�ݺ�ߣ

Intellectual Asset Management January/February 2014 11www.iam-magazine.com
The brokered patent
market in 2013
Last year we wrote an article on the
brokered patent market (“Turning the
spotlight on the brokered patent market”,
Intellectual Asset Management, issue 57),
because we felt that we had enough data
to shed light on this otherwise obscure
business. We were delighted to receive
positive feedback, as well as requests for
more information, so we decided to update
our work with new data and further analysis
of our existing data.
Brokers regularly contact corporations
to try to sell them patent packages. Some of
our clients see well over 50 potential deals
per month. Not all of these packages fall into
the brokered patent market, but we believe
that the quantity is significant. Buying
objectives and processes for identifying
deals of interest, tracking, bidding and
closing are critical when a company might
see more than 600 deals in a year. Without
some basic information about what a normal
deal looks like, it is extremely difficult to set
the parameters of a buying programme. We
have used the data in this article to help our
clients set up successful buying programmes
– we hope that readers will also find the
information useful.
This winter we examine the public
but not yet particularly well-understood
market for bare patent transactions created
primarily by patent brokers and regular
While the brokered patent market
in the United States is going from
strength to strength, there is still
little analysis of this obscure
business. A review of 222 deals
offers welcome insight into what is
becoming a lucrative market
By Kent Richardson and Erik Oliver
patent sellers. Brokers typically provide
a detailed package for every potential
deal – we use the specific package as our
fundamental tracking unit. Our focus data
sample uses packages seen between 1st
June 2012 and 31st May 2013, and has 296
packages containing 4,316 US patents and
9,359 total patent assets worldwide. There
is no overlap of packages between the 2012-
2013 sample and the packages discussed in
last year’s article. Where applicable, we have
included comparisons to last year’s paper.
Our analysis is based on a sample of the
market, not the entire market. We estimate
that while running buying programmes,
we saw somewhere between 50% and 90%
of the available packages, depending on
the time and the particular project. Also,
because of our clients’ locations, we tended
to see more US-focused transactions. We
excluded very large deals (over 200 patents)
or deals that were not widely distributed,
even though brokers may be involved.
Because of rounding, some tables may
not add up to 100%.
Patent brokers
We continue to focus on the brokered patent
market because it is open to all buyers,
brokers provide an initial filter of the assets,
the seller’s pricing expectations have been
established and the outcomes can be tracked.
Brokers bring useful skills, including
the following:
• A selection of viable sellers, including
some certainty that the seller is willing
to sell.
• The filtering of patents and identification
of important patents and claims.
• Pricing guidance.
• Guidance to the seller on sale terms and
timelines.
• Definition of a process for diligence,
bidding and sales.
Good things in small packages

www.iam-magazine.com12 Intellectual Asset Management January/February 2014
For a list of brokers and contact
information, see http://richardsonoliver.
com/brokers/.
Patent broker data
Out of 55 brokers from which we received
packages, 25 (Table 1) had at least three
packages on the market over the 12-month
period studied (12 brokers last year). We saw
substantially more brokers with multiple
sets of assets on the market. This year, we
had a more active outreach programme.
AST, Intellectual Ventures and RPX
are some of the repeat sellers of patents,
while small and medium-sized enterprises
continue to be the vast majority of the
sellers in our study.
As in our last article, we did not always
find strong correlations where expected.
Generally, there were no correlations between
brokers and deal price or package size
(except at the extreme ends). Again, we saw
almost no technology specialisation among
the brokers – all of the brokers handled
all high-tech areas. The exceptions were
those affiliated with larger semiconductor
reverse-engineering houses (eg, UBM
TechInsights and TAEUS). The rationale for
the semiconductor correlation is apparent
when one considers the deep semiconductor
reverse-engineering experience that these
companies have. Also, for semiconductor
implementation patents (eg, circuits and
process technology), claim charts are
particularly difficult to obtain without some
level of complex reverse engineering.
