This series, brought to you by Yahoo! Finance, looks at which upgrades and downgrades make sense, and which ones investors should act on. Today, our headliners include downward price tweaks at Zumiez (ZUMZ 1.77%) and Splunk (SPLK), but for ConocoPhillips (COP -0.43%), a big buy rating. Let's start with that one.

Fill 'er up!
The week ends on a bright note for oil investors, as ace analyst Wells Fargo comes out with a new outperform rating on ConocoPhillips. Praising the company for its "favorable reserves and acreage positions," its "expected production" going forward, and its "balance sheet strength" today, Wells is initiating a buy recommendation on the oil major with a price target of anywhere from $72 to $78 a share.

So essentially, Wells is looking at Conoco as an assets play, and focusing on whether the company has the staying power to eventually make these assets pay off. But how does the company look from a more traditional, price-to-earnings perspective?

Unfortunately, the answer there is: not good. Sure, on the surface, the stock's P/E of just 7.0 looks attractively low. But it's the reason investors are paying such a low multiple earnings that matters. Namely, most analysts on Wall Street see a future full of falling profits at Conoco. Declining earnings next year push the company's forward P/E up toward 10. Over the next five years, as earnings continue to fall, analysts see12% annual shrinkage in Conoco's profits. Meanwhile, free cash flow at the company is all but nonexistent -- a mere $8 million in positive free cash flow generated over the past 12 months, versus purported "profits" of $10.4 billion.

Long story short, Conoco may be worth something... but it's not worth a buy rating.

Zumiez doesn't
Meanwhile, as Wells puts a bright face on Conoco's bleak future, other analysts are taking a more skeptical look at Zumiez after its weak Q3 report. $0.40 per share in profits were down a nickel from last year's Q3. And this being despite a 17% increase in revenue, so you know that margins got hurt badly as well. Meanwhile, the company is forecasting earnings as low as $0.59 per share for the current quarter, which, if correct, will fall 17% short of analyst estimates.

Little wonder, therefore, that analyst Janney is cutting its price target on the shares by $3, to just $21 a stub. While on the surface, Zumiez stock looks attractive at 15 times earnings, and a growth rate pushing 19%, it's worth noting that free cash flow at the company -- less than $30 million -- is fully a fifth lower than reported income. It's also down significantly from the $43 million in cash the company generated last year.

In other words, Zumiez is getting worse, not better. And yes, it deserves the price target cut.

Splunk goes splat
Similarly, "real time operational intelligence" company Splunk is getting hurt this morning as an early backer (Needham, which initiated the stock at "buy" in September) cuts $6 off its price target. Needham's move reverses the 5% rally Splunk enjoyed after reporting earnings yesterday, and it's shaving about 2% off the stock's share price in early trading.

But why? After all, Splunk did "beat earnings" yesterday, reporting 67% revenue growth, and losing "only" $0.06 a share. Splunk also forecast estimate-beating revenues for the current fourth quarter, raised guidance for the year, and raised expectations for operating margin as well. Isn't that all good news?

Well, yes, in a manner of speaking. Collecting more revenue and losing less money on it is good news. But it's not quite as good news as collecting more revenue and earning a profit on it would have been -- and so far, Splunk isn't promising anything of that sort. Instead, the money-losing firm is promising to lose more money, while Wall Street analysts predict it will continue losing even more money for at least the next three years. Three years, as you can imagine, is an awfully long time in the Internet business, and it seems Needham is feeling less inclined to wait for Splunk to turn profitable than it perhaps was back in September.

Personally, I've got a good feeling about Splunk. I like, for example, the fact that while everyone's worrying about "GAAP earnings," Splunk continues to calmly plunk down cash profits on the table -- more than $24 million in positive free cash flow over the past year alone. Needham, for the record, is also still optimistic, and continues to rate the stock a buy. It's just being a bit more cautious with its price target, and I can hardly fault Needham for that.

Fool contributor Rich Smith has no positions in the stocks mentioned above.

link