Geoff Gannon March 18, 2006

On Google Finance

Google Finance launched yesterday. I have mixed feelings about the service. It looks to be a great finance portal. But, finance portals still have a lot of room for improvement.

Google finance begins with search. You enter the name of a publicly traded company into the familiar Google search box. The search is not limited to ticker symbols. Entering either the ticker symbol or the company name will immediately load a company page.

Entering a term like “restaurants” will return a list of relevant stocks and funds (in this case, 2,168 companies and 1 fund). A few of these companies are private; but, many are publicly traded.

The search results for generic terms (like restaurants) are fascinating. There seems to be a bias towards large cap stocks. By this I mean that large cap stocks that are somewhat related to the search term often get placed above highly relevant, smaller companies.

Google Finance’s search feature is very weak. Obviously, it’s an improvement over the other finance portals; but, the generic search feature is badly flawed. For instance, neither Electronic Arts (ERTS) nor Activision (ATVI) is among the top ten results for video games. On the other hand, terms like steel work pretty well.

If the search term isn’t in the company’s name or the company is engaged in more than one industry, the results tend to be very poor. The search feature can’t really be used as an industry search unless you know the exact term that will work best. For instance, you’d have to know to use waste instead of garbage or trash; and poultry instead of chicken. It’s not exactly intuitive.

Some search terms will lead directly to a company page. This is a wonderful feature that lets you find a company without having to know the ticker. Here, Google Finance gets a little cocky. It’s always convinced it knows exactly what you’re looking for, even when it clearly doesn’t have a clue. For instance, the search term “video games” returns Navarre Corporation (NAVR). There are two problems with this. Obviously, a generic term like “video games” should not return a specific result. Also, while Navarre Corporation is a relevant result for the search term, it isn’t even close to being the most relevant result for “video games”.

Google Finance’s search feature doesn’t seem to be able to consistently return subsidiaries. This is very unfortunate, because my first thought upon seeing the Google search box was that users would be able to perform true subsidiary searches. That doesn’t seem to be the case – yet.

For instance, a search for Kentucky Fried Chicken returns Kentucky Fried Chicken Japan instead of YUM! Brands’ KFC corporation division. On the other hand, a search for Duracell does return the maker of the CopperTop. Strangely, Duracell’s (public) parent Procter & Gamble (PG) is not listed in the related companies section of Duracell’s company page.

These are understandable mistakes for a computer to make. But, that’s the problem. A search feature used by humans shouldn’t make mistakes that are understandable coming from a computer, but would never be made by a human. For a demonstration of how a good subsidiary search should work, see Hoovers.

I’m still hopeful about Google Finance’s search feature. It’s a much needed improvement over the status quo at finance portals, and Google has a good record in search. So, it comes as a bit of a surprise that Google Finance’s stock search capabilities are far inferior to Hoovers’. The answer is probably that the people at Hoovers understand what their users are searching for, while the people at Google Finance don’t – yet.

So far, this post doesn’t exactly read like a ringing endorsement of Google Finance. After all, I’ve just said that the company whose name is synonymous with search has designed a miserable search feature. That’s pretty embarrassing. But, in Google’s defense, search probably won’t be an integral part of the site for most regular visitors. Both the company name and the ticker symbol consistently return the right results – and that’s what really matters.

Still, unless the search feature is improved, Google should drop the “cars” example. True, the search term “cars” does work; however, the fact that Google suggests a term like “cars” might lead visitors to the natural assumption that terms like “video games” work. They don’t. So, Google should pull the “cars” example – or, better yet, improve the search feature.

Once you get past the god awful search feature, Google Finance really shines. The company pages are designed to look a lot like Google. Visually Google Finance will immediately feel like an old friend. I have mixed feeling about this.

I may be suffering from some sort of Google fatigue. I’ve always liked that Google search had a clean, neutral look. Sure, it’s kind of cartoonish. But, you’re just visiting Google to go somewhere else anyway; so, it’s better for the site’s design to be understated rather than overstated.

