Investigation is tough in a roleplaying game. First off, your players are bound to miss a clue that you think is so plainly obvious. In fact, there’s a “rule” about that: the three clue rule.
For any conclusion you want the PCs to make, include at least three clues.
Whenever I run an investigation, I try to keep this in mind and have at least three clues spread out for the players and maybe a few more in my back pocket in case they miss those, but generally that’s enough to get them on the trail to find the ones they may have missed. From there, it’s up to them to finish the investigation, but much like yesterday, it’s worth adding some time pressure to their investigation. Maybe someone’s life is on the line or there’s a bonus payout for finishing sooner.
As a side note, I want to mention my general dislike for Dungeon & Dragon’s 5th edition Investigate skill (along with Perception). I don’t love that players assume they can just roll the dice and not ask interesting questions. I probably need to twist it around so that there are grades of success where an utter failure finds nothing and a huge success means they find exactly what they want, but there are shades of gray in the middle. I need to think more about that.