As with last year, brokers continue to
prefer to sell patents with higher visibility.
For example, user-interface patents and
social patents tend to present better than
semiconductor patents, and brokers select
packages accordingly. Other areas that
remain uncorrelated include the number of
offerings relative to the higher perceived
quality of the offerings – some of the better-
quality packages came from brokers that put
fewer packages on the market.‘Quality’here
is defined as how well the patents match
the target market identified by the broker’s
materials compared to our analysis of the
market, technology and patents.
Last year, we reported that brokers which
provide the highest-quality packages include
an evidence of use (EOU) with their packages
more than 80% of the time. This year, we
drew a link between the provision of an EOU
and a substantial increase in both the price
per asset and the probability of a sale.
Package flow
Generally, we believe that package flow has
increased substantially. We estimate the
number of new packages at 52 per month. The
most packages that we saw in any one month
was 48. However, our buying programmes
focus on particular technologies – meaning
that we do not see all packages on the market
all the time. This 52 per month number
conforms with how much of the market
we believe we are seeing and also with an
informal survey of some of the largest patent
buyers; it puts the overall market at 624
packages per year (last year we estimated 276
packages). We think that we underestimated
the number of deals in the market last year,
and that there are more deals in the market.
Figure 1 illustrates the number of new
packages that we received. The February
and March spikes relate to the start of new
buying programmes.
Private packages are not included in
our data set. Private packages were sourced
from our clients’ business development
departments and from private corporate-to-
corporate sales. Private packages typically
add between about 5% and 30% of the total
packages in a month.
Key diligence data
For a given buyer, few packages have
values that exceed the expected sales
price. Efficient buyers set buying criteria
to quickly eliminate packages that will not
meet the value versus price requirements.
Typical buying criteria categories are the
focus technology fields and a maximum
price, or price per asset. For example, a
buyer may say: “We will only review LTE
standards-essential patents, priced at no
more than US$3.5 million per family.” We
also continue to recommend using the bid
due date as a flexible buying criterion. It is
rare that a bid due date cannot be pushed
out far enough to allow for review.
The following section analyses the
application of buying criteria. Last year’s
article focused on pricing as the first
piece of data used to eliminate packages.
However, this year we looked more deeply
at our decisions to identify the more
common diligence path for a package.
We first looked at the technology fit
as the primary reason for passing (not
recommending a bid) (Table 2).
Why would “Technology area does not
fit” be the most common reason to pass?
Reasons might include the following:
• We have cast our net widely in order
to capture packages outside the focus
technology buying area, so some
packages will necessarily fall outside the
specific client need.
• The technology is the first and quickest
criterion to evaluate.
Adapt IP Ventures
Blue Sage Enterprises
Capital Legal
Epicenter IP Group
Global IP Law Group
Hilco Streambank
Hoffman Patent Firm
ICAP
Iceberg
Intellectual Profit
IP Offerings
ipCapital Licensing Company
IPInvestments Group
IPVALUE
Munich Innovation Group
NG Consulting
NextTechs
Patent Monetization Inc
Patent Profit International
PATEV Associates GmbH
Quinn Pacific
Red Chalk Group
TAEUS
Transpacific
UBM TechInsights (Semiconductor Insights)
Table 1. Broker list
• Evidence of use (EOU): An evidence of
use is a mapping of the elements of a
claim to a potentially impacted product
or service, together with reasoned
assumptions, including an identification
of any additional information needed
to complete the analysis. In contrast, a
claim chart includes further supporting
details, such as definitions, support
in the specification identified for each
term, reverse engineering concluded
and documented, and/or references to
terminology used in the prosecution.
• Patent assets: We define a ‘patent
asset’ as a single, unexpired pending
or issued patent asset somewhere in
the world. We have also found that
the number of issued US patents is an
additional relevant factor for evaluating
packages, so we have included both
metrics in many cases. Note that
pending applications are not included
in the US-issued count. Generally,
pending applications do not affect
the metrics unless there are very few
assets. Of course, there are situations
where a pending application is the most
important patent in the package.
