There is a great debate that a team can have internally about the quality of the visual assets (i.e. images) to use in a site or app’s front end design. The trade off is cost (money to spend, page load time for huge high res images) vs. professionalism and impression that is made upon the user/consumer who sees it for the first time (what might be called “imputability”–that you impute a sense of high quality and professionalism). The user will usually make up his mind about something from his first impression after looking at it for only a few seconds, so imputation is critical. Cost can be managed through cheap high quality stock photo image repositories such as Shutterstock, but the nicer the photo, the longer it takes for the page to load, so the trade-off is real.
I personally inject an element into this analysis that is not commonly found: that, to a coder (or an investor) an amateur home-made image looks ghastly (obviously), but to the end user, they have no idea what the difference between the various potential sources of images are, they don’t know whether it was made by a professional or a semi-amateur, nor do they have the time to care, and a homemade image, if it looks good, tends to feel friendly, warm, and non-intimidating and non-threatening, which can be the difference in making a sale or getting the sign-up. This, and whether anyone on the team has native artistic skill and/or a relationship with a good yet affordable artist, then also goes into the cost-benefit trade-off analysis.
Ultimately there is no one right answer, but it is a debate that every front end design team for web applications or apps must have at some point, especially early on.
Of course, if nobody wants to take responsibility for the potential embarrassment of making visual assets that end up looking poor, this responsibility will inevitably be outsourced to Shutterstock, and with it will go any uniqueness that the visual assets might have had if they had been custom-made.