How long does a PIM implementation take? Three phases

What this episode covers

The seventh episode takes on a question whose honest answer is “it depends”, and explains what exactly it depends on. Karolina breaks an implementation into three phases, shows which steps tend to stretch the timeline, and gives ranges from the team’s experience.

Key takeaways

• The minimum is five months including the analysis. Complex projects take a year or longer.
• Pre-implementation analysis: around two months. Implementation: three to six months, sometimes more.
• Integrations are the most time-consuming stage, and the time grows with the number of connections and the transformations the data needs.
• Data migration takes as long as the current state of the data allows. One source is quick; a dozen files and drives is a tidying project.
• The third phase, post MVP, is continuous: integration tuning, training for new users, and new features from live requirements.

Phase one: analysis

A series of meetings between the implementation partner and the client: examining needs, reviewing the systems that hold product data today, analysing processes and defining the goal. On that basis come a system recommendation, a budget and a timeline. Around two months, longer for more complex processes. The previous episode covers this stage in detail.

Phase two: implementation

Three to six months, sometimes longer, and it is worth knowing what that time consists of.

It starts with a kick off, a formal confirmation of what the analysis established. Then installation, or access to the purchased licence.

Next comes modelling the product catalogue structure, meaning the design of the data skeleton: attributes, attribute groups, product templates, categories, channels and the rest of the configuration. This is the foundation everything else stands on.

Integrations can begin during modelling: pulling data from systems such as an ERP and sending it out to e-commerce. This is the most time-consuming stage, and its length depends on two things: the number of integrations and the number of transformations the data needs on the way. The more information flows and the more it has to be reshaped, the longer it takes.

An optional stage is adapting the system to specific needs, meaning features absent from the base version. That extends the timeline too.

With the structure in place and integrations roughly ready comes data migration, sometimes called hydrating the system: technical specifications, marketing copy, images, documentation. The pace depends entirely on the state of the data today. One source makes this a smooth stage. Data spread across many files and drives means tidying it up first.

Finally accounts, permissions and training for the people who will look after the product base.

Phase three: maintenance and development

Also called the post MVP phase. Work that was not high priority at the start comes back, integration tuning arrives, and there is ongoing care that the system grows correctly.

This is a continuous process: the PIM system becomes part of the company’s system architecture, so it needs periodic reviews, training for new users and new features requested by the users themselves.

The conclusion

Three phases, and the total depends on the company. From the team’s experience: an implementation including the analysis takes at least five months, and in more complex projects a year or longer.

If you are planning such a project and want to estimate your own timeline, book a free consultation. Our offer is described on the PIM systems page.