Terminology

• When reviewed in detail, the technology
is unlikely to be adopted in the market
of interest.
• Although the package may be within the
technology area, the claims are not.
We still recommend starting with a
broader scope of technology for inclusion,
sampling packages and refining the scope of
technology accordingly. We find that clients
can better define their needs this way. Also,
a broader scope better protects a client’s
identity and exact search needs.
EOU issues represent 13% of passes.
Failures occur in a number of ways, but the
most common is a clear mismatch between
what the claim requires and what the
seller purports the claim to cover. Usually,
the problem arises where a key element
is missing, is used inconsistently or is
assumed to exist without justification for
the assumption. We found it rare that the
file wrapper affected the EOU. Firstly, the
EOU often failed before the file wrapper was
reviewed; and secondly, file wrappers rarely
resolve interpretation issues.
“Unresolved prior art” failures occur
from already known art or after we have
done a prior art search. Here, we believe
that we have found art that knocks out the
most important claims. Usually, we will raise
the issue with the broker and ask whether
they would like to see the art. Some brokers
do look at the art and provide responses
that change our minds; others do not
review the art and we pass on the package.
Unfortunately, we know of a few packages
that sold where knock-out art existed.
“Actual market adoption is too small”
falls into a challenging category for the
seller. When we look at actual adoption of
the claimed technology, we find that the
technology is used much less extensively
than sellers might hope. Usually, this arises
because the technology is not as compelling
as the seller thought. On a number of
occasions, we have been surprised to find
that a clearly interesting invention just does
not seem to be adopted widely – the market
being satisfied with ‘good enough’.
We examine pricing and bid due date
buying criteria below.
Turning to the remaining asset life,
many of our clients regularly have a
minimum remaining life criterion (eg,
minimum remaining life is five years).
Package technology
The packages that we reviewed fell into
diverse technology areas (Figure 2),
representing the desires of various clients
to address specific problems.
Out of the 296 packages, almost
half came from specific search projects.
Although the search projects of our clients
skew the data, we also see what happens
when a buyer does a targeted outreach.
If you are looking to buy patents in a
particular technology, you will be able to
find a healthy market offering them for sale.
Pricing data – improved pricing targets
For any package, one would think that the
first and simplest question to answer is
whether the package is within the buyer’s
budget. The broker’s pricing guidance
is critical to determining whether any
resources should be spent analysing the
package. Unfortunately, some brokers
simply do not supply a target price or
supply one much later than is helpful.
Consider how difficult the absence of
pricing information is for the buyer – are
the seller’s price expectations exceeding
market prices? Does the broker have a
good sense of the market? Or is the seller
just testing the waters? To focus diligence
resources, we ask for ballpark price targets
Figure 1. Number of brokerage packages (n = 296)
50
60
0
40
30
20
10
Jun-12
13
Jul-12
12
Aug-12
16
Sep-12
16
Oct-12
22
Nov-12
42
Dec-12
19
Jan-13
19
Feb-13
41
Mar-13
48
Apr-13
26
May-13
22Packages
Table 2. Reasons for passing on a package
Reason for passing % of time
Technology area does not fit 64%
Evidence of use fails to map properly 13%
Unresolved prior art 7%
Actual market adoption is too small 6%
Pricing 4%
Remaining asset life too short 4%
Bids due too soon 2%

to eliminate packages well outside of market
prices. We find that longtime brokers
have a good sense of market price. First-
time sellers and sellers that look only to
the public sales data for large deals for
pricing validation represent challenges. For
these sellers, we recommend testing price
targets before starting diligence. It is not
uncommon to see pricing targets drop by
more than 50% with some sellers during
the course of the sales process.
