First of all, it should be pointed out that the Turkification of Anatolia was not a conscious process or policy. It is not that the Seljuks were "able" to do it in one case and failed in another, it's simply that in Anatolia it happened and in Iran it did not.

 

Now, the standard work on the Turkification of Anatolia is Vryonis Jr., Speros, The Decline of Medieval Hellenism in Asia Minor and the Process of Islamization from the Eleventh through the Fifteenth Century. University of California Press, 1971.

Turkish nomadic groups began migrating into Byzantine Anatolia in the eleventh century, particularly after the Battle of Manzikert in 1071. However, they constituted a relatively small group in comparison to the native non-Muslim population. Vryonis argues that conversion to Islam and Turkification were closely linked, with converts to Islam eventually adopting the Turkish language as well. His main thesis is that the primary impetus for conversion was the breakdown of the Byzantine church organization in the region. Cut off from much of its former revenue and from Byzantine state support, the Orthodox Church in Anatolia was unable to maintain its cohesion and fell apart. It ceased being able to fulfill its previous social functions, which were taken over by new Muslim institutions. The regions still under Byzantine control also suffered from Turkish raids, putting strain on the Church and causing peasant flight even prior to their conquest. This was especially pronounced during the periods when the Seljuk state was weak and unable to maintain control over the Turkish nomads and frontier lords (such as in the fourteenth century). The Orthodox Church was only able to regain its cohesion at the end of the fifteenth century, when the Ottomans reestablished strong state control, ensuring the security of the population. By that time, most of the population of Anatolia had either converted or fled.

 

If we adopt Vryonis' view that conversion was the main facilitator of Turkification, then perhaps we can say that Iran did not become Turkish due to the fact that most Iranians were already Muslims. Native Anatolians adopting their new religion adopted the Turkish language from their overlords as they sought to build a new identity for themselves. Iranians had no need for a new identity and thus never became Turks. Of course, this is only one of many potential explanations.

Lastly, the Seljuk era actually did leave a mark on Iran, as there is indeed a significant minority of Turkic language speakers living in Iran today.

 

Edit: A side note - the "converts become Turks" theory is strengthened by two further examples: Cyprus and Crete. These were two islands conquered by the Ottoman Empire in the sixteenth and seventeenth centuries. Both eventually developed a large local Muslim community, but did so in very different ways. Cyprus was conquered in a quick campaign in 1570. The major fortresses were given Turkish garrisons, and the countryside was partially settled with Turks from Anatolia. The Muslims of the island all continued to speak Turkish, while the Christians spoke Greek.

Crete, in contrast, was conquered over the course of a long war, from 1645 to 1669. Ottoman armies thus remained on the island for more than two decades, and worked closely with the native population, many of whom converted to Islam. These converts, while they learned Turkish, continued to speak Greek as their primary language, and the Muslim population of Crete remained Greek-speaking after the Ottoman army left. It seems as though the continued presence of the Ottoman army as a source of opportunity for local volunteers produced a critical mass of Greek converts such that converts did not feel as though they had to wholly adopt Turkish in order to participate in the Muslim community.

Cyprus, like Anatolia, developed a Muslim community through converts attaching themselves to a small "core" of Turkish-speakers, and thus became Turkish. In contrast Crete was able to develop an independent "core" of Greek-speaking Muslims within a single generation, and thus remained Greek. I think these two contrasting examples add a great deal of weight to Vryonis' theory by showing the criteria which was needed to make it happen.