This presentation was provided by Bert Carelli of TrendMD, during the NISO event "Researcher Behaviors and the Impact of Technology," held on March 25, 2020.

1. Promoting Content Discovery Within the Reader/Researcher
Workflow
Bert Carelli
Director, Partnerships

2. About TrendMD
Founded in 2014 by professionals from academic research, scholarly publishing, and
digital technology
Nurtured by Y Combinator — the startup accelerator that incubated Reddit, Dropbox,
and Airbnb.
42 employees, with management in Toronto, California and Germany
Used by over 400 scholarly publishers on nearly 5,000 websites

3. 8,000+ new articles per day
50% of articles are never read (Eveleth, 2014) and a much higher percentage are never cited (Meho, 2007)

4. Publishers, Authors
What can I do to promote discovery of my
content?
Researchers
How do I find the content that will be
most interesting and useful to my work?
Two Sides of the Problem

5. Browsing is core to discovery
“Browse and search are becoming more blurred and more similar.” How Readers Discover Content in Scholarly Publications, Gardner & Inger, September
2018
A specialist
bibliographic
database (A&I)
4.0
3.0
2.0
1.0
0.0
5.0
6.0
When you wish to view the latest issues of your favourite online journals, how do you navigate to those journals?
Library web pages,
library discovery
tools, library
search engines
A journal collection
such as EBSCO,
ProQuest
A social or
professional
networking site
A publisher’s
website
By following links
in journal alerts
The journal’s
homepage
A general web
search engine
An academic
search engine
E.g. Google
Scholar
A scholarly society
web page
2005, n=413, ±0.20 at 95% confidence 2008, n=762, ±0.15 at 95% confidence
2012, n=2340, ±0.09 at 95%
confidence2015, n=4961, ±0.06 at 95%
confidence
2018, n=1980, ±0.09 at 95%
confidence

6. Browsing on publisher websites
“Related articles remain the most popular feature of a publisher website.”
How Readers Discover Content in Scholarly Publications, Gardner & Inger, September 2018

7. TrendMD recommendations add browsing to the reader’s flow
Readers discover new content,
based on:
What they are reading
What other users like them
have read
What they have read in the
past
3
1
2

8. TrendMD generates recommendations via collaborative filtering - similar to how
Amazon product recommendations are generated

10. ‘People who read this also read this’

11. A/B tests over 6-weeks; read the full report
Quality of recommended articles: TrendMD outperforms PubMed related
citations by 272%
5.0%
Week 1
Collaborative filtering (TrendMD)Semantic similarity
4.0%
3.0%
2.0%
1.0%
0.0%
Week 2 Week 3 Week 4 Week 5 Week 6

12. A cross-publisher network
Find new readers:
Grow your readership by showing links to your
articles across top scholarly publications
Increase page views:
Increase article views 4-6% by recommending
additional relevant content to all visitors
Increased article usage
measured by:
Article pageviews
Mendeley readers
Altmetric scores
Citations
1 2
Your content
Other publishers in
TrendMD Network
Non-sponsored click
Sponsored click
Sponsored click
Your content

13. TrendMD is used on over 5,000 scholarly websites by 300+ publishers
Publishers Integration partners

14. 180M+
Monthly Uniques Globally
10M+
Monthly Clicks
40%
Mobile
60%
Desktop
90%
US MDs reached
85%
US academic reached
TrendMD audience reach

15. Country Sessions Percent
United States 49.0M 56.36%
United Kingdom 9.5M 10.93%
India 6.2M 7.13%
Canada 5.2M 5.98%
Australia 3.6M 4.14%
Philippines 1.6M 1.84%
Germany 1.4M 1.61%
China 1.2M 1.38%
Japan 1.1M 1.27%
South Africa 896K 1.03%
France 829K 0.95%
Netherlands 811K 0.93%
Spain 800K 0.92%
Italy 783K 0.90%
Brazil 714K 0.82%
Singapore 683K 0.79%
Nigeria 675K 0.78%
Malaysia 670K 0.77%
New Zealand 639K 0.74%
Ireland 637K 0.73%
Global Distribution

16. TrendMD Traffic Shaping
Professional & Enterprise Plans
Goal Tactic
Increase Impact Factor Promote only content from the previous 2 years.
Promote articles based on Most-Read, Most-Cited, or most-
mentioned in altmetric data.
Launch a new site Use credits from high-traffic sites to promote articles from new site
Highlight a sub-discipline or conference Use credits to promote a collection, an issue, or selected articles
Increase author submissions Use backlink data to determine to target authors publishing in 3rd-
party journals
Increase institutional subscription sales Use custom TrendMD analytics data on where new readers are
coming from, how many are returning, what articles they read on other
journals, etc.

17. Impact: TrendMD traffic increase is correlated with 50% increase in mean citations
over 12 month study
Control
0.0 5.0 10.0 15.0 20.0
Mean citations over 12-
months
“The Citation Advantage of Promoted Articles in a Cross‐Publisher Distribution Platform: A 12‐Month Randomized Controlled Trial”; Kudlow et al; Journal of the Association for
Information Science and Technology, First published: 23 December 2019, DOI: (10.1002/asi.24330)

18. 0 10 20
TrendMD significantly increased citation counts in Health and Medical Sciences (82%),
Life Sciences and Earth Sciences (80%), and Physics and Mathematics (59%)
Control
TrendMD
Health and Medical
Sciences
Social Sciences
Chemical and Materials
Sciences
Business, Economics,
and Management
Humanities, Literature
and Arts
Engineering and
Computer Science
Life Sciences and Earth
Sciences
Physics and Mathematics
30 40 50
“The Citation Advantage of Promoted Articles in a Cross‐Publisher Distribution Platform: A 12‐Month Randomized Controlled Trial”; Kudlow et al; Journal of the Association for
Information Science and Technology, First published: 23 December 2019, DOI: (10.1002/asi.24330)
60