Pricing guidance continues to be
communicated as a broad range – for
example: “We think this package will
sell in the low seven figures.” That said,
we found more brokers willing to put a
specific price on the package. We still use
pricing buckets that correspond to these
broader ranges.
Pricing – the numbers
This year, 246 of 296 packages had pricing
guidance (87% this year as opposed to
56% last year). Importantly, some of this
increase was because we pressed more for
pricing information. However, brokers are
volunteering pricing guidance much more
frequently. Generally, the more frequently
that brokers offer packages, the more likely
they are to provide pricing guidance. We are
seeing a shift in how pricing information
is being provided. Previously, phone
calls seemed to be the preferred way to
communicate price. Now, we see brokers
provide pricing information in emails, with
some of the most experienced brokers
providing it in their sales packages.
To normalise the data, we then removed
unusual deals – specifically, we removed the
top and bottom 5% priced deals, reducing
our sample set to 222. Other factors can
skew the pricing (bigger packages can have
lower per-asset pricing), but we believe that
this method has allowed us to select a good
representative sample of the market.
Where pricing was available for
these 222 deals (see Figure 3), there was
a clear preference (73%) for deals in the
US$500,000 to US$2 million range. This
was also true last year.
To enable comparison with publicly
reported deals, we computed an asking price
per issued US patent, as well as an asking
price per listed patent asset. In doing so, we
used the midpoint of the price bucket as the
estimated deal price (eg, US$1.5 million was
used for the US$1 million to US$2 million
range). Two exceptions were made – one for
the US$20 million-plus range, where US$20
million was used as the price; and the other
for the US$4 million to US$10 million
range, where – like last year – the deal data
indicated that US$5.7 million, rather than
US$7 million, provided a better indication
of average pricing guidance.
We did not include a minimum price in
the data, as split packages appear to distort
the minimum price. Looking across all
deal sizes, we computed an average asking
price per listed asset of US$305,000 (as
opposed to US$344,000 last year), with an
average asking price per issued US patent
Figure 3. Package pricing for packages with broker provided pricing guidance (n = 222)
70
80
0
60
50
40
20
US$500k
50
30
10
US$500K-1M
67
US$1-2M
44
US$2-4M
27
US$4-10M
20
US$10-20M
10
US$20M+
4Packages
Figure 2. Package technology areas (n = 296)
100 20 30 40 50 80
23
60 70
27
2
60
6
10
5
9
6
2
8
27
11
24
76
Wireless
System infrastructure SW
Service provider networks
Semiconductor
Miscellaneous
Imaging
Displays
Development SW
Consumer electronics
Computing equipment
Components
Communication services
Communication equipment
Cloud computing
Application SW

of US$467,000 (as opposed to US$577,000
last year) (see Table 3).
We attempted some testing for packages
that sold and there did not appear to be a
substantially different asking price. However,
because packages take a long time to sell (see
below), this data is difficult to analyse.
Although the pricing seems to be more
consistent this year, the standard deviation
to be problematic. The first standard
deviation is 81% of the asking price per US
patent. Although better than in the past,
there is clearly a lot of pricing variation.
Importantly, the pricing guidance
continues to be much lower than what the
publicly reported deals might suggest. The
graph shows how badly the public data
distorts the average asking price per patent.
For public deals in 2010-2013, the average
price per sold asset was US$1.029 million
per asset (in 2010-2012 it was US$970,000
per asset) (see Figure 4).
A note of caution: these prices are
based on the broker’s pricing guidance, not
on actual closed transactions. If we apply
a discount of 35% to the asking prices,
we have US$198,000 per listed asset and
US$303,000 per listed US patent. We
believe that a 35% discount is reasonable
based on the deals that we have seen
close, as well as the prices paid when we
surveyed buyers. Last year, we used a 25%
discount, but we believe that 35% more
accurately represents the appropriate
discount. Our data indicates that the larger
packages have a lower price per asset
(approaching bulk pricing of US$50,000
per asset). However, we believe that
US$303,000 per US issued patent is useful
for deals with fewer than 10 US patents
(representing 75% of the packages).