At first, I didn’t think this look worked with Google Finance. It really lacks a sense of place. That’s great for a search engine, but I wasn’t sure it worked for a finance portal. After a few hours, I came around. Now, I think Google Finance is more aesthetically pleasing than any other finance portal. It’s easy on the eyes, and makes you feel real comfortable real fast.

For me, the ideal finance portal would look a lot like a page from Value Line. For those that don’t subscribe, you can see a sample page here. Obviously, this page is cluttered. But, it presents much of the most important information about a stock on one page. Trends can be observed without reference to a lot of different growth rates. I’d like to see a page like this (cleaned up and brightened up) instead of what you find at most finance portals.

Google Finance comes closer to this ideal than any other finance portal. A lot of information is presented on a single page. For the most part, the information that matters gets top billing. For instance, the “Company Facts” section consists of the following stats: annual revenue, net profit margin, operating margin, return on average assets, return on average equity, and number of employees. There’s a link to see more ratios at Reuters. The company’s homepage is provided as are several key pages within the company’s site.

The “Company Financials” section provides key figures from both the income statement and balance sheet. Both quarterly and annual figures are provided. Obviously, annual balance sheet data is useless. Balance sheets are as of a certain date, not for a particular period. Reading annual balance sheet data is kind of like remembering how full your gas tank was at 4:37 p.m. last Thursday – theoretically, it can be done, but it has no practical significance. Google Finance presents the income statement, balance sheet, and cash flow data in two columns (quarterly and annual). So, the annual balance sheet data is probably presented for purely aesthetic reasons.

As of this writing, Google Finance’s cash flow data is worse than useless. Google claims the data is provided by Reuters. Well, someone forget to fill out the “capital expenditures” line, which is kind of important. Presenting cash flow without capital expenditures is a pet peeve of mine, so I’m probably being too harsh on Google. I know that Google Finance is in beta, but I still think the cash flow data should either be corrected or pulled.

Presenting cash flow from operations without also presenting the activities that consume that cash is irresponsible. This is especially true, because some of the lines contain no data for legitimate reasons (e.g., no dividends were paid). For now, I’d advise everyone to steer clear of Google Finance’s cash flow data. Go to Morningstar to see an example of what good cash flow data should look like.

Google Finance links quite liberally. The “More Resources” section consists of links to specific content at finance sties like: Yahoo, MSN, MarketWatch, and the Street.com. News links are presented “above the fold”. For the most part, these news items are highly relevant.

But, I have noticed problems with a couple companies. Google Finance doesn’t seem to be able to differentiate between news that would interest an investor and high-profile news stories that repeatedly reference a company or its products but do not provide information of interest to investors.

For instance, the top news item for CEC Entertainment (CEC) is an article titled: “Officer Within Policy to Use Taser For Salad Theft”. Obviously, the incident occurred at a Chuck E. Cheese restaurant. Still, I’m not sure this is the news item prospective shareholders would be most interested in.

Overall, Google Finance does a good job of finding relevant news items. In fact, on a couple occasions, Google Finance surprised me by presenting a highly relevant article appearing in an obscure news source. For instance, a reprint of my article “An Analysis of Energizer Holdings (ENR)” makes the list of news for ENR, despite appearing at DailyIndia.com.

While I’m glad Google Finance linked to some of my work, that isn’t the point. The point is that Google was able to correctly weigh the “lack of authority” of the source (DailyIndia.com) against the fact that the story itself was highly relevant to Energizer Holdings (especially from an investor’s perspective). That’s something I have yet to see any other finance portal do.

Google Finance links to blogs. This is big news for bloggers. It means more traffic will be fed into the community. I like the idea. I think some blogs are a great resource for company specific information. However, many blogs aren’t.

Many blogs (like many discussion boards) have a tendency to lean towards the weird, the nasty, and the shrill. Google Finance seems to be somewhat effective in skimming posts from the better blogs. I include the qualifier somewhat, because it looks like Google did something wrong here. The blog posts presented on a Google Finance company page are sometimes far less relevant than the blog posts returned from a ticker search at Google Blog Search.

Why? I don’t know. Two problems seem to be cropping up. If Google Finance is searching blogs using the company name instead of the ticker, it will tend to return far less relevant results. This is pretty obvious. Anyone who searches for company information online could have told you this (and told Google as well).

