Start by running two for loops, with the outer loop (i) running from [0, N - M - 1] and the inner loop (j) running from [0, M - 1], where N is the length of the text and M is the length of the pattern.
The outer loop runs until N - M because for the pattern to fit into the string as a substring, the remaining characters in the string from index i should be at least as long as the pattern (i.e., there should be enough characters from index i to match the entire pattern).
In each iteration of the outer loop, store the current index (i) in a variable temp, which will represent the starting position of the substring check.
The inner loop (j) iterates over each character of the pattern, comparing it with the corresponding character in the text starting from index temp.
If all characters of the pattern match the corresponding characters in the text, check if j == M, meaning the entire pattern has been matched. In this case, return the index i where the match was found.
If the substring is not found after completing the loops, return -1.

Code

#include <bits/stdc++.h>
using namespace std;

// find first index where pattern occurs in text (naive search)
int foundPattern(string text, string pattern) {
  // cache lengths
  int N = text.length();
  int M = pattern.length();
  // try all start positions where pattern can fit
  for (int i = 0; i <= N - M; i++) {
    // copy start for scanning
    int temp = i;
    // index in pattern
    int j = 0;
    // scan while characters match
    for (j = 0; j < M; j++) {
      // stop if mismatch
      if (text[temp] != pattern[j]) {
        break;
      }
      // advance in text
      temp++;
    }
    // return start if full pattern matched
    if (j == M) {
      return i;
    }
  }
  // not found
  return -1;
}

int main() {
  // define input strings
  string text = "takeuforward";
  string pattern = "forward";
  // run search
  int foundIdx = foundPattern(text, pattern);
  // print result
  cout << "The substring starts f