EOUs impact on prices
This year, we also analysed whether the
package included a claim chart or EOU.
Approximately 33% of the packages had
EOUs.
EOUs are expensive and take time to
produce. However, they also allow brokers
to demand a 27% premium per US patent.
Assuming that an EOU can cost between
US$10,000 and US$40,000, the EOU
usually pays for itself in the premium
alone. The impact on sales is even more
pronounced and is discussed below.
Considering whether an EOU is
included in the package, pricing generally
aligns with one of three groups:
• No EOU or the EOU being for an
insignificant market results in below-
market pricing guidance (for package size).
• An EOU on a significant market, but
untested results in market pricing.
• An EOU which either has been tested
(ie, litigated or licensed) or targets an
exceptional market (eg, LTE essential)
results in above-market pricing.
Diligence and bid timing
Buyers need to know when they are
expected to have their diligence complete
and bids ready. Few packages are eliminated
from consideration because of short bid due
dates (2%), because buyers regularly ask for
– and receive – extensions.
Our data shows that packages are sent
out with an average of 70 days prior to the
bid due date (51 days last year). However,
we think that bid due dates are very
flexible. We are seeing some brokers go
with completely open-ended selling dates.
In more than a handful of cases, bidding
extended well past the one-year mark (see
the sales analysis below). This creates a
disincentive to bid early.
Most packages are received via an email
that provides a summary of the offering.
Many brokers include a link to a website
where additional data can be downloaded,
Table 3. Sold packages: broker asking price requested and impact of EOUs
Top and bottom 5% of data points
from each set removed
EOU top/bottom 5% removed
US$ per US
patent
US$ per asset US$ per US
patent
US$ per asset
Average US$467,230 US$305,180 US$594,240 US$421.33
Max US$1.5 million US$1.14 million US$3 million US$1.5 million
StdDev US$389,190 US$264,240 US$502.65 US$337,450
NumData 214 222 71 71
Figure 4. Per asset price (US$K) in relation to broker asking price (n=222)
US$500K
1200
800
600
400
200
0
1000
US$500K-
$1 million
US$1-
$2 million
US$2-
$4 million
US$4-
$10 million
US$10-
$20 million
US$20
million +
Per US patent
Per asset
Public deals
Table 4. Receipt to package disposition
Days from receipt
of package to
disposition (includes
purchase)
Percentage of
packages
= 7 9%
8 to 30 69%
31 to 60 17%
60 5%

such as PDF copies of the file histories and/
or EOUs. The most commonly missing item
in packages is an easy-to-use spreadsheet
listing all of the patent assets. It continues
to surprise us that packages arrive without
this. The better brokers provide the list
in a normalised form that is suitable for
direct copy-and-paste input into patent
analysis tools, such as IPVision, Thomson
Innovation and Innography.
The review period from when we first
receive a package until we close the review
of that package – including negotiating
and signing a patent purchase agreement
– averages 24 days (21 days last year).
As before, the average time presents a
misleading picture. Drilling down further,
four groupings of review time become
apparent in Table 4.
In under one month, 78% of packages
are closed out from further review. These
are packages that do not meet the buying
criteria. One can see the impact of clear
buying criteria: with 64% of the packages
not meeting the technology fit requirement,
these packages are eliminated quickly.
Number of assets per package
The market’s focus on smaller packages
continues. This year, 75% (78% last year)
of all packages had 10 or fewer US patents.
The average package size was 14 US patents
(eight last year) or 19 total patent assets (14
last year). We used a median of two US and
five total assets. However, we found that a
few bigger packages, with more than 100
assets, skewed the averages.
It is also helpful to consider the total
number of patent assets in a typical package
(all packages analysed), as shown in Table 5.
As before, most packages have a
handful of US patents – often only a
single patent family. Our data suggests
that higher-priced packages generally have
multiple families with a greater number
of foreign counterparts and often with
pending continuations.
We continue to believe that once
a package has more than 10 US issued
patents, the sellers benefit by segmenting
the package and identifying focus assets
in each segment. We find that smaller
segments help buyers to focus their
diligence resources. Without segmentation,
packages larger than 10 US patents or with a
wide array of families may be removed from
consideration early in the diligence process.
Bidding and buying
Switching to a new sample set of 287
packages, all received pre-2013, we analysed
package sales. The 287 packages included
packages that were analysed in our paper
last year. However, we have improved our
sales analysis methodology and changed
how we track sales, so revisiting these older
packages is helpful.
The percentage sales in the first nine
months of package receipt was 16% (16%
last year, 20% after adjustment; see below).
We use a window of nine months from the
date of package receipt to look for a recorded
assignment that indicated a sale of the patents.
The adjustment from 16% to 20%
in last year’s sales relates to partial
package sales. We built tools to automate
assignment checks for all US patents in a
package. This year, we checked all issued
US patents in all of the packages and found
partial sales. When we reviewed the 96
packages from last year’s sample, we found
that an additional 4% of the packages sold
as partial packages.
We define a package as being sold when
we find at least one patent from the package
reassigned to an entity other than that
selling the package.
What happens if sales take much longer?
Nine months from date of receipt seems
like a reasonable period for a package
Table 5. Sales package size
Number of assets Percentage of packages
(counting US issued only)
Percentage of packages
(counting all assets)
No US issued 3% 0%
1 34% 21%
2 to 5 30% 32%
6 to 10 8% 14%
11 to 25 10% 11%
26 to 50 7% 7%
51 to 100 3% 6%
101 to 200 4% 9%
Table 6. Sales date compared with bid due date
Months Percentage sold packages
where sale occurred within
this number of months of
receipt of package
Percentage sold packages
where sale occurred within
this number of months of bid
due date of package
Sold before due date 0% 4%
In bid due date month 0% 4%
Bid due date unavailable 0% 36%
1 to 3 15% 14%
4 to 6 18% 11%
7 to 12 36% 24%
13 to 24 18% 4%
24 13% 5%

to sell. Diligence, bidding and closing
should easily be completed within this
time. However, our broader assignment
checking suggests otherwise. Looking at
sales of packages from 2009 on, we see a
very different picture. For this analysis, we
continue with the pre-2013 sample.
Eighty-five of the 287 packages have an
assignment indicating a sale. This is a 30%
sales rate. So, if the seller is willing to wait
– and we recommend that the seller does
– it can improve its chances of selling the
package from 16% to 30%.
Looking at the sales date compared to
the bid due date, we see that 69% of the 85
packages sold within 12 months of receipt
of the package. The following year, 18% of
the sales took place, with 13% the year after.
Only a few packages (4%) sold before the bid
due date, suggesting that early bids are rarely
received or compelling. Also, 36% sold with
no bid due date provided by the seller. This
suggests that an open-ended process may be
an effective sales practice (see Table 6).
The importance of patience in selling
your portfolio is even more pronounced
when we look at very old packages. For
packages in the pre-2013 sample that first
went on the market in 2009, we see that if
you waited long enough, just over 50% had
sold as of mid-2013 (see Table 7). We will
continue to look at long-term trends.
For the 85 packages that sold, the timing
from the bid due date to the date on which
the assignment was recorded averaged 270
days (180 last year). Looking at all of the
steps, the process took 345 days on average
(173 last year) – specifically, from the
distribution of a package, as measured by
our receipt, to assignment recordation.
Who bought the patents?
For the 85 sold packages, we see non-
practising entities (NPEs) continuing to buy
the majority of assets at 56%, corporations
at 35% and defensive aggregators at 8% (see
Table 12).
