4 Tips for Becoming a Better Data Migration Client

It’s all too easy to blame your data migration solution provider when your project hits the rails but are you really doing everything on your side of the fence to ensure project success?

Use these tips as a starting point for determining what your company should be doing to ensure project success for all concerned with the implementation.

1. Be Realistic

Several years ago, whilst working for a consulting group, I was approached by an existing major client who wished to have an external specialist migrate substantial volumes of highly complex data.

The challenges were varied:

  • 20-30+ systems, some of which went back 15 years, client had no idea how many systems would be required
  • No real data quality measures aside from finger in the air guesstimates
  • Complex political “turf-wars” and lots of power plays
  • Fixed deadline for completion due to regulatory timelines
  • Requirement for a fixed price implementation
  • Strong desire to offshore the development and slash costs across wherever possible

So, outwardly this may have looked like the dream project right? Big name client, multi-million pound contract, bonanza time! Sadly, no.

The client had actually committed the first rule which is to be totally realistic. Some simple calculations based on even optimistic time scales and resources revealed that it was physically impossible to migrate the volumes of data they were expecting, simply from a complexity perspective.

Even if your pockets are deep, do not assume that simply throwing huge amounts of cash at a problem will make it go away.

Be totally realistic about what can be achieved. A lot of consultancies will tell you what you want to hear because they have their goals to consider, not yours.

2. Measure twice, cut once

If you’re looking for external help with your data migration it pays to take the initiative in a number of areas, requirements management being the most critical.

I’ve seen too many projects kick off with only some vague notion of what measures will determine project success. You need to take the steer on the exact requirements your service provider must fulfil.

A classic mistake is to leave the issue of data quality floating precariously between both parties. Many system integrators will offer to measure and assess the data but cleansing and improvement will typically be additional items, the classic “data quality get-out clause”.

You as the client must take data quality by the scruff of the neck and measure it, define it, scope it and negotiate price on it accordingly. Get data quality at the forefront of your project strategy from day 1 and ensure it is fully measured and understood.

3. Grant access to expertise

One of the biggest obstacles to success on any data migration project is the failure to have the right type of legacy or target skills available when required.

As a client this is firmly your responsibility. You need to ensure that Jeff, the 15 year vet who developed that crumbling CRM system from C and Informix back in 1995 will be on hand to decipher those 4GL function definitions that have never been documented.

The same applies to the target system. Many integrators will not be responsible for the new application environment but they will need specialist knowledge on how to load data into the API or back-end database. This is your job as client to provision this knowledge.

4. Structure the Project for a Win-Win

Most migration projects fail or overrun. With these statistics ringing in the clients ears there is little wonder why many contracts and project relationships are based around a punitive strategy.

“You will deliver this project or else” seems to be the way forward for many clients who are (rightly) nervous about the future success of the project.

The key here is to look at the project as a real collaboration. If you were entering into a business partnership you would focus on building a culture of sharing the victories, not regular beatings for missed opportunities.

  • How can you structure the contract so that it is based around a reward strategy?
  • How can you set up the team environment and culture so that there is less “them and us” and more “we’re in this together”.

Have you ever been a data migration client? What did you do to help ensure the success of the project and beat the statistics? Please share your views below.

Comments are closed