Dear Ilaria, I experienced the same case in evaluating a new development model. In fact, there is no much research in model evaluation but I found some publications of Barbara Kitchenham discussing this issue, they are very valuable. Among them is this paper:

DESMET: a methodology for evaluating software engineering methods and tools

Hope it helps

I am happy this question is seeking a documented meaning. 

This (...  of a new method/tool from a user perspective) objective has always made an inclusion in all designs jobs. the only aspect isin the seriousness and sincerity of Effectiveness.

Validation is necessary when things can not be measured.

If such is the case, then validation may be done w.r.t quality or critical parameters defined by user @ his/her own standards. After the user has okayed the delivered for deliverables, getting process honing back over real time should help.

Hi,

I understand how much it is a complex case and has no direct easy answer. Therefore, in such situation, the researcher need to do the following:

- Search meticulously the literature, be lucky if you find some thing close to what you want to measure, if you find it then you use is a reference to compare with (concurrent validity).

- So, you can develop a new measure, {which you plan to validate} and this is the one which you will compare with what you found. Now, if you did not find any,

- you may do Triangulation method of measurement: your quantitative Likert tool, beside qualitative questions.

- you run your data collection, and next step you measure the employee satisfaction and compare it / correlated it with the tool which you developed

- you can also present the new tool to expert in the field to judge it.

- you need to run Chronbach's alpha in check the reliability of the total scale and for each item.

I know it is a long topic which cant be presented in detail here. but I hope the above points can help. ready for further clarifications.

You can try propagating and canvassing your truth till you find the people who are really convinced of your ideas and get ready to implement your tool/model. Remember, the world is a multi context and multi perspective reality. Your tools will certainly match some one's interests.

Would like to suggest the following:

1) if no real / suitable case, can you find the closest / nearest case based on your literature review and supported by literature e.g. respondents / participants from different industries / professions etc.

2) try to conduct an experimental study with both experimental & control groups to provide empirical evidence that your tools are workable.

3) after the experimental study, pen down as limitation in your research report that you can't obtain the right respondents / participants but this can be future research for further knowledge contribution.

4) you can also perform a quantitative research by surveying respondents' opinions toward the new tool based on the literature review outcome.  This quantitative research can be used as a precursor for your above #1-3 experimental study or concurrently for triangulation purpose i.e. opinion-based & experimental-based.

You are correct to understand that in design science approaches that you need to provide some sort of validation to show that it improves practice as your thought it should. You are also correct to think that to actually using it in production might be a good first approach. However, you don't have the capability to get one. There are a couple of things you can do. Fung has rightly pointed out that experimentation is an option. You can set up an experimental process and then use that technique vs. the traditional techniques to show if it provides improvement. Another way you can go is to use a simulation study. If you have data on how the process currently works, you can simulate the old method using simulation software and then simulate with the new method/process/technique to see if based on simulation that it works well. Either of these would be an appropriate validation of your innovation.