The following are the definitions used
to categorise purchasers:
• ‘Company’ is a known operating
company (eg, publicly traded).
• ‘Aggregator’ or ‘defensive aggregator’ is
a known defensive aggregator such as
AST, RPX, OIN or Unified Patents.
• NPEs are all other purchasers.
Impact of EOU on closing
As we noted above, the impact of an EOU
on the price of the asset is pronounced
(27% increase in asking price). In our
previous paper, we hypothesised that an
EOU improved the chances of selling. We
have now specifically analysed our sales
data and know that there is a 52% better
chance of selling your patent when an EOU
is included. This improved chance of selling
– along with the increase in asking price –
clearly demonstrates that providing EOUs is
an effective practice.
In the pre-2013 sample used for the
assignment analysis, 88 of 287 packages
(31%) had broker-provided EOUs. Of
those 88 packages with EOUs, 34 (38%)
sold. Contrast this with 51 sales in the
remaining 199 packages, or a 25% chance
of sale (Figure 5).
Brokers and sellers should spend time
developing an EOU for select patents. Even
when buyers do not rely directly on the EOU,
it helps to clarify where the seller believes
that the value in the package is focused.
Lots of NPE litigation?
Last year, we reported that few of the
purchased patents were litigated (two out of
15) after the sale. This year, looking at all of
the patents in all of the sold packages (pre-
2013 sample) gave a similar result. Of the 85
sold packages, only seven had patents with
litigations after the sale. NPEs litigated at
a higher rate – five out of seven litigations.
However, the NPEs litigate a much larger
number of times using patents from the
same package. NPEs tended to start a
median of three litigations per package. The
mean was much higher because one NPE
initiated 19 separate litigations.
The two corporate litigations were both
counter-assertions: Facebook defending
against Yahoo! and Salesforce defending
against Microsoft. Based on the assignment
records, it appears that in both cases the
defendants in the initial litigation went
to the patent market to buy patents.
Interestingly, Salesforce bought the patents
after Open Invention Network had bought
them. We continue to find the brokered
patent market an effective tool for finding
patents to use in counter-assertion.
Table 7. Long-term sales view (2009
packages)
% sold as of Cumulative % of
2009 packages sold
31/12/2009 3.57%
31/12/2010 21.43%
31/12/2011 42.86%
31/12/2012 48.21%
30/6/2013 51.79%
Figure 5. Pre-2013 sample: purchases by type
(n=85)
NPE 48Company 30Aggregator 7
Table 8. Brokered patent market (200 assets)
Number of packages per year 624
Sales rate 30%
Sold packages/year 187
Price per US issued patent US$303,000
Number US issued per package 5
Average sales price per package US$1.515 million
Total market US$283.305 million
Total commission US$56.6618 million

Also of interest is that 13 (15%) of the
85 sold packages had had patents litigated
before the sales. However, none of these 13
appears to have been sold because of the
litigation – in other words, the buyer was
not the target of the lawsuit. Only 15%
previously litigated may at first appear low,
but when one considers that between only
1% and 2% of patents are ever litigated, we
see that litigation history is an important
factor in sales.
How big is the 2013 market?
Again, we extrapolated our data on pricing,
sales and packages to determine the size of
this market. We size the market for deals
with fewer than 200 assets in the package,
sold by brokers, to the open market. We
estimate US$283 million (US$153 million
last year) in annual sales, with broker fees
of about US$57 million (US$34 million last
year). The biggest differences from last year
are as follows:
• There is twice the number of packages.
• The sales window has been extended
from nine months to 36 months.
To derive the market size, we first
estimated the number of new packages
brought to the market in a year (52 a
month). We then used the 36-month sales
rate (30% sell), recognising that the number
could be as high as 50%. We discounted the
sales price from the asking price (65% of
the asking price). We believe that the actual
sales price is one of the more difficult areas
to estimate because so few transactions
are reported. We adjusted the number of
US issued patents per package to take into
account the large number of small packages
and the pricing differences of big packages
versus small packages. We believe that
the price and number of assets could be
50% below or above this number. Using
the adjusted number of issued US patents
per package brings us to a total market of
US$283 million per year.
