Competing with Tech industry giants can be HARD. This of course happens constantly in the software and hardware business. It’s all part of the cycle of innovation and hardware and software competition that makes our business so dynamic. One of the more high-profile competitions that is happening now is in the Artificial Intelligence market. Many pundits are expressing the view that giants, such as Microsoft and Alphabet/Google, have already won. The argument is that they have a lead and the cost of training and running the artificial intelligence models of today (large language models, or LLMs) is so excessive that startups won’t be able to compete. More on that later.
Many years ago I remember reading an article in the business section of our local newspaper about a new product named CUIL (pronounced Cool). This company was trying to create software competition in the Search Engine category. I already knew about CUIL, because I had noticed that it had just recently indexed the PJM Consulting website.
Enough differentiation?
One of their claimed differentiating factors was that their search index was twice as large as Google’s. In addition, they believed that they had improved the ranking algorithms. They also presented the search results in a different, and they thought better way. The results offered fewer results per page, but more comprehensive information on each site. They often included a photo or other graphic image. All these things Google has of course since adopted. The premise of the article was that it could have a chance to be a real competitor vs. Google. Or at least compete with Yahoo and Microsoft for secondary market share in the huge search business. The founders had impressive pedigrees and came from Google on the technical side.
The article gave credence to the possibility of CUIL being a potentially serious competitor to Google, Yahoo, and MS. But it also pointed out that quite a few companies had attempted to enter this fray. Collectively these upstart competitors created barely a blip in search engine market share after their entry.
I took a quick peek at CUIL at the time. The presentation was different and possibly superior for some tastes. But in my quick look, I wasn’t impressed with the relevancy of the search results. No matter how you present the data, the relevancy of the results is paramount in the search business. I stuck with Google, as it appeared that just about everyone did.
Just one of many startup failures
I’m pretty certain none of you ever heard of CUIL. Did CUIL meet with any success at all? Well, they’re no longer in business, closing up shop in September 2010. CUIL was barely around for two years. They took on – head on – what is one of the 2 or 3 most powerful technology companies in the world today. The startup attempted to compete with them directly, in their core area of strength. So you can’t say that the odds of success were high, which they rarely are for any startup anyway. And this might have been the ultimate toughest market to enter. Microsoft over a long period has continuously poured massive amounts of money into competing with Google, with modest success. Most recently with BING and the Generative AI partnership with OpenAI.
I later worked as a consultant with a startup Search Engine company with some great technology. The company had little success. This was true even with a more appropriate strategy than CUIL. They didn’t try to take on Google directly in their core web search business but in more niche market segments. Search is a really difficult business for new competitors. We’re talking about a REALLY tough market segment here. But this IS the technology business; everyone gets at least a puncher’s chance.
Once upon a time, Google was an unknown upstart competing against the well-known names at the time. Companies such as Excite, Ask, Lycos, AltaVista, and WebCrawler to name a few. All of which had at least modest success until Google came along. A big key to having a real chance at survival (if not success) is usually how well you execute. In addition to a questionable strategy, it didn’t appear that CUIL executed very well.
What’s the best way to compete?
But execution aside, what’s the best way to compete in the software and hardware industries today? Should you just steer clear of the elephants of the industry? Many believe this is prudent, but I think it is not always necessary. Well, maybe charging like an angry bull directly at Google’s search engine isn’t the best bet! But it wasn’t so very long ago that it was nearly impossible to get a venture capitalist to fund a company that was perceived to compete in any category vs. Microsoft. At the time, this non-fundability could have been viewed as MOST categories of the software business. Even if MS was rumored to be CONSIDERING entering a segment, funding might be a no-go. Yet Microsoft, while still powerful these days, certainly isn’t considered unbeatable in any segment they choose to compete in.
Can you Dance with the Elephant?
If it isn’t insane to compete with the giants, what are some best strategic practices to do so? Practices that an early-stage tech company can adopt to give it the best chance to survive and thrive when entering market categories with large, entrenched competitors? Let’s take a look at a few ideas:
Make sure that you can differentiate from large hardware & software competition
This would seem obvious for any business, but what do you do when you go up against huge hardware or SaaS competition with a dominant brand? Well, don’t even try it without significant differentiating factors! These factors don’t necessarily need to be product-related. It could be free and outstanding support, better (cost-driven) price points, exceptional ease of use, or a completely novel business model. Or, many other things. But don’t kid yourself. You will need REAL differentiation.