A few of the blog posts appearing on the company pages of consumer products companies lead me to believe Google is not performing a ticker search. I’ve noticed a couple totally irrelevant blog posts. In fact, I found a couple blog posts that are too bizarre to quote from.

A couple of the posts are stream of consciousness pieces that have absolutely no relevance to the company, except for a title that begins with the company name. In one case, the company name appears in the title, but never appears in the body. Again, I have to fault Google for a poorly designed search. Isn’t it obvious that a post that doesn’t reference the company or the ticker within its body is totally irrelevant? Relying heavily on the title (and particularly the first word of the title) sounds like a recipe for disaster.

The correct format for a stock search is: (tickerfull company name”). This format consistently generates highly relevant search results. Placing the company name before the ticker biases the search, because some writers (like me) follow the convention of giving the ticker symbol for any public company referenced. So, placing the company name before the ticker would unfairly favor sites like mine. I’m not sure why Google couldn’t figure this one out. It seems obvious that the ticker symbol followed by the exact company name would lead to the best search results.

The company name alone often leads to poor search results. I can’t confirm that this is precisely the format Google is using to generate the blog posts on Google Finance’s company pages, but it seems very likely Google is using this format or some variation of it.

Google Finance may also be overweighting how recent a post is. I noticed it returned one of my more recent blog posts above one of my older blogs posts. This wouldn’t be a problem if the two posts were of comparable relevance. But, they weren’t. The more recent one mentioned the stock in passing. The less recent one was a post of several thousand words focused entirely on that specific stock.

There’s another danger with placing great importance on how recent a post is. Google Finance’s blog posts could become nothing more than a collection of echoes. The blogosphere is an echo chamber; skimming off the most recent posts may leave you with echoes of the same story.

So far, I haven’t seen a lot of evidence that this is happening. But, the blog results at Google Finance are currently so muddled it would be hard to tell if the lack of such echoes was the result of forward thinking on Google’s part or just such a poor search feature that Google’s ineptitude was being masked by the random nature of its search results.

Google Finance features discussion boards. Obviously, I don’t have anything to say about these yet. I hope they don’t end up resembling the boards at Yahoo. I think we are all too familiar with the poisoning of that well. In all likelihood, Google’s boards will suffer the same fate. Google seems to be aware of the potential problems, as evidenced by this excerpt from the Google Finance FAQ:

We’ve created a few tools to help ensure the Discussion Groups are filled with real people with real opinions. We also have a team at Google dedicated to ensuring these discussions are as high-quality as possible, so you can enjoy thoughtful conversations about the companies or mutual funds you care about.

Despite all my criticisms, I really like Google Finance. It surprised me in two ways. The concepts are surprisingly good and the execution is surprisingly poor. Google Finance has already become my favorite finance portal – which isn’t saying much.

Finance portals still have a long way to go. The discussion boards are terrible. The search features are very weak (despite all of Google Finance’s blunders in the search department, its search feature is still an improvement over the competition).

The biggest problem with the finance portals is their lack of imagination. There is so much more they could be. Users leave a thousand different clues about themselves and the stocks they research.

It’s not enough to list recent quotes and “related companies”. The portals should take a page out of Amazon’s playbook and (gently) recommend related items. If I spend most of my time looking at high ROE, low P/E stocks, the portal should be able to throw up the names of a few more such stocks.

How about a ticker that slips a couple (algorithmically) related stocks in every so often? If a lot of people who view Energizer Holdings also view Sealed Air (SEE), then make sure the people searching for ENR know SEE exists and vice versa. The universe of publicly traded companies is so big very few people can traverse it alone. The portals can help.

Anyway, if anyone from Google is reading this (who am I kidding?), you might want to think about my criticisms of Google Finance’s search feature as well as the problem with the cash flow data. Google Finance has a lot of promise. I’d like to see it realized.

Google would be wise to “appropriate” some of the better ideas from these sites: MorningstarGuruFocusStockDiggVinvesting123jumpHoovers, and Fat Pitch News. You may want to check them out too, if you haven’t already.

Visit Google Finance

Share: