Tobacco Documents Online

BROWN & WILLIAHSON TOBACCO CORPORATION INTERNAL CORRESPONDENCE TO: CC: FROM: DATE: SUBJECT: R. A. B1ott T. E. Whitehair, Or. A. O. Mellman O. 0, Woods B. R. Pellett December 23, 1985 FAVOR SMOKE-FREE CIGARETTES-- RESEARCH OUTLIHE E. E~_]~olzl~I o r_~lL__ ..... ~ec~.:f S~( BAT ~Sout hm;t:~n }'~ BACKGROUND Advanced Cigarette Products, Inc. has introduced a new product: Favor. This unique entry into the cigarette marketplace is shaped like an ordinary cigarette, but contains no tobacco leaf. Each stick contains nicotine and "tobacco flavors", and the user simply inhales through the "filter" without lighting the cigarette. Packed six to a box, this smokeless product is positioned as providing the satisfaction of smoking without the social, legal, and physical (matches and ashtrays) restrictions of cigarettes. Advertising is via TV, outdoor, and newspaper In the Dallas, Houston, and Austin areas of Texas. MRD has been requested to monitor Favor's performance. RESEARCH OBJECTIVES There are two major questions to be addressed by marketing research: 1. Consumer reaction to the offering 2. Method and extent of distribution in the trade RESEARCH OUTLINE MRD proposes that a preliminmary Awareness and Usage Study be conducted with a total of 500 smokers to measure the impact this new product has had on smokers in the test area. Questioning would concern: Awareness of Favor (product) Awareness of Favo~ advertising - Media Favor advertising seen in Attitudes towards smokeless cigarettes Purchase Interest Trial Usage (occasional and regular) 0 C li b P D F ~wwwTfa'stioTc'o'm

-Z- If the results of the preltnflnary Aau Study show sufficient consumer lmpact, then a l~o-pronged approach w|ll be used to address the above questions. le Continue monitoring consumer behavtor via additional waves of A&U or a purchase panel study. 2. Determine distribution panels vta sales force information. - If Favor Is being distributed through tobacco wholesalers then ICC data can be uttllzed for measuring wholesale activity. - If Favor ts being handled through food and drug d|str|butors perhaps a short-term measurement through syndicated distribution services such $AMI or Ntelsen could be utilized. Alternatively, In-store Interviews with store personnel and/or purchasers of Favor may be necessary tf awareness and distribution are Inordinately low. NEXT STEPS A schedule and final btds for the preliminary A&U will be avatlab]e by January 2, 1986. Inquiries are being made to B&W sales force regarding distribution channels of Favor. A fo~] proposa] wl]l be ]ssued no ]ater than January 6, 1986. /BRPOOOO6:fmr B. R. Pellett~ 0 C~ C li b P D F - ("v vv v:v. f a s t i-o-c'o'm

KAX [U.K.and Export) Limited RESEARCH & DEVELOPMENT CENTRE SOUTMAMPt'~ |NG ~.,J~O TO !~ A L HEARD REF CCG/JP FROM DR C C GREIG DATE ZO January 1986 Re - FAVOR AS A "NICOTINE-ADDING FILTER" - FURTHER WORK A previous note for you showed that FAVOR (Lights) elements acted as a relatively low efficiency filter for TPM but added nicotine to smoke. The amount of nicotine added to smoke was higher ~the amount that would have been derived from passage of the same amount of air, leading to the surmise that the many chemical species in smoke caused increased nicotine transfer. However, the author was concerned that the data did not indicate where, in the whole process of smoking the cigarette, such increased transfer occurred. On purely chemical grounds it appeared likely that, with the hotter, more humid smoke of the last two or three puffs, the bulk of the transfer might be occurring there. If true, this implied that the apparent benefit of an increased nicotine delivery would not apply where most needed, in the earlier puffs. An experiment to clarify this has been performed using the cigarette construc- tions in the original note. Deliveries were measured on a per puff, ie puff- by-puff, basis from 15 cigarettes and, as is now established, these data do not necessarily agree with standard smoking conditions, but are internally consistent and are presented in Table I. The graphic form (Figure I) shows that the author's fears were unjustified and that transfer is almost equally enhanced during the whole smoking process. This finding confirms that FAVOR is indeed a viable candidate "filter" for adding nicotine to smoke. However. the amounts added by this FAVOR Regular element are far greater than is likely to be required for product amelioration. Thus, taking 15% as a just noticeable difference (JND), the kind of increases measured here can amount to 5 JNDs and would probably utterly unbalance the smoke even If, as in the present instance, transfer of flavourants from the FAVOR element were not concurrent and confusing. The results do the following: (1) They confirm the earlier small-scale investigations'o{ increased smoke nicotine; (2) They suggest a preferential increase in the earlier puffs; (3) They suggest that use of the FAVOR element in a market place product would require (a) either a lower content of nicotine per FAVOR element or (b) a shorter length of element at the same loading. For mechanlcal reasons, (a) is likely to be preferred. (4) They confirm that, for behavioural studies where it is deslrable to alter nicotine in isolation, the FAVOR element offers only the possibility of nicotine fortification. Thus if a set oF nicotine:tar ratios of low, medium and high are required from a given blend, this blend will have to be the low nicotine:tar ratio blend plus FAVOR in two increments to give medium and then high. Unflavoured but nicotine-containing elements would be required for these products. C~ C7~ -.0 0 CJ~ co .. C Ii b P D F~w~,','Jwda'st~i'o-.c'o'm - -

