HYCS #36 – Which Data Collection Method is Best?

HYCS #36 – Which Data Collection Method is Best?

It’s tempting to think the more data you collect, the closer to the truth you’ll be.  But data isn’t the truth.  You need people talking to each other for that.  It’s best not to overwhelm them with more data than they can manage.

+     +     +     +     +

Reading Time:  3 minutes

Assignment Time:  2 minutes (It’s a thought assignment)

+     +     +     +     +

Collecting data is an art. You want to strike a balance between data that overwhelms and bullies your client, and data that is too skimpy to provide ample fodder for conversation. You want just enough data, and you want a good mix of data.

While more data isn’t always helpful, more types of data are.  I like to use a mix of the data collection techniques below.  Like walking into the same room through a different door, each of the techniques listed below provides another point of view.

But first, I need to limit my inquiry.  I need a focus that will keep my client and I from choking on too much data. That’s why I use a model to guide my inquiry.   If I’m working with a team, I might use the 5 elements of team of a team charter, or a team model like the Drexler-Sibbet model.   It will focus the data I collect and help me organize it for my client’s use.  If my client is also collecting data, a model will keep us working in harness rather than at odds.

 Data Collection Methods

Existing documents – This is where I like to start.  I’ve looked through medical records, nursing notes, meeting minutes, org charts, agendas, the forms people use and the emails they send one another.  There is a great deal of information available to you here.  How readily available this information is will tell you as much as the documents themselves.

Direct observation – Go to the workplace and wander around.  Observe a meeting.  Notice everything – what’s on the walls, where people sit, who talks to who, what they say and do not say.   If you can’t go there, ask someone to send you a video or start a video call on their phone and give you a tour.  If you can’t get that, ask them to walk around and describe what they see.  Ask questions, like “where is the team charter?”  This is the kind of information you can’t get any other way.  A caution:  You don’t yet know what any of it means. You are collecting impressions only.

Drive-by Interviews – When I’m observing and something I don’t understand happens, I find an unobtrusive moment and I ask.  I am careful to maintain a detached, neutral presence when I ask a question and fade into the background as soon as I can.

1:1 interviews – I avoid these unless there is no other option.  When I do use them, I bring the information I’ve gleaned back to the group or pair quickly.  Individual interviews do not help people talk to one another, and, in my experience, talking to each other is what they need to do.

Group interviews/focus groups – It’s a rare focus group that doesn’t succumb to group dynamics.  Good meeting skills can take care of this, but it’s more time consuming than my preferred option, which is to survey first so the group has some data to respond to.

Surveys – This is a great way to get data that’s not limited by a group’s typical way of interacting.  Of course, the downside is they don’t take it seriously.  That’s OK with me:  I’m looking for all the data I collect to spark a better conversation. The survey information can disrupt even the most entrenched group dynamic.  Survey’s also provide the option of running the same survey in a few months to see what has changed.

Two Pitfalls of Data Collection

What does it mean?  A common pitfall is thinking that the data you collect means something.  Data is a trigger for conversation, that’s all.  Your clients are the alchemists, turning the data you bring them into the actions they can take at this time in their organization.  You have not idea where the conversation will lead, and that’s as it should be.  Being the expert on your client’s data can leave you holding the bag for implementation.  That’s not a successful intervention.

Taking too long to collect data and feed it back. The more data you collect and the longer you take, the more likely you’ll be seen as the expert with the truth.  Which makes it harder to transfer accountability to your client.  Best not to hold that accountability in the first place.  Data collection is best done quickly.  I like to limit it to 1-2 weeks.  You should feed the data back while it’s still fresh, 2-3 weeks after you’re finished collecting it.  I’ll cut the data I gather before I’ll extend the timeframe.  I want to get it into my clients hands ASAP.


Consider using one new method of data collection with a client.