19. Control (n=198)
0 2 4 6 8
Read the Full Report in the Journal Scientometrics
Mean Mendeley saves over 4-weeks
(n=197)
Impact: TrendMD traffic increase is correlated with 77% increase in Mendeley saves

20. Reader Engagement: TrendMD readers view more articles per visit than all other
traffic sources
0 1 2 3 4
Pages per session

21. Reader Engagement: TrendMD readers have the lowest bounce rate of all traffic
sources

22. Attracting authors
0.0% 10.0%
Q3 What finally convinced youQ2 How did you hear
Read a paper
published by BMC
Published previously
with BMC
Asked to peer review a paper
published by BMC
Recommended by
colleague / friend
Saw an
advertisement
Read an article in scientific
or general press
Received a call for
papers by e-mail
Received a call for
papers by direct mail
Heard presentation/received visit
from BioMed Central
Received information from my
institute/library
Saw BioMed Central at
a conference
Received an abstract
book of journal articles
20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0%

23. Case study: JMIR Publications uses TrendMD to drive author submissions
Goal
JMIR Publications, is a leading eHealth
publisher of 24 Open Access journals
Ue TrendMD to drive author submissions
Strategy
JMIR Publications launched a TrendMD
campaign to promote 500 articles from the
Journal of Medical Internet Research (JMIR
IF=5.175) to targeted researchers in the
TrendMD Network
Links to JMIR content were featured in
TrendMD article recommendations, across a
network of >4,000 journals such as The BMJ,
Lancet, Pediatrics, and others
10,000 clicks were purchased for the campaign
As controls, JMIR launched 2 additional
campaigns: Google AdWords and Facebook,
with the goal of driving author submissions
Conversion rate vs Source of referral
0.00% 0.10% 0.20% 0.30%
Results
TrendMD campaign resulted in 26 author submissions at a conversion
rate of 0.26%
ROI was nearly 10x
TrendMD outperformed Google Adwords by >2x and Facebook by >5x
Source of referral Sessions Author submissions Conversion rate
TrendMD 10,000 26 0.26%
Google AdWords 5,000 6 0.11%
Facebook 12,500 7 0.05%
Grand total 17,500 12 0.08%

24. Targeting
While TrendMD finds those readers most likely to be interested in your content, finer
targeting is available under Enterprise Plans
Country or region (Global)
Institutional (Global)
Target hospital networks, universities, colleges, organizations
User-targeting (Global)
Target specific types of researchers and/or HCPs in the network, globally

25. Case study: TrendMD drives HCP signups
Goal
Drive new HCP signups to Elsevier PracticeUpdate
- http://www.practiceupdate.com/explore/
Strategy
Elsevier Practice Update installed the TrendMD
widget and launched a campaign targeting HCPs
Practice Update content was promoted in
TrendMD network
Results
Pilot campaign resulted in over 500 new HCP
signups in first month
Over 7,500 HCPs were directed to Elsevier
Practice Update in first month
Conversion rate was >5%
Breakdown of HCP signups to Elsevier Practice Update
from TrendMD referrals:
Professions for TrendMD,
All channels: Jan 2017
0 25 50 75 100 125 150 175 200 225
Physician
Consumer/Other
Student
Other Health Professional
Faculty/Educator
Researcher/Scientist
Nurse
Fellow/Resident
Nurse Practitioner
Pharmacist
Physician Assistant
Media/Press
Optometrist

26. Case study: TrendMD<>Merck promotion of Nexplanon
Background
In the sample data below, TrendMD was one of two
sources of referral for a microsite sponsored by
Merck.
Strategy
Merck’s goal was to promote the use of Nexplanon
to targeted physicians and drive conversions which
were measured as email signups.
The brand had a small match list of 8,464
physicians with NPI numbers.
TrendMD initially matched to 96% of the list.
Results
500 clicks were purchased from TrendMD with
visitors only coming from the match list.
The conversion rate with TrendMD was the highest
compared to all other channels (3.71%)
TrendMD alone succeeding in reaching over 5% of
the original match list provided by Merck.
Conversion rate vs Source of referral
0.0%
Source of referral Sessions Pageviews Pages/session Bounce rate TOS Conversions Conversion rate
Google AdWords 180 224 1.24 76% 0:01:15 1 0.28%
Facebook 220 276 1.25 82% 0:01:22 2 0.90%
Email 73 161 2.21 62% 0:02:03 1 1.30%
TrendMD 500 1116 2.23 47% 0:02:14 19 3.71%
Grand total 973 1777 1.83 67% 0:01:44 22 1.55%
Email
1.0% 2.0% 3.0% 4.0%

27. Article-level Campaigns - Promoting Time-Critical Content: Covid-19
The Problem: In a glut of unreliable information, with the cancellation of scientific
conferences, how can a publisher get their vital information noticed?
Publisher #1
Started campaign Feb 1, starting with 16 articles with keyword ‘Corona’
Total budget of $2000
9.6 million impressions and nearly 2,400 clicks (pageviews) in February
Publisher #2
Started campaign Mar 6 to promote 10 articles related to the coronavirus
Initial $1,000 budget, set with weekly limits.
March 10 - Upon seeing high demand, increased weekly budget to $500
March 17 - Increased budget to $1,000/week
As of March 18, 5.6 million impressions, and over 1,600 clicks.

28. Thank you!
Bert Carelli
Director, Partnerships
bert.carelli@trendmd.com
510-551-3784