Okay, maybe not failed, per se.
Sometimes—actually often—clients don't end up choosing the logo that I feel is the strongest, for a wide variety of reasons. The thing you must understand is that the work presented on my site is almost all compromise. Sometimes client feedback strengthens the design, but sometimes it doesn't. There is a certain amount of subjectivity to design.
So without further adieu, here are a few logos that I felt were either as strong as the final logo, or stronger, and the journey that led to their dismissal.
Finn AI's branding exploration went through a whopping five rounds of revisions, with dozens of iterations of logos (the combinations were almost endless when we started mixing and matching different typefaces, different icons, and different casing of the letters in Finn AI). It was a bit tricky getting consensus as there were quite a few people giving feedback, and the feedback was often quite conflicting.
The brief began with a focus on heavily emphasizing the "conversational" aspect of the technology, and not lean as heavily on the AI. However as we iterated through the different versions, Finn decided that they wanted more of a fusion of "conversation" and "AI", so we shifted gears in the iconography.
Presented in first round. Focus of the icon is entirely conversational (as per the initial brief before we switched gears and brought more of a tech feel into it). It's still a strong logo, but it definitely doesn't conjure up a feeling of technical innovation, and I think that in some ways the client needed to see it to realize they actually did want to lean more heavily on the AI aspect of the business.
Presented in the final (fifth) round. Overall shape is a chat bubble (similar to the final logo), but shapes forming it are clearly more AI-focused (reminiscent of a circuit board). I preferred this one to the chosen one because I thought it balanced the conversational aspect and the technology aspect the best, and just generally had a more interesting shape.
I really liked this one. The concept here was a transition from a sound wave into data nodes. I thought this was a nice translation of conversational AI and I really liked the way the dots in the nodes are reflected back in the wordmark. Feedback on this one was very mixed—people either loved it or didn't like it at all.
All of the above logos were actually presented in black and white only—I've just converted them to the colour palette that I preferred. Colour palette was an area where we switched gears early on as well. Initially we wanted to stay away from blue and green (Finn AI's old palette) because it didn't differentiate at all from competitors (turns out there's a lot of blue and green in the banking and AI world), so I had started the brand exploration using cool deep greys and vibrant salmon shades. Ultimately they ended up gravitating toward teal (blue and green mixed) and we didn't end up utilizing the bolder grey and salmon palette.
They also ended up shifting away from the serif typeface we had initially played with—in addition to merging "conversation" and "tech" concepts, the brand also needed to appeal to large financial institutions, which tend to be more conventional in their typography. I felt a serif typeface captured that nicely, but was offset with a more modern feel by using all lowercase lettering. In the end however, we went to capitalizing the name in a sans serif typeface. We also removed the dot (something that was heavily debated throughout the process as well).
In the end, what they were able to reach consensus on was this logo. We dialed things back quite a bit in terms of palette and typography, pulling it away from some of our earlier "edgier" concepts. Finn AI is a startup, and I think that ultimately they weren't sure they had the recognition or capital in the industry to go in a bold direction. While I don't particularly mind the logo they settled on, I also don't find it to be very memorable or eye-catching. To me it isn't as "cutting edge technology" as what Finn AI is doing in the banking industry.
But ultimately it's their brand and their company, and I happen to know they're very happy with the logo we arrived at.
Feast Culinary Consulting
This was a bit of a weird one! And a truly "failed" logo because ultimately the client and I ended up parting ways without a final logo. It was unfortunate, as we'd had a really good working relationship on the previous project, but a third party had entered the picture and it seemed to really confuse the dynamic.
Feast came to me knowing "exactly what they wanted". (As mentioned previously, this is usually a really bad thing and tends to doom a project before it even begins). So, I delivered exactly what they asked for in the first round (which was a simple wordmark with a clearly dictated brand environment). What usually happens when you give a client exactly what they ask for is they realize once they see it that it's not quite what they want.
So we went back to the drawing board, and this time the desire was to have an icon instead of a simple wordmark. Feast is run by a business consultant and her partner, a former chef. Basically Feast comes in to your restaurant and tells you how to improve it, from the food to the decor to the accounting, and everything in between. The restaurant industry in Vancouver is pretty dog-eat-dog (ouch, pun not intended), and restaurants are always looking for a leg up.
After a bit of exploration into what kind of visuals they felt represented the brand, they settled on knives, both a prominent feature of the kitchen, and a very literal representation of their "cutting edge" business offering. As one of the partners was a chef, she was keen on featuring a knife/knives in the logo icon.
We started here. The desire was to be bold and conjure up strength and edginess. I used steely greys for the color palette (our original wordmark was vibrant blues at the client's request) and a no-nonsense typeface. I kept the knives detailed enough to be recognizable, but not overcomplicated.
Feedback on the first version was that it was too straight up and down, too expected, so we tried the knives on an angle...
Can we integrate the icon into the wordmark? The crossbar of the A is represented by one of of those magnetic metal strips.
Can we integrate it even more? Maybe the entire wordmark is mounted on a magnetic strip.
Here's where things started to go off the rails a bit. Feedback I kept getting was that the knives needed to be more realistic—much more realistic—almost photorealistic. I had also chosen a knife shape that would be very obvious at small sizes, and recognizable to anyone. The chef side of the partnership had a very specific knife style in mind that looked a lot like a boning knife. Unfortunately it's a very non-descript knife and really doesn't convey the sharpness and boldness of a chef's knife, and gets absolutely lost at small sizes.
I tried to show them that more realistically rendered knives don't look as iconic, and don't scale at all well down to small sizes. So what if we use a single knife, more realistic (a single knife will be bigger overall in small versions of the logo)?
Single knife, about to slice (nothing)? Oof, nope, still not realistic enough, and still not the style of knife they'd like to see. This is when they decided that we just weren't going to reach a solution together and we parted ways.
As a conclusion to this tale, I should point out that the logo they did end up with (designed by someone else), ended up being a very simple wordmark (which was where we started!!), and I can sleep at night knowing that I did everything I could but sometimes people have to take the long hard road to end up where they need to go.
And just to reiterate, this is why it can be a very bad thing to go into a project knowing "exactly what you want", because it doesn't leave you very open to alternatives when things don't go as planned. Once the client had formed a very specific idea of a knife in her mind, it was nearly impossible for her to switch gears.
In an effort to save money, startups will often end up soliciting a logo from these websites that charge $50 or $100 for a logo. I don't recommend this. Even if you don't hire me to do your logo, don't do this. For one thing, the quality of the design is generally bad. Here's where "you get what you pay for" really comes into play. These "designers" are often people working in developing countries with a little bit of software know-how and very little design know-how, working for low wages and pumping out logo after logo. They also often end up stealing or copying rights-managed graphics from stock sites or from other brands' logos. This is how they can produce so much for so little. One of the first logos I ever designed post art school was for a university sports team, the Trojans. The original artwork that I created for that logo has been stolen and used in at least a dozen other logos (that I can find in a reverse Google image search). It was even put in another designer's portfolio until I asked him to take it down.
So beware. What seems like a good deal might be, at the very least, poor design. At most, it might result in you getting a cease and desist notice for using graphics you don't have the rights to.
Flosonics came to me needing brand support for some printed materials. Their logo was new, purchased from one of these $99 logo sites. Since it was new and they were just starting to develop a lot of their materials, I saw an opportunity to improve it. Logos (and design in general) in the medical industry are notoriously awful, and it wouldn't take much to set them apart from the competition.
Flosonics Medical is a company that produces a disposable ultrasound patch to monitor fluid administration in emergent patients. The icon in the original logo was meant to represent the "doppler" effect of fluid monitoring, as well as the wireless (bluetooth) capabilities of the device (and you can see an 'f' in there as well). Sound concept, but poorly executed. The circles weren't perfect, and nothing really lined up properly. The 'f' felt forced into the icon, and ended up more of a distraction than anything.
