← All Kits · All Guides · Data Migration Series
Stage 1: Kickoff and Scope
A kickoff is not a get to know you call. It is where you define the project, and anything you fail to settle here you will renegotiate later at a worse moment. This article gives you the agenda, the four questions that matter, and the email that closes the stage.
The short version. Leave the kickoff with a written answer to why we are here, what success looks like, what is not coming across, and who your one contact is.
The four questions
Which of these do you think clients find hardest to answer? Guess, then read.
Why are we doing this? There is always a reason underneath the software purchase. A failed audit, a merger, a system the vendor stopped supporting. The reason tells you what the client will care about when a tradeoff comes up.
What does success look like? Push until it is concrete. "All client records available in the new system on day one, with visit history back to 2019, and no client appearing twice" is a success criterion. "A smooth migration" is not.
What does failure look like? Ask this one out loud. It surfaces the thing they are quietly afraid of, and it is usually specific. Losing signed documents. Losing the audit trail. Staff refusing to use the new system.
What are the risks? Name them together: timelines, people going on holiday, an expert who is only available two days a week, a technical dependency you do not control. A risk the client named is a risk you can talk about later without it sounding like an excuse.
The exclusion list beats the inclusion list
Agreeing that seven years of history comes over is easy and it is not the useful half. The useful half is what does not come, and what that means in practice.
Say out loud what happens on the day a staff member cannot find a record you agreed to leave behind. If you cannot picture that moment, the exclusion has not been agreed properly.
Write the exclusions in the client's own words, not in system terms. "Anything closed before 2019 is reference only, which means you look it up in the archive, not in the new system" lands. "Pre-2019 records are out of scope" does not.
One point of contact, not an address book
Ask for a single named person on the client side who routes everything. Meeting invitations go to them. Questions go to them. They forward as needed.
This is not about convenience. Trying to manage a client's internal address book means you are guessing who needs to know, and every guess is a chance to miss someone who later says they were never told. One contact makes forwarding their job, which is where it belongs.
Separately, find out who signed the contract. That person is rarely your daily contact, and they are the person whose attention moves things when the project stalls. You are not going to email them weekly. You need to know who they are before you need them.
What you say about yourself
Spend two minutes, not ten. Who you are, that you have done this before, and how you are going to get them from here to live. Then put the plan on the screen and let it do the talking.
A visible project plan at kickoff is worth more than any credential. It is also the thing you will point at every time someone asks for a change, which is why scope creep is a solved problem when the plan exists and an argument when it does not.
The agenda, in order
- Why we are here, in the client's words.
- What success looks like, written down where everyone can see it.
- What failure looks like.
- Risks, named by both sides.
- What comes across, and what does not.
- Who does what, including the single point of contact and the experts who will validate data.
- The plan, with the milestone dates and the gates.
- What happens next, and by when.
The gate
The stage closes with an email, sent the same day. It states what was agreed, what is not in scope, who the contact is, what the next actions are with dates, and it links the plan. It asks the client to confirm that this matches their understanding, and it copies the people who need to know.
Naming a specific action with a when and a who is not a stylistic choice. Gollwitzer and Sheeran's meta-analysis of ninety four tests found that turning an intention into a stated plan of when, where and how produced a medium to large improvement in whether the thing got done. Your follow up email is that, applied to a client.
Cheat sheet
| Leave the kickoff with | Written where |
|---|---|
| Why this project exists | Follow up email, first line |
| Success criteria, in numbers | Follow up email and the plan |
| What is out of scope | Follow up email, in the client's words |
| Named risks | The plan |
| Single point of contact | Follow up email |
| Who signed the contract | Your own notes |
| Next actions with dates | Follow up email |
The one habit
Do not leave the kickoff without knowing what is not migrating. That sentence prevents more disputes than any other in the project.
What is the question you wish someone had asked you at the start of your last project?
Up next
Stage 2: Profile the Source →Previously: The Eight Stages of a Data Migration · All eight stages
Counting rows, grouping to find duplicates, and joining to find orphans covers most of what a migration asks of you. The SQL Kit teaches those moves in order, with the data in front of you.
Open the SQL Kit →Or start typing straight away: open SQL Drill, thirteen queries that each add one thing to the last.
You are moving a client’s records into a new system, and the part that worries you is the week the client goes quiet. The Data Migration Playbook is all eight stages in order, 63 pages, from the first kickoff meeting through hypercare, so you can run one without finding the important question three weeks too late.
The Data Migration Playbook, $29 →References
- Haller, K., Matthes, F., & Schulz, C. (2012). A detailed process model for large scale data migration projects. Business Information Systems (BIS 2012), Lecture Notes in Business Information Processing, 117. Springer.
- Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119.