Using an average commission rate of
20% (22% last year), the revenue from this
market for brokers is US$91 million per year
(see Table 8).
We back-tested the market size by
estimating the average loaded labour rate per
broker (US$300,000 a year). This gives 189
brokers. Assuming that four brokers work
in each brokerage shop, this results in 47
brokerages (our data shows 55). Each brokerage
would bring about 13 packages to the market
per year; however, our data shows that a few
brokers bring many deals to the market, with
the majority bringing a few deals.
Opportunities
Our conclusions do not differ significantly
from our previous paper. With more than
50 new packages coming to market each
month, the opportunity to find something
to fit your needs exists. Nearly 70% of the
packages have a target price below US$2
million, allowing companies to purchase
assets to address specific problems at a
relatively low cost.
Even extending the window to 36
months to look for sales, we still see a low
percentage selling. The broker business can
be tough. Corporate buyers continue to be
seen as significant buyers, but NPEs make
up the majority of buying. Litigation post-
sale seems lower than one might expect,
given the NPE purchasing.
We continue to evaluate, bid on and buy
patents for our clients, and will continue
to track the data and refine our database.
This being our second year of reporting on
our data, we have had the opportunity to
ask more penetrating questions (eg, what
Summary of data
Packages studied June 2012 to May 2013 296
Number of issued US patents 4316
Total assets 9359
Packages old enough to have sold Q1-Q3 2012 112
Percentage sold in first nine months 16%
Asking price per US issued patent US$467,000
Asking price per listed patent asset US$305,000
Average number of US issued patents per package (excluding packages with over 200 assets) 14
Median number of US issued patents per package (excluding packages with over 200 assets) 2
Percentage of packages with 10 or fewer issued US patents 67%
Average days from receipt to disposition by ROL Group 24
Annual sales US$283 million
Receipt of package to assignment record 345 days
Number of people employed as brokers (est) 189

happened to all the patents we saw, not just
one representative patent). As soon as the
data started answering these questions, new
detailed and in-depth questions presented
themselves. We look forward to continuing
to answer these.
Action plan A
When buying patents:
• State the business case for buying –
what problem are you solving?
• Model a return for your buying
programme.
• Your buying operations need to reflect
that over 90% of the patents will not fit
your needs – eliminating those patents
from consideration early will greatly
reduce your costs.
Programme parameters include:
• Timeline.
• Budget.
• Buying team authority and
responsibilities.
• Buying criteria.
• Listing of acceptable sources of patent
packages.
• Special requirements such as a whitelist
of unlicensed companies.
Fail-fast triage process for eliminating
undesirable packages quickly:
• Extract criteria from the business case
to identify interesting markets and
technologies, and define the diligence
needs.
• This is a multi-part analysis of markets,
technical knowledge and legal analysis
where a failure in any one area
eliminates the package from further
review.
• Track basic information about your
programme so that you can learn from
your past.
Bidding and buying:
• Your valuation model determines a
maximum bid price.
• Assume that diligence will take longer
than planned.
• Consider adding a consulting
agreement with the inventors if they
are available.
Kent Richardson and Erik Oliver are
partners in the Richardson Oliver Law Group
kent@richardsonoliver.com
erik@richardsonoliver.com
The authors gratefully acknowledge Lex
Machina’s assistance in the litigation portion
of this analysis. www.lexmachina.com

�ݺ�ߣ

IAM_63_The Brokered Patent Market - Kent Richardson and Erik Oliver - from IAM

More Related Content

IAM_63_The Brokered Patent Market - Kent Richardson and Erik Oliver - from IAM