either way, it’s poor and that kind of inconsistent data can cost you as a business.

I also look at accuracy. That’s a little bit harder to measure but, from my experience, we can create accuracy metrics by getting customer feedback on wrong data and scoring it. So we might have a total score for high-level data quality, with combined fill-rate (completeness), consistency and accuracy scores. That gives us a kind of overall idea of the data quality.

Is there any pushback from clients regarding the need for this scoring of product data quality?

Well, at the level of legal compliance and data governance, not so much, because I can easily present instances which prove that something’s not quite right with the data. If you explain the traceability of the problem, its fairly clear – for example, we might need to be a little more ‘clever’ with the taxonomy, and for the taxonomy, we need the right product attributes, and before we collect the data, we need some standards, guidelines and protocols. Otherwise, it’ll just be a mess again. They can see the clarity and logic of that message.

So that leaves the data ‘clean’

…which helps the PIM system to be clean from the outset. Then the data migration specialist is in a position to migrate good data into the PIM. It’s a new system, so it’s better and more agile than the previous one, which means you can make changes as and when required.

Is it a challenge reaching consensus with the client stakeholders?

It depends on the size of the company. I’ve worked with really large organisations and at times, it’s seemed like there’s more than one person for some roles. So a key challenge is that there can be a lot of opinions without you being able to get to the bottom of who exactly the decision-maker is. I mean, these are all good, valid opinions, but if we start working on the basis of one opinion and it turns out we weren’t speaking to the decision-maker, it becomes,…well…


Mmmm. I know what needs doing but the trick is, how to get decisions – and who to ignore!…

Do you think that’s a problem of data ownership?

Maybe. It’s like knowing where to go to find the person who is accountable, who’s responsible. In terms of the organisational structure of large companies, that can be a bit time-consuming. As I said before, I can be asking myself “why have they got six people for a job that could be done by one?”. But if they don’t have an efficient, effective and successful PIM, they may need those people. That’s the whole point of getting a good, clean PIM system and more fluid processes – you won’t need so many people and resources. AND, with fewer people involved, you’ll get more decisions and fewer opinions!

What floats your boat about your role?

I have a real passion for data, and I’ve done this for so many years. It’s a bit of a niche, and maybe some bits of what I’ve explained sound obvious. But to some companies that potentially don’t have that ‘data mindset’, what I communicate to them can be a revelation. Organisations often see data as being primarily concerned with business intelligence, but what we do is something separate to that.

I really like working with large data sets. I have a good eye for detail and can spot errors, things that just don’t look right and stand out. This comes easy to me because I’ve been doing it for a long time! Part of the enjoyment for me is making improvements and seeing a clearly positive outcome.

If you would like to find out more about how our product data management consultants can create value for your business, we’d love to hear from you – Ben Adams, CEO Start with Data

We’re always looking for talented people! Read more about our culture, the experience required and our current roles available. We’d love to hear from you – Joanna Hall, Head of Talent, Resourcing & People Operations