7 ms·

People inexperienced in online adversities often think that they can pay for conversions. That is not the case. You’re paying to get your message out. Very very

by ttul 9y ago

People inexperienced in online adversities often think that they can pay for conversions. That is not the case. You’re paying to get your message out. Very very few people will immediately act because of an ad. But if you hit them enough, they may later convert. Tracking those late conversions is imprecise. This does not mean advertising is useless. But it is not generally a game for those with a tiny budget or a short time horizon.

petercooper 9y ago
This lesson struck me hard when I was working with a major conference organizer. Even upon surveying them, they'd struggle to establish how attendees had heard of the event first. For the majority of attendees there was no direct way to correlate ticket purchases with ad spend because almost no-one immediately buys a $1000+ ticket after clicking an ad :-)
- jackgolding 9y ago
  Market research is the easiest way to solve these problems!
  - ChristianGeek 9y ago
    What if you provide a discount code in the ad that’s unique to the ad? Subsequent ads have slight lower discounts so there’s an incentive to use the code from the first ad they saw.
    - lozenge 9y ago
      Not being able to find a discount you saw can be pretty frustrating. Less important in this case as employers often pay for these conferences.
    - icebraining 9y ago
      Might work for some people, but the probability that I would even remember the first ad had a code - let alone the exact string - is minuscule.
    - petercooper 9y ago
      Discount codes was one method used, the only problem was that at the $1000+ ticket level, it's often another department buying the tickets and they frequently don't bother to use the codes :-D
libertine 9y ago
Precisely - "marketers" these days don't grasp the concept of brand building. Just because you can measure directly results it doesn't mean the goal is to get a direct conversion - specially from a platform like youtube where people are actively consuming content - not buying stuff. The idea of compound investment into brand building doesn't seem to exist. I don't know if these guys thought that decades of TV advertising campaigns were made by clueless people, with no education, with no market research, benchmarks, that dumped millions to build reach and frequency to try to fill a small spot in the mind of consumers.
- DisruptiveDave 9y ago
  I'd venture to say it's not the marketers acting on their own in many cases. Shit runs down hill. VCs want fast returns, CEOs want their investors to be happy about their CAC, marketers want to prove their value in this system. And EVERYBODY is obsessed with analytics, often as a poorly constructed crutch. You try pitching brand building in that environment.
  - e_b 9y ago
    Exactly this.
  - libertine 9y ago
    But I understand that, so maybe it's a matter of choice. What kind of brand you want, and set it right away from the beginning because it will define your media channel/production investment. I think that brands who had strong branding advertising are still well positioned in people's minds, and it's indeed an investment that should be made with a massive ROI. Now, it's a long term game. Doesn't suit the status quo like you well said.
    - e_b 9y ago
      The long term brand plan must be set by the CEO and in the example provided by the parent, the CEO only cares about keeping investors happy in the short term, alas no spend without immediately identifiable ROI.
jehlakj 9y ago
Indeed. People are subjected to feel more secure about a brand that is more familiar to them (e.g., Tide).
sfifs 9y ago
There actually are ways like market mix models and multi touch attribution models to account for impact of those "late conversions" across multiple channels but requires fairly sophisticated data integration and modeling and is generally worth it only if you are optimizing millions of dollars in ad spend. The tools Google/Facebook currently provide are not really sophisticated enough for such work. Source: I work on these techniques
- KodiakLabs 9y ago
  Could you possibly point to by resources on these models?
  - sfifs 9y ago
    While details tend to be proprietary, MMMs are well established and lots of academic papers online. For instance https://web.archive.org/web/20120813035903/http://faculty.fuqua.duke.edu/~mela/bio/papers/Ataman_Mela_Van_Heerde_2006.pdf https://web.archive.org/web/20120813035903/http://faculty.fu... High quality multi touch attribution models are still fairly new and I've not seen much public literature on the designs. We've had to spend a fair amount of time in development ourselves vs. relying on existing methods.
- istorical 9y ago
  I work in analytics in the sales and marketing department at a late stage startup. I was recently tasked with moving from just reporting to working in modeling and prediction. A first step I thought of was to build some multiple variable models / apply multiple linear regression to try to measure the relative influence of our marketing campaigns and touches to a sale. With your experience in this area - are there are particular models you recommend trying to build as a first step for someone who's more of a beginner in modeling and predictive analytics? Or business questions you find go the furthest within this specific domain of sales and marketing?
ekianjo 9y ago
Indeed you mainly just buy awareness of your service. The road from awareness to trial can be quite long.
ggg9990 9y ago
Disagree. I had a startup selling a piece of software that immediately solved a major problem for a small niche of users. Paid for conversions on common search terms and they were ROI positive since each click converted at 25% to signup, and then the product spread around the companies where one user adopted. Sold the company for >$10m.
- dylanha 9y ago
  How did you find the niche? Any reading material I should see?
  - ggg9990 9y ago
    I had a CS degree but was working outside the field in pharmaceuticals. I was having the problem myself and wrote a crazy Excel spreadsheet with tens of thousands of lines of poorly written copy pasted code to do it. I knew that there would be a market for it so I contacted an old classmate of mine who rewrote it much better as a web service. That is why I think CS and software engineering should be part of every college degree, so that people can see the opportunity for software where it exists. I wasn’t good enough to be a programmer but I had enough knowledge to see an opportunity whereas most people thought “hm, this is just how this is done.”
    - nathan_f77 9y ago
      Reminds me of this AMA answer by @patio11 [1] > Here's an exercise you can do: do you understand what a life insurance agent does all day every day? Make it your mission for a week to do so, well enough to explain it to a close friend who has no access to your sources. All you have to do to learn this is read and make conversations happen. (People are happy to talk to you!) I really need to take that seriously, and talk to people in a lot of different industries. [1] https://www.indiehackers.com/forum/im-patio11-patrick-mckenzie-creator-of-appointment-reminder-ama-fb96fd1f86?commentId=-L5qqQU1KexqFnGATD7I https://www.indiehackers.com/forum/im-patio11-patrick-mckenz...
- 282883392 9y ago
kraig911 9y ago
Uhh look up funnel marketing. There's a whole system in place that exercises and excels in exactly what you say we in advertising are inexperienced? I mean you can technically pay for views it's just not anything over 5% of impressions though... retargeting though works.

by ttul 9y ago

petercooper 9y ago

This lesson struck me hard when I was working with a major conference organizer. Even upon