Clicking expand / collapse on each point below will reveal the detail.

https://ngec.saas.nipltd.com/assess/

Certainly no technical reason we can’t use it again and all the past data is already there, so we’ll assume so.

Confirmed that will will use the existing site.

We had to get things sorted with CA IT department so that the site was accessible and worked properly, do we need to check if that’s still in place?

Confirmed that CA should be fine, and Steffi has confirmed that the MOD can also.

Outstanding items were:

• Whether it’s worth linking the VCs to the SERs
• Should we use the one-parameter-per-page option

Steve B is in touch with Steve F & Steffi on point 1 – think it’s mostly resolved.

Steve F liked the one per page – he asked if comments could be made larger, but that shouldn’t stop us using it.

As discussed the first thing to decide will be what characteristics about the participants we should collect to be used in the reports. These are things like job title, site, functional level etc. that are used to categorise the responses and shed more light on the breakdown of scores and comments.

For this number of respondents we might recommend up to 5 could be used sensibly, but each one only makes sense if it would be useful.

When choosing the characteristics it’s worth making sure:

• the information can be easily collected about everyone
• there is a common set of values across everyone for each dimension

As an example, in the past with you we have split respondents up into mostly regional groups (SW, S&NI etc.) and did have a single ‘Role’, but the roles were not able to be compared between DIO and CA.

This information needs to be agreed before the data is collected as you need to know what to collect. It’s also needed by us at that early stage as we can then setup the reporting.

Steve has proposed a field called ‘Job Family’ from the CA data, however it’s unclear yet if a corresponding field exists in the DIO respondent data.

The normal reports (both graphical and comments) are available as soon as one response is received.  In order to create and send these out usefully it’s worth considering who would benefit from seeing them, what they would be particularly interested in seeing, and to which sub-set of the data (if appropriate).

For example, there may be people who would benefit from seeing participation stats whilst the diagnostic is open, analysts who are only interested in their region or site, or those with overall responsibility who would just want summary information.

We need to know who needs to see the reports when, and to which sub-set of the data, before the diagnostic begins as it will affect how we choose to represent the user data mentioned above in the system.

Reports should be in Regional groupings (and Leadership team), and overall.  However, for the most part we won’t need to be too precious about making sure authorised report viewers get access to reports they shouldn’t be allowed to see – it’s largely assumed it would be fine if they saw everything (anonymity of individual respondents aside).

Once the dimensions are chosen we will give you a template spreadsheet to capture this information and then send to us.

You might collect this information from disparate internal sources, so we may be able to help in re-formatting and combining this information. We would treat that as a separate piece of work and quote accordingly.

You can send us examples of the spreadsheet you’re collecting so we can advise if the format is being followed correctly, but we will not upload these versions to the system. We will need the final version in one spreadsheet by the day before the invitations are intended to go out.

We have the CA respondents, which seem in good order.  DIO respondent data still outstanding.

When will the invitations go out, how long will respondents have to respond and when is the deadline.

This should be chosen in light of the data collection process outlined above. We would normally suggest 2 weeks for the length of time to allow responses.

The kick off date is still unknown, but it should be open for two weeks.

The default is 7,3,2,1 & 0 days before the deadline, and is fine for most circumstances.

We will use the default.

It is relatively common to allow a short period after the deadline for stragglers, this often gets a small but significant amount of extra responses. Whether this is done or not can be decided when the deadline arrives, but it is something to bear in mind when setting the deadline. It’s perfectly fine to decide to not do this whatever happens.

Unless there is a surprisingly high response rate we will extend for a week (but only inform respondents at that point).

We suggest that you send an email introducing the diagnostic to the respondents. This means their first point of contact won’t be an email from an unknown source, which they might be inclined to disregard without due consideration. The email should at least include:

• some big picture context
• the specific reason(s) why they are being invited
• ‘what to expect next’, that they will be receiving another email, this time an invitation to participate from the system, with login information and details of how to complete the questionnaire.

You can see and edit the letter that was sent introducing the process last time here.

Sending this email is also an opportunity to check the email addresses you have collected as our experience suggests that there will inevitably be a proportion that bounce. Usually this is because a typo has been made in data collection or that the person has recently left the organisation. These corrections to email addresses can then be used to update your spreadsheet of respondent data (correcting email addresses or removing people from the list).

Note – this step should be done before you send the final spreadsheet of names, email addresses etc. to us so that you can incorporate the information gathered here into that spreadsheet.

Steve F has sent a proposal, everyone is fine with that.

The text that respondents will see when they are contacted by the system with the instructions to participate. This should typically include:

• some more ‘what to expect’ information – how the diagnostic is organised, likely time commitment etc.
• more detailed generic explanation of how to respond to each criteria.
• the deadline date for completion.
• if possible, reassurance that their investment of time is of value and will make a difference.

You can see and edit the previous versions here.

This will need to be agreed at least the day before the diagnostic starts.

I’d asked Steve B to give there a once over – Steve F and Steffi seem broadly happy, though they have them and could revise.  It probably needs checking to update e.g. references to Tim even if the rest is fine.

As the diagnostic progresses, we will need to communicate back to you:

• Participation statistics
• Reports as they stand (if appropriate)
• Any issues that need your input, typically:
  • Bad email addresses (see below)
  • Are specific reasons for declining appropriate
  • Questions respondents ask in emails to us about the business context and process

At least one person should be nominated to us before the diagnostic opens to deal with this.

It will be Steve F and Steffi.

When we send out the invitations we will see any email addresses that bounce.  These bounces mean that the email address is incorrect and the potential respondent will not receive the invitation. At that point we can either:

1. Collect those and pass them to the nominated contact who can then investigate the addresses and send us corrections. We will then send out new invitations to the corrected email addresses.
2. Cancel those respondents.

We would suggest number 1 as most mistakes in email address are typos that are easily corrected once spotted, however at this stage correcting email addresses and re-sending invitations is a time consuming process. As a result if there are more than a handful of corrections we will need to charge on a time & expenses basis (minimum £450) for option 1. That is why we strongly suggest above checking the email addresses yourself so that any bad email addresses can be found prior to this stage.

The decision to choose option 1 or 2 can be taken when the number of bounces is known.

We will use option 1, passing bad emails back to Steve F & Steffi.