For a long time, it has been difficult to design websites with a clear structure that users can easily navigate. One major cause is the gap that might exist between web developers' and consumers' views on the optimal structure for a website. Although many techniques have been presented for relinking web sites to enhance their navigability by analyzing user navigation data, the totally rearranged new structure may be very unexpected, and the cost of confusing people after the changes has not been quantified. In this article, we will discuss how to tweak a website without completely overhauling it. To enhance user navigation on such a website while making as little changes as possible to the site's present structure, we offer a mathematical programming approach. Extensive testing on a real-world data set that is available to the general public shows that our approach not only improves user navigation dramatically with minor adjustments, while also being efficiently addressed. Further, we have shown that the model scales up well by testing it on extremely large synthetic data sets. We also create two measures for assessment and employ them to measure the enhanced website's effectiveness on a live data set. These evaluations show that the new structure is much easier to navigate for users. Interestingly, we discover that users who are more severely bewildered are now more likely to experience the benefits of the new structure