Pick a niche, any niche–at least to start
It is important to pick a tight enough niche so that you can provide that true differentiation discussed above. Your outside investors may want you to attack a huge market. But if you don’t have that sometimes BAD influence pushing you in that direction, pick a smaller, tighter niche. One with less hardware or software competition that you can have a better chance of dominating when you’re new. If you are successful in your initial niche, you can then broaden out into adjacent segments and become a giant yourself (see Amazon.com). Down the road, maybe you take on the giant “head-on”. But at startup, it is NOT the time for this.
An AI software competition example
Going back to the competitive situation in the AI market that I alluded to at the beginning of this article. At this early stage, some software companies are focusing on small language models, rather than the LLMs that have so publicly been created by the industry giants. These small models are of course not as generally capable as the LLMs. But many can run on a local computer CPU and some even on a Smartphone! Contrast this with the massive cloud resources (and corresponding financial and energy costs) required by the current LLMs.
These smaller models can be trained quite well for specific vertical tasks, even sometimes outperforming the LLMs in these more narrow applications. Some think this will be the future of AI. That remains to be seen. But this is a great example of competing with industry giants without a suicidal, direct attack like the one that failed in the CUIL example above.
Another that some companies are choosing to compete in AI is by leveraging the LLMs, but verticalizing them for specific markets. This allows them to be more capable in that niche when compared to the horizontal LLM “platforms” themselves.
These are just two examples from a very current battle going on in AI. The key takeaway is to compete, you need a differentiated, and usually niche strategy.
Consider raising more money than you think you will need
Every once in a while a new company will “hit on all cylinders” from the very beginning. But in my consulting practice, I rarely see this. A significant part of my practice is helping companies “pick up the pieces” and turn around after their initial business plan or execution has gone awry. No one likes to give up more equity than they need to. But things usually take longer in a startup to begin working than you initially project. There are usually too many things that you just don’t know until you really get into the marketplace. Plus, it’s generally easier (and cheaper!) to raise a bit more money at first than it is after that first misstep. A little extra funding in the bank can be a good insurance policy against a capital crisis early on.
DON’T try to be like them
A common mistake that I often see early-stage companies make is trying to be like their giant hardware or software competition. Sometimes this comes from an inferiority complex. Other times it’s because the founders come from one of the giant companies themselves. The last thing you want to do is create a big company bureaucracy in a startup. In almost every way, you want to operate VERY DIFFERENTLY from your huge, slow-moving hardware or software competition.
Resist the urge to create huge amounts of process before your company size dictates it as necessary. Be very careful about hiring away senior executives from your giant competitors. Don’t do it unless you are certain that they also have successfully operated in an early-stage company before. Stay as fast and nimble as possible for as long as you can. That’s the primary advantage vs. the big guys at this early stage of a company’s development.
Recognize the giant’s execution weaknesses and beat them there
Analyze the large hardware or software competitor’s business, and try to create your differentiation where they are weakest. It could be faster customer service, better channel relations, better ease of use, much lower total cost of ownership, a free entry-level product version, etc. If you concentrate your differentiation where they are doing the poorest job, it will accentuate that difference to the marketplace. This will provide a better chance of your advantage being recognized.
Focus, Focus, and Focus
This advice can be viewed as the culmination of the points above. Make sure that you don’t try to do any more things than you can do EXCEPTIONALLY WELL at this stage. You can always expand your focus later. Remember, there is a chance we would all be speaking German if Hitler hadn’t prematurely opened up a second front with Russia in World War II! The tech landscape is littered with companies that followed an analogous “spread-to-thin” business strategy, with similar disastrous results. Upstart software competition Novell and Netscape are two former high-profile cases that immediately come to mind.
SUMMARY
As an early-stage company entering a market where a major company or two are the known leaders, ensure you don’t “bite off more than you can chew”. You can always expand your focus after initial success. Contracting your focus later is usually quite a bit more painful. And many companies don’t make it through that transition. That’s my advice on how to attack a large, entrenched hardware and software competition. As usual, I’d be interested in seeing your comments–post them below to expand the discussion.
Follow Phil Morettini and Morettini on Management via Twitter, Facebook, LinkedIn, RSS, or Subscribe to the Morettini on Management Newsletter hosted by LinkedIn. Contact Phil directly at info@pjmconsult.com
Leave a Reply