So we set out to explore a few icons that might better represent the device capabilities, and have a more elegant, professional feel to them. This is all about confidence—if you want people to invest in your groundbreaking device, it helps to convey your expertise through a professional looking logo (I should also say that the client liked this original logo, and it makes it tougher for me to sell them on a refresh when they don't see anything wrong with what they've got.)
Keeping their colour scheme (mostly—I shifted the harsh black to a navy blue), we attempt to show both stroke volume and doppler effect without trying to force the letter 'f' into it...
As I began to explore, the client wondered if we could portray blood in some way, since the device is all about monitoring fluids via stroke volume.
I have an entire Illustrator document full of versions like this, showing different ways to integrate a blood drop, stroke volume and the doppler effect—all concepts the client wanted to see in the icon.
Once again, the client decided to stick with their original logo (allowing me to clean up the artwork a bit and keep the new navy blue and cyan colour scheme). I'm speculating a bit here, but I find engineers and medical professionals the toughest to work with when it comes to logos and branding in general. Their professions are focused on being logical, literal, and devoid of the abstract. While these are critical qualities of their fields, they often stifle creative exploration, and as previously mentioned in the above Feast example, literal representations of complex objects or concepts rarely translate well to an iconic logo. Imagine if the Apple logo showed every bruise, every imperfection, discolouration, bent stem, and misshapen edge. It would be an accurate representation of an apple, but the detail would be distracting and it would become an unreadable blob at small sizes.
Two Peas in a Pod Catering
This was an almost failed logo, in that I had to beat my head against a wall quite a bit to get to the final product.
And I knew from the start it was going to be tricky. The client had been using an actual photo of a pea pod as their logo, and wanted something "as close to that as possible". I had explained that a logo shouldn't be photorealistic (especially when you want to put it on aprons and t-shirts and hats), and that with an extremely limited budget, our only option was to create something more iconic and simplified. They agreed, or at least they appeared to, but I don't think that meant the same thing to them as it did to me.
Once again, this is an example of extremely literal interpretations going awry...
Since I thought we had agreed on a simplified, illustrative representation of a pea pod, I went very geometric in the first round. This logo would reproduce well on clothing, and had the "cute" feeling that the name implies. This went too far into cute for them, though. (I believe they described it as "cartoony").
Feedback was that they wanted a very realistic looking pea pod, with realistic colours (admittedly my colours aren't terribly realistic as they needed to have contrast or the pea pod would just be a solid green blob at small sizes). They kept repeating the same thing over and over: "we want it to look fresh". Here's the thing: I didn't realize that what they meant by that was they literally wanted it to look like a freshly picked pea pod, in as realistic a depiction as possible. ("Fresh" in design terms means something fairly different than it does in produce terms.)
In addition to this "fresh" direction, they supplied me with some illustrations of pea pods that they liked. They were all fairly cheesy clip art examples, and I tried desperately to figure out what to glean from them. They also stressed they wanted a very realistic colour scheme, which meant losing some of the contrast I had been trying to establish.
The final result. I don't mind it, but I still had to fight hard to keep the contrast that this one has, and to steer away from some of the "clip-art" examples that they had given me. I still had to exaggerate the shapes quite a bit (real pea pods aren't so plump and curvy), and I had to gently steer them away from an extravagant curly vine that integrated into the text (very fussy and difficult to pull off in the vertical version of the logo, not to mention the focus should really be on the two peas in the pod rather than the stems and leaves).
In the end I think that both the client and I were a little frustrated at how the process had gone (it seemed as though we had a few breakdowns in communication, and we also just weren't on the same page in terms of how realistic an icon should be), but in the end I believe they were satisfied with the logo, so I guess this one was a near failure that I pulled out of the fire.