Welcome to Cisco Week! Q3 FY2011 Earnings

10/05/2011

Welcome to Cisco week.  CSCO reports earnings after the close on Wednesday.  When I was on the buy side this would mean a mind numbing night of numbers after a conference call that was too long and too repetitive.  Cisco is an amazing company that provides a far better set of financial data than the vast majority of technology companies.  It is the broad data set provided within CSCO’s results that are relevant for tech.  Here is a partial list of companies that have revenue exposure to CSCO:

NETL	OPXT	FNSR	JDSU	AVGO	AVT	JBL	FLEX	CLS	SANM
BKHM	IDTI	CAVM	MERX	PMCS	BRCM	MSP	AMCC	APH	MRVL
MSP	ALTR	XLNX	SIGM	FTNT	EZCH	ATHR

Then there are the competitors:

JNPR

FFIV

BRCD

EXTR

HPQ

QLGC

APKT

NTGR

ARUN

RVBD

As mentioned in a previous post, the stock has been a loser for the past year.  Of late there has been more positive sentiment on the stock by sell side analysts.  This comes from three sources: (1) management changes (CEO’s letter to the employees), (2) strategy/product changes and (3) positive channel checks.  The stock has bounced a little and management has set a high bar for this quarter (Q3) and next quarter (Q4).  This is the area that I want to explore because I would contend that the financial data set provided by the company really does help an investor understand the velocity of business within the technology industry.  The company provides backlog data on an annual basis, they provide product bookings trend on a quarterly basis and in early 2009 the CEO went as far to provide monthly product bookings data.  When I state bookings and backlog I am only referring to product bookings and backlog which does not include services.  Look at the information that the CEO provided on the last three calls:

–         Q4 2010 “From a global perspective, Q4 product orders grew 23% from a year-over-year perspective.”

–         Q1 2011 “From a book-to-bill perspective, while product revenues were very good at 21% year-over-year growth, total product order growth was only approximately 10% year-over-year; therefore, book-to-bill was below 1.”

–         Q2 2011 “Moving on to a more general picture of our products, total product book to bill was greater than 1, and backlog increased. Product orders grew approximately 8% year-over-year, while total product revenue growth was 3% year-over-year in Q2.”

If you go through the financial numbers when filed with the SEC (note the results on Wednesday will be preliminary) and take what the company tells you on the quarterly call, it possible to build a model to predict the linearity and velocity of CSCO’s business.  I have such a model and I will not share it on this blog.  It culls a decade worth of performance data, management commentary, sell side notes, etc into viewpoint designed to predict how management will guide the business versus market expectations.  This can be reproduced, especially if you have no life and like to spend many hours reading and calculating numbers on a spreadsheet.  If I look at the model here is what it is telling me for Q3 which they report on Wednesday and Q4.

The model calculates backlog, bookings and other ratios and average expected values based on the data set Cisco provides.  I am sure the model’s backlog calculation is not accurate to the actual company’s backlog, but the model is providing an output that is close enough to be of value in predictive terms as to how the management team will guide the business.  If I look only at the data set for last 14 quarters a few statistics are apparent:

•  On average the guidance for the next Q revenue is 51.47% of the backlog going into the Q.  This number has had a low of 46.28% and a high of 59.20%.  This number varies because management sees the forward looking sales funnel and can predict in quarter book to bill turns – but they do not report the funnel numbers.  These unknown numbers will be in play Wednesday night.  Going into Q3, I think CSCO started the Q with ~$4.094B backlog.

• Another metric is to look only at the product bookings as multiplier to the product guide by stripping out the services business.  This multiplier typically overestimates the revenue guide because I have to tweak the model to handle deferred revenues better.  I simply correct for it in the model.

• CSCO’s book to bill ratio (product only) has ranged from 0.91 (Q1 ’09) to 1.08 (Q4 ’09) in last 14 quarters.

• CSCO’s product revenues have never topped $9B in a Q.

• Only once (Q4 ’08, which was the Q before LEH blew up) has CSCO’s product bookings topped $9B.

On their Q2 FY11 call, CSCO guided revenues to $10.9B for Q3 and $11.85B for Q4.  I am using the mid-point of the guide for Q4.  This revenue level is a tad aggressive.  Assuming that the services business continues to grow by ~$100M a quarter, Cisco is expecting two big back to back bookings quarters or they have deferred revenues coming off the books they expect to recognize.  In the past 14 quarters, CSCO has missed the revenue line once (Q4 FY10) by 28bps.  That tells me they have an excellent near term ability to predict the follow-on quarter revenue.  If the management team says they are going to print $10.9B for Q3 FY11 I believe them.

What is surprising about the $10.9B number is it implies a big bookings number in Q3.  I estimate to get to this number the company needs $9.1B in product bookings followed by $9.75B in Q4, which is typically their best business quarter.  I think it can be done, but we are assuming that in a quarter of organizational changes, mea culpa letters, soul searching that the company was able to be focused on the business.  I am going to take them at the word on the Q3 revenue number; I just wonder what they had to do to get the numbers done?  Will product margins suffer?  The two most important forward looking numbers that management will provide for Q4 and Q1 FY12 will be (1) revenue and (2) margin trajectory.  The stock will work or fail off these two metrics.  If you are long CSCO, then you are making the assumption that management can run the business to achieve a high rate of customer wins, margins are stable and this implies that the economy is getting better.  If you are bearish CSCO, you think they have negative product cycles in switching and routing, margins will suffer and the company is unlikely to grow a pace to move the stock.  See the product revenue by quarter chart, the downtick in switching has been a big concern for investors.

Here are some charts that make you cautious on the forward expectations:

The Broader Meaning of Cisco’s Results

As soon as the CEO provides some bookings commentary the model I have can calculate the expected guidance for the forward quarters.  Beyond this exercise in buy side / sell side / management alignment, I think that CSCO’s report is going to provide some clarity to three questions:

–         Can the market specific competitors (e.g. APKT, RVBD, JNPR, ARUN, FFIV) continue the momentum against CSCO?

–         Will technology spending in general see a positive or negative inflection going forward?

–         Has CSCO stopped the negative product cycles (e.g. switching) and momentum?

If the forward guidance is soft and the product segment data looks poor, I would say this further proves in the theory that focused competitors are hurting CSCO and this is good for FFIV, APKT, RVBD, JNPR and ARUN.

Reaffirming or increasing revenue projections would imply that the US and maybe the global economies are getting stronger and that caution should be considered on being long the CSCO competitors.  The argument can be made that a rising tide lifts all boats, but some consideration of competitors eating CSCO’s lunch in specific markets must be considered and whether CSCO has decided to defend and take market share.  Margins will be an indicator.  Net/net…I wait to hear what management has to say before I make my choices.

Any lower guidance on revenues or poor margin guidance would be bad for the stock.  That would mean that you want to be long the CSCO competitors and that the management team needs to go back into the Telepresence chamber and figure it out.

/wrk

** It is all about the network stupid, because it is all about compute. **