Feature the projects that represent the work you want more of. That sounds obvious, and almost no firm does it. Most portfolios get built from a different set of criteria entirely: the biggest projects, the newest ones, the ones that won awards, and the ones a founding principal is personally proud of. Those projects might be great. They are often not the ones that bring in the next client you actually want.
A portfolio is a sales tool that points forward. The projects on it tell a prospective client what kind of firm you are and what you would like to do for them. Choosing them should start with where the firm is going.
How Most Firms Choose, and Why It Drifts
The default selection process usually happens once, during a website redesign, and then gets updated by whoever has time. New projects get added when photography comes in. Old ones rarely come off. Over a few years the portfolio becomes a record of everything the firm has done, weighted toward whatever happened to get photographed well.
That is how a firm trying to grow its healthcare practice ends up with a portfolio led by three custom homes and a tenant improvement. Nobody chose that. It just accumulated.
Start With the Work You Want Next
Before picking a single project, write down the two or three types of work the firm wants more of over the next few years. Be specific. "Commercial" is too broad. "Ground-up medical office buildings for regional health systems" gives you something to select against.
Then go through your completed work and ask one question of each project: does this make a prospective client in that target market more confident in us? If yes, it is a candidate. If no, it might still belong in an archive, but it should not be in the featured work.
What Makes a Project Worth Featuring
Once you have filtered for relevance, a few other qualities separate the strong candidates from the rest.
Relevance beats recency. A six-year-old project that matches your target work is more valuable than last year's project in a market you are moving away from. Clients care whether you have solved their kind of problem. They care much less about when.
The project has a story. The strongest portfolio pieces involve a real challenge and a clear decision. A tight urban site, an occupied building that had to stay open, a budget that forced a creative structural approach. Projects that went smoothly and look nice are fine. Projects where the firm's judgment visibly changed the outcome are better.
You can actually show it. Some of your best work may lack good photography, have a client who will not allow publication, or involve an outcome you cannot talk about. Those constraints are real and common. A slightly less impressive project you can fully document often does more work than a great one you can only mention in conversation.
Your role is clear. Engineering firms and contractors are often not the lead on a project, and a portfolio full of architect-designed buildings with no explanation of what your firm actually did can mislead in either direction. Feature projects where you can say plainly what your scope was and why it mattered.
What to Do With Everything Else
Taking a project off the featured list does not mean deleting it. Many firms, especially engineering firms that pursue public work, need a complete project list for qualifications and past performance. Keep that list. Make it searchable or filterable. Just keep it separate from the handful of projects that carry your homepage and your market pages.
The same goes for impressive projects outside your direction. A well-known building that does not match where the firm is heading can live in the archive. It doesn't have to lead.
A Simple Exercise
List your target project types across the top of a page. Under each one, write the two or three strongest projects you have that prove you can do that work. Some columns will fill up quickly. Others will be thin or empty.
The full columns are your featured portfolio. The thin ones show you where the firm needs to build proof, whether that means pursuing a specific project, documenting a past one better, or being honest that the firm is still growing into that market.
