If you choose to use a surface which is not considered "runway", best to walk it first. If it could be a runway, and is not, there's probably a reason. Walking it is a good start to detecting characteristics which could get in trouble. Soft ground would be the first thing I'd be looking for.
At the small grass runway where I used to keep my 150, I was the "caretaker", in return for free tiedowns. I took the care. One spring, an end of the 1600 foot runway was too soft for airplanes. So, determining that the rest was suitable, I got a bag of sports field marking chalk, and neatly placed displaced threshold chevrons on the soft part. From my new threshold, it was perfectly fine, albeit short. Well, the boss came home later that day, flying one of his several planes (14 at the high point of his ownership), today, it was the Piper Tomahawk - with its 5.00-5 mainwheels. He didn't take my displaced threshold markings seriously, and landed in that area. He stopped in only a couple of hundred feet, up to the tops of the tiny tires in mud. There was so much mud splattered everywhere, that it was muddy on top of the T tail! 'Lucky thing he didn't flip it. I helped him pull it out, but left him to wash it. After that, he appreciated my caretaking efforts.
His friend had me fly him up to his cottage after he lost his medical. Cessna floatplane 180 . After a check out in the tiny lake, I could drop him off, and have the plane for the weekend. He checked me out, taxiing around, pointing out every rock. He knew each one, as he had hit each one at some point. (He also showed me where he crashed and sank his previous C 185). Well, the first few weekend trips went well. Then with some tricky winds between the trees, and a too loose turn, I hit a rock, "thump, scrape, thump, scrape and so forth, the full length of the float. I was imagining a Titanic type situation, with a gash through many compartments. So, it was sink, or fly, right then. Sinking there wound be bad, not a road within 20 miles, so a very expensive helicopter hoist. I poured on the power, hoping to get up on the step before I took on too much water. Any I took on, would drain in flight, and I could beach it easily at home base. I got airborne, Whew! When I landed back at base, I beached it, and went for the dolly to pull it out, and see how much metal I'd have to patch that weekend. Well... I could not tell which of the many scratches was mine! So, relived, I gave it another good look, pumped the floats again, and wend flying for the rest of the day!
When I was asked to demonstrate the C182 amphibian into a very short runway in central Norway, I was skeptical. The runway was commonly used by powered hang gliders, and is only 900 feet long. This was 40 feet longer than the landing ground roll performance value for the conditions. Knowing that I would use all of the runway, I drove over first, and walked all of the runway. In particular, what was at the far end, and past the fence, in case I did not get stopped. It was fine, no surprises.
I flew my flying boat into a small lake at a fishing camp in the very far north of Quebec. The camp operator had buoy marked the lake for floatplane use. I landed in the buoyed area, but had trouble turning around in the wind, so went outside the buoys a little. After a very nice welcome by the camp staff, the boss, who was equally welcoming, gently, sternly said to me: "stay inside the buoys, they're there for a reason.". He knows the lake much better than I do, and that would be a very inconvenient place to punch a hole in the hull.
So if you're going to operate outside the lines, walk them first!