We need a research architecture and better tools

by Kris_Tuttle on June 3, 2008

We’ve been doc­u­ment­ing a broad range of sources and ser­vices we have adopted to per­form our reg­u­lar research tasks.  In the process we have hit upon what we think is an impor­tant ques­tion for us and any­one else who is actively invest­ing cap­i­tal in the pub­lic markets.

Years ago the hedge funds helped cement the con­cept of the “mosaic” for a par­tic­u­lar stock idea or mar­ket con­di­tion.  When look­ing at buy­ing or sell­ing a name these port­fo­lio man­agers con­sulted a few sources that cre­ated a pic­ture for them from a set of well defined aspects.  The clas­sic case was a set of inputs that sug­gested near-term busi­ness was track­ing bet­ter than con­sen­sus expec­ta­tions, insid­ers are buy­ing, new ana­lysts have been vis­it­ing the com­pany imply­ing that new cov­er­age is com­ing, and the val­u­a­tion is attrac­tive.  A set of con­sis­tent pos­i­tive inputs means an infor­ma­tional mosaic that says “buy.” As we have said many times before a suc­cess­ful analy­sis includes fun­da­men­tals (busi­ness trends), expec­ta­tions (con­sen­sus think­ing) and valuation.

There’s been a huge pro­lif­er­a­tion of infor­ma­tion sources, sources that vary by type, scope and qual­ity.  There’s also been a con­tin­u­ing trend of increased spe­cial­iza­tion.  Now insti­tu­tions reg­u­larly con­sult firms like Ger­son Lehrman for inter­nal morale or insight on a poten­tial invest­ment, use another to do chan­nel checks, get mod­els and con­sen­sus think­ing from bro­kers, bring in raw data from ser­vices like NPD, or Majes­tic, com­mis­sion a cus­tom sur­vey, use a too like Bloomberg, or Reuters for insider trans­ac­tions, lock-up expi­ra­tions and so on.   Some are using con­sumer ser­vices like Google and Yahoo more to get a retail view of stocks and the mar­ket.  Add to that the grow­ing num­ber of inter­est­ing alter­na­tive sources of infor­ma­tion like Monitor110, Covestor, Stock­pickr and other ways to glean changes in sen­ti­ment or mar­ket adop­tion.  The full range of sources often also shifts based on whether an insti­tu­tion is in “idea gen­er­a­tion” or “due dili­gence” mode. 

As we begin to doc­u­ment and semi-automate aspects of our own invest­ment process we are forced to con­sider some archi­tec­ture to pull these diverse and chang­ing ele­ments into our uni­verse of knowl­edge and analy­sis so we can fur­ther opti­mize our port­fo­lio and the qual­ity of our decisions.

Every week we find and try to incor­po­rate more good infor­ma­tion sources, addi­tional indus­try experts and smart peo­ple, dif­fer­ent meth­ods of get­ting and shar­ing infor­ma­tion (first email then Skype now Twitter?)

We’re try­ing to incor­po­rate a few over­lap­ping lay­ers and a fairly com­pre­hen­sive view but even doing the basics begs for a sim­pler way to match fun­da­men­tals with con­sen­sus and val­u­a­tion.  Why aren’t the online tools bet­ter?  Who has a sim­ple and flex­i­ble archi­tec­ture to put all these research ele­ments in a con­text for deci­sion making?

Bet­ter tools are needed.  Right now it looks like we need to develop most of what we want our­selves.  If noth­ing bet­ter appears we’ll pro­duce them for the rest of the world when we are done.  Mean­while, we wel­come any good ideas.

– Kris Tuttle

Tags: , ,

Tagged as: Research

Comments on this entry are closed.

Previous post:

Next post: