There is no general rule to that. An AHP works after the first interview. The question is, whether it really represents your sample.  We have been running surveys with very small groups (n=3) and huge groups (n=8000) using AHP questions between other forms of asking questions. An AHP does not require  more interviews. We can sometimes see that the results are very stable after only a bunch of people interviewed. Have an eye on the rank reversal phenomenon which I would use as an indicator to see how many people you actually need to have a stable result. It largely depends on the modell that you apply. If you get in touch with people that never saw an AHP in their life I would suggest to use a warm-up round explaining how it works.

Good luck with your survey.

We are using questfox.com from Switzerland for AHP and Mixed Mode interviews. Try it.

AHP may be applied to responses of large as well as small groups; however, number of respondents may depend upon the nature of problem and availability of experts.

http://www.isahp.org/uploads/isahp16_proceeding_1155987.pdf

It depends on the occurrence of the events. Use appropriate sample size formula to calculate it.

Hello , my case is Weighting Criteria for Sustatinability of power plants and im limited with 20 experts with my Excel Template , how can i deal with the sample size in this case ?

what if i used sample size formula and i got high number of experts 100 for example !! then there will be a problem of entering experts data into 100 pair wise comparison matrices , any suggestions please ?

Ali Azzam go and have a look at http://bpmsg.com/ahp, it can accommodate large group AHP research and then aggregates the results from the whole group to give you a single view on the group's responses

Dear Santhosh Channappa

It all depends on whether your group is homogeneous since you may find it difficult to weight the criteria.

You could read my paper

Article Determinación de las Ponderaciones de los Criterios de Suste...

The extant literature on AHP applications in management and engineering studies indicates that there is no strict requirement on the minimum sample size for AHP analysis. Some studies used sample sizes ranging from four to nine.

Reference ( Amos Darko, Albert Ping Chuen Chan, Ernest Effah Ameyaw, Emmanuel Kingsford Owusu, Erika Pärn & David John Edwards (2019) Review of application of analytic hierarchy process (AHP) in construction, International Journal of Construction Management, 19:5, 436-452)