Why I Sketch Every Project Before Touching a Shovel

The first thing I do on every project, after the site walk and the conversation, is sit down at a desk with a pencil and a sheet of trace paper laid over the survey. The first sketch usually takes two or three hours. The second one, after sleeping on it, takes another hour. By the third sketch the plan is ninety percent of the way to what we will install.

This is unusual now. Most landscape designers in 2026 work directly in CAD or in a 3D rendering tool that produces photographic-looking outputs in minutes. The renders are pretty, the clients love them, and the conversion rate from rendered presentation to signed contract is high. I do not begrudge anyone the workflow that works for them.

But I keep drawing by hand, and here is why.

A sketch is honest about what it is

A pencil sketch does not pretend to be a finished image. It shows masses, edges, paths, and proportions, with notes in the margin about what each blob represents. The client looking at the sketch is reading a plan, not a photograph, and they engage with it as a plan: what is this shape, where is it relative to the house, what does the shadow do at three pm in August, why is the path curving away from the corner of the deck.

A photographic render, in contrast, looks finished. The client engages with it visually, the way they would engage with a photograph in a magazine. They miss the structural decisions because the structure is presented as a fait accompli. The conversation about whether the path should curve more or less, whether the bed should be three feet wider, whether the seating area should rotate ten degrees, never quite happens; the render has already decided.

A sketch costs me less time, which costs you less money

A render takes me four to six hours to produce convincingly. A pencil sketch takes me two. The render is more impressive at the meeting; the sketch leaves us with more time to talk about what actually needs to be different about the plan.

Across a typical project, I produce eight to twelve sketches before installation begins. The first three or four are exploration, working out the basic geometry. The next several are refinements, after we have walked the site again and thought about edge cases. The last few are construction-level: detailed enough that the crew can lay out the design with stakes and string. None of them needed to be rendered.

A sketch is changeable

Every iteration in CAD or in a 3D tool requires editing the digital model, regenerating the lighting, re-exporting the image. A sketch revision happens in five minutes with a fresh sheet of trace paper. The lower friction of revision means more revisions, which means a better final plan.

I have watched projects in CAD freeze around an early decision because the cost of changing it was too high; the client said "OK" because they were tired of waiting for the next version. The pencil-and-trace workflow keeps the design loose for longer, which is when the design is most likely to improve.

What a sketch cannot do

A sketch does not show a client what their yard will look like in three years when the plants have grown in. For that, I sometimes pull a photograph of a similar finished project from my own files and use it as a reference. Photographs of real installations communicate the future of a planting better than any render.

A sketch also does not produce an irrigation cut sheet, a planting list, or a quantity takeoff. Those are produced separately from the sketch, in a clear written document, before the work begins. The sketch is the plan; the documents are the contract.

The point

The sketch is not nostalgia and it is not a stylistic choice. It is the design tool that produces the best finished gardens for the time spent on them. Every project I am proud of started with a pencil and a sheet of trace paper, and the rendering software is unlikely to change my mind on that.

If you are curious to see the working sketches behind a recent project, I usually pull a few out at the first meeting. They are not pretty in the polished sense; they are the actual artifacts the project was built from. Most clients find them more useful than the finished construction documents.

← Back to the journal