-2- TABLE I Puff Number 1 2 3 4 5 6 7 Sum (a) Standard SEFK, TPM 20mm CA filter Nic PMWNF 0.89 1.14 1.45 1.75 1.80 2.00 2.59 11.9 0.053 0.090 0.I10 0.133 0.148 0.160 0.180 0.87 0.83 1.04 1.29 1.53 1.55 1.69 2.16 10.4 (b) SEFK with lOmm TPM CA filter NIc (lOmm removed) PMWNF 1.09 1.49 1.84 2.15 2.51 2.69 4.16 15.9 0.070 0.120 0.147 0.153 0,180 0.193 0.220 1.09 0.99 1.31 1.59 1.83 2.12 2.22 3.55 13.6 (c) SEFK with 10mm TPM CA filter andNic 9mm of FAVOR PMWNF Regular element 1.13 1.35 1.58 (1.82) 1.99 2.36 (2.85) 13.1 0.170 0.213 0.230 0.240 0.251 0.272 0.293 1.67 0.92 1.06 1,27 (1.47) 1.59 1.89 (2.27) 10.5 { ) - Values Interpol atod/e~itrapolated Note - Data for [a) may seem at variance with standard smoking. This is because under puff-by-puff conditions integer puff numbers only are measured and, since deliveries in the later puffs are high, the loss of one or one plus of these later puffs has a pronounced effect. TABLE 2 - Derived data from Table i Nicotine increase of FAVOR cigarette, in %, relative to: Puff Number 1 2 3 4 5- 6 7 Sum (a) SEFK with 20m CA filter 220 136 109 80 69 70 63 92 (b) S~K with lOmm CA filter 143 77 56 47 39 42 33 53 Sul/lIlar~ FAVOR source elements appear to be an attractive proposition for increasing nicotine deliveries to smoke. This work demonstrates an across-the-board increase wlth a preferential increase in the first few puffs. However, before enthusiasm becomes rampant, It t s going to be necessary to obtain FAVOR elements without added ~avours and~ith much lower nicotine contents, before product testing with consumers ts attempted. Further, there are patent conslderations,-and cost impllcatlons If a dual or triple filter configuration is involved. Dt strlbution: Mr A L Heard (2 copies) Or T Hlrjl/Dr R R Baker I cr~ L~ P~j m C IJ b P [~" F~I ~+~t'l b~c~6- n'l --- "

Ill ilt !lllllll iii ,,.,,,, I!, +,, ~i lilii'ii Illllill+ I! IIllllllll !11111ll . i!111111 I !INIIlI! INIIIII Itliiiil IIIIIIII Iit11111 !11111"~ !lllllll !:i: t'L • .-0~0069201

B Our Ref: RPF/BCH 22 January 1986 Mrs. C. McBride Impertal Tobacco Ltd. P.O. Box 6500 C.P. Montreal P.Q. Canada H3C 3L6 Dear Cathy, Further to our telephone conversation regarding consumer test methodo|ogy for the FAVOR product/concept, I am pleased to enclose relevant details and materlals used in the U,K, approach: 1. OUTLINE OF METHODOLOGY AGENCY LIAISON DETAILS 3. QUALITATIVE SESSION PROTOCOLS 4. QUESTIONNAIRES FOR STAGES 1-3 5. SUPPLEMENTARY PRODUCT TESTING QUESTIONNAIRE 6. TOKEN ALLOCATION EXERCISE FORMS 7. PRODUCT USAGE OIlY 8. SET OF I9 X 35ff4 SLIDES USED FOR CONCEPT EXPOSURE These represent all the materials you wtl1 need to reproduce our methodology. However, I strongly recommend that we 11atse by telephone to cover aspects of Intervle~Ing/questlonnalre administration and group work. The U.K. research Is now complete and data analysls/Interpretatlon ~11 be completed by January 30th. Our experlence of the methodology usa has been very favourab|e and we have no consumer attrltlon. I w11], of course, brief you on our deta11~ flndlngs when~hey become avallab}e. Wtth best regards, You?s stncere_ly, R.P. FERRIS" CeCo Mr. A.L. Heard, BATCo. Rtllbank Dr. R. Btnns Dr. T, HlrJl P~ CT~ ~u3 0 L J< C IJ b P D'P-'~-,'o,"?~:,',; ~:,~YT fra's't'i'o .-'. C~0~il

JOHN MUMFORD & ASSOCIATES LIMITED Dr. R.P. Ferris Group Research & Development Centre B. A.T. Ltd Regents Park Road Southampton Hants Vt'Er7 H~se ~-IoI ~ s~ vv~ 7HB Tel: 01-4:37-1224 2nd January 1986 Dear Rob, Project Goodbov Thanks for your recent visit and your brief on the above project. This is to confirm the major points of our discussion and to detail some of the logistics. i. The project design is covered in your Consumer Research Proposal dated 17th December 1985, a copy c~which was passed to me. 2. JMA involvement in the project is as follows: a) b) d; Printing the Questionnaire. Booking, briPfing and paying Interviewers. General Admi istratien: including planning interviewer outes, handcounts, liaison with yourself, etc. Computing results. a) Printing.the O, uestionnaire The-questionnaire is divided into three sections: Stage i Stage 2 Stage 3 Administered in Group) " " Home) (Self-completion collected at second group- convention; You will be writing the questionnaire; we shall code it, lay it out and print it. To ensure ease of administration we will print each section in a different colour. Could we please have your handwritten questionnaire by lOth January. The questionnaires will probably be brought to Southampton on 15th January, however, if there is time we will send them in advance of this date. b~ Interviewers I have booked five interviewers (all with cars) to work between 1.30.pm and 7.30.pm on the 15th and ibth January. cent .... Ob'~r.ccs: ,Jcl~ Mum~r.c:l, Sally IVk..'m~r-~. ~e~...~l~l No.1810~S8 F:l~l;ePe~l off.e: 5 Greys~C~-e CouP~ CPow~c~'~e, BePl<s. IVlembeP Of ~~ ~ ASsoCJ.Gio~ of S~h M.~I~; Rese~ COI'v~I~S 0 O~ 0 LJ~ ClibPDF - www.fastio.com

Dr. R.P. Ferris B.A.T. Ltd -2- 2nd January 1986 On each of these two days the following time-table would apply:- 9.30-10.OOam RF meet with Sally Cruickshank to discuss research design 10.00Jll.OOam RF conduct Group 1 ll. O0-11.15am SC arrange recall times with each respondent. Mark each respondent address on the map. ll.15-12.15pm SC plan routes, allocate interviewers. 1.30-2.30pm Interviewer briefing, allocation of interview assignments. 2.00-3.00pm RF conduct Group 2 3.00-3.15pm SC arrange recall times with each respondent. Mark each respondent address on map. 2.30-4.30pm Each interviewer to visit and interview two respondents. Return to Dolphin Hotel. 4.30-5.30pm Interviewers re-briefed for second phase of interviews. New addresses issued. 5.30-7.30pm Each interviewer to visit and interview two respondents. Return to Dolphin Hotel. Handcounts. 7.30-8.30pm Given that you do not want more than about four hours to elapse between the end of the group and the re-interview, the following apply:- . End of Group Interviewer leaves hotel Intervieer arrives at first R. Interviewer leaves first R. Interviewer arrives at second R. * viz. ** vim. 11.15 .am 3.15.pro 2.30.pm 5.30.pm 2.50.pm* 5.5O.pm* 3.2O.pm 6.20.pm 3.40.pm~* 6.40.pm** 3 hours 35 minutes after end of Group 4 hours 25 minutes " " " " Average 4 hours The above timings assume: a twenty minute journey time, a half- hour inter~ew length. c) The project will be run by Sally. She will meet with you at cont ..... O Cr~ O t~ ClibPDF - www.fastio.com

Dr. R.P. Feris B.A.T. Ltd -3- 2nd January 1986 9.30.am at the Dolphin. Your secretary will book a room with bath for Sally for the night of the 15th. Handcounts of results will be available at around 9.00.pm on both the 15th and 16th. d) The questionnaires (for all three stages) will be computer analysed by JMA. The questionnaire for Stages 1 and 2 will be brought back to London by Sally. The questionnaires for Stage 3 will be sent to JMA by you once they have been checked. When we have the questionnaires for all three stages we can process them. This will take about two or three days to put on disc: once this is done you can either visit us to select and run-off tables or we can let you have a basic set which can then be subsequently augmented by ad hoc runs. 3. The cost of the project as described will be £1,600 (exc. VAT). This cost excludes incentives. If you wish we can pay the StaEe 2 incentives (i.e. 40 at £5) and bill you for this. Please let me know if this would be useful. I would like to allow a contingency of £300 to cover any additional processing {such as analysis of variance) where we may need to put the processing out to a d.p. company. Additionally, in order to ensure (as far as possible) that we allow for bad weather and interviewer problems, a further £100 should be held in reserve for taxis and recruiting another interviewer at short notice. I think this covers the major points. I will be in touch in the next few weeks to discuss developments and progress. Thanks ~gain for your brief. Best Wishes for a Happy and Prosperous 1.986. Yours si-~rely, J.W. MUMFORD i O C~ O ClibPD'F - ~,'c'~"l['lO.COlll .........