"It is an immutable law in business that words are words,
explanations are explanations, promises are promises
but only performance is reality."
-- Harold S. Geneen, CEO of ITT Corp.
This aspect of management has a bad name. some people shudder at the
suggestion of authoritarianism. Yet, it is essential to success and needn't be done
in a dictatorial way. The alternative to controlling is "out of control".
We also see the activities here as filling a need for project administration
to be accountable -- to its membership and to the general public.
Managerial controlling consists of assessing and regulating work in progress and
evaluating results attained, taking action when indicated.
The two basic methods of controlling are
- By inspection -- allows rapid correction of any deficiencies, but takes
much effort, is expensive and tends to limit others' initiative
- By exception -- only matters outside of standards need attention
"Without a standard there is no logical basis for making a decision or taking action."
-- Joseph M. Juran, professor and consultant
Managers set performance standards to determine the criteria for
measuring work and results. Standards need, as
objectives do, to be measurable
Some managers are reluctant to commit to tangible & measurable criteria.
They may fear revealing deficiencies or be uncertain of the level of
performance required. However, administering without standards is like archery
without targets; how do you know if an arrow flies true?. Or, like steering a
ship without a destination.
Performance standards should
- flow from the enterprise's objectives,
- be tangible and measurable,
- have tolerance limits built in,
- be realistic,
- be written and widely communicated, and
- be reviewed periodically or as conditions change.
In order to serve their purpose,
they must receive understanding and acceptance. Standards that don't meet these criteria create more problems than they
purport to solve.
TFG Performance Standards
We've developed objective measures (metrics) for assessing performance.
We recommend this
link. After several years of monitoring the metrics (the rare project to
do so), we may now propose tentative standards for assessing performance.
Measurements alone are not, technically, standards because a
standard requires a
minimum acceptable performance statement. Although more of each
is "better" and less is "worse", we hadn't fixed
targets, but we think we've identified the critical metrics. Analyzing the
numbers over time has suggested possible targets.
Assessment as of Summer 2016: Size is adequate for some purposes, but see penetration
- Total members, all test types
- Y-DNA (STR)
- Standard: None specific
Penetration is the ratio of DNA-tested Taylors to the Taylor population,
usually expressed as number per 100,000.
For more about penetration,
see this page.
- USA & world-wide
- United Kingdom (England, Scotland, Wales & Northern Ireland)
- Standard: For yDNA, 80 per 105 Taylor males.
Standard: 80 per 100,000 (10^5) is a possible penetration index
Assessment Summer 2015: US penetration is adequate, but UK & Ireland
penetration is seriously deficient.
Resolution measures the adequacy of tests to the purpose of finding genetic cousins
Standard: For ySTR, at least 80% testing to 37 markers, 50% testing
to 67 markers.
- Percent w/ >12 markers, >=37, etc.
- Average number of markers tested
For ySNP: ?
Assessment Summer 2016: Good (better than most projects) and continuing to
Participation measures the extent to which members contribute to
Standards: EKA >= 90%, GEDCOM >= 25%; Trees >= 50%.
- Earliest-known (most distant0 ancestors stated
- GEDCOMs uploaded to FTDNA
- Taylor lineage trees submitted to project
Assessment Summer 2015: Likely to be a continuing deficiency, as many
members are little motivated. The admin challenge is to
(In truth, admins can not change the DNA with which they work; they can
only do a competent job at seeing who matches who.)
Standard: >= 70% of qualifying
members match 1+ other members.
- Match rate, # matches over # ySTR
- Number of genetic families identified (& as to compared to estimated
- Number of estimated Taylor lineages in project (an abstract
estimate). See also below
Assessment Summer 2016: Overall rate better than 50%, >70% after
adjusting for inadequate resolution, NPE &
NTP., Less than 30%
of those who could conceivably match are matched;
this seems a reasonable attainment.
Comment: Match rates seem to greatly depend on penetration &
resolution. Achieve good performance on both and you'll have a
good match rate; at low performance for those, the match rate for most surnames is a
Standards need also be set for various aspects of project work, such as
important one concerns what is considered an adequate ("significant")
match and what is not.
"The only man I know who behaves sensibly is my tailor;
he takes my measurements anew each time he sees me.
The rest go on with their old measurements and expect me to fit them."
-- George Bernard Shaw
Managers measure performance by recording and reporting the work
done and results attained. The reports must be accurate, timely,
concise and understandable. Periodic reporting and consistency over
time provide comparability; improvements and regressions become apparent.
The type of measurement will vary with the enterprise. A complex management
information system may be needed, or the manager may keep a log.
Some data the admin will need is easily available; some may be difficult to come by.
Sources we've used include:
- FTDNA project statistics page,
This page has most of the statistics we use. We periodically download it in
a XML spreadsheet, then copy and paste the data into an Excel file
containing (also) prior data.
- FTDNA Member Information page,
We make (somewhat detailed) notes (e.g., test types, matches, residence) on each member in a standard format,
then periodically download into a XML file for analysis.
- FTDNA Y-Results page,
This page, downloaded periodically as a XML file & converted to Excel,
provides the data on matching -- if the admin has ketp current
with the matching and grouping task.
- Trees: As this information is not in FTDNA purview, we must
generate the data ourselves. Basically, we just ocunt the trees.
- Comparables: Each comparable project's total membership is displayed
on its public FTDNA page. They number of ySTR tests is more a problem;
many projects do not choose (an oversight?) to display this statistic.
TFG Performance Measurement
We haven't set standards, but we have defined their metrics. The reason
(excuse?) for not setting specific targets is that we
ventured into unknown territory. (Only one other project admin, to our
knowledge, has gone so far as to describe performance metrics.)
It is recommended to periodically (monthly or quarterly) download information from the
FTDNA GAP. (For a "how-to", see the
page on Performance Monitoring.)
- Project statistics
- Member information, with admin notes
- Y-DNA results
- mtDNA results
This data provides a snapshot of the project's current status. An ongoing
series of snapshots makes a "progress movie". See "Project Status"
and scroll down to the performance section. For example,
Depicts improvement in ySTR resolution from Jan 2009 to present
"In God we trust, all others bring data."
-- W. Edwards Deming
Evaluating performance consists of analyzing, interpreting, and
determining the worth or quality of the work done and the results. Part of
evaluating performance is determining whether corrective action is needed.
Evaluation is ever so much easier with good performance standards.
TFG Performance Evaluation
See "Assessments" above.
- Project size: Size is adequate, but see penetration
- Penetration: US penetration is adequate, but UK & Ireland
penetration seriously deficient
- Participation: Likely to be a continuing deficiency, as many
members are insufficiently motivated. The admin challenge is to
- Matches: Seems reasonable. Overall rate better than 50%, >70% after adjusting
for inadequate resolution,& NPE &
NTP., Less than 30% of those who could conceivably match are not matched
with at least one other project member.
"What gets measured gets done,
what gets measured and fed back gets done well,
what gets rewarded gets repeated."
-- John E. Jones
"...If you can't control it, you can't improve it."
-- H. James Harrington
Correcting performance follows on the heels of evaluating it. It's about
rectifying or improving the work being done and the results. The direction
and degree of course correction -- with proper standards and monitoring --
should be obvious from the evaluations.
This may involve technical changes, a procedural adjustment here, a re-training
there. Or, it may require management action -- better communication, new or improved plans or
organizational structure or system of controls. The latter are typically longer-term
TFG Performance Correction
The next steps, dear new admin, in correcting the project's performance are up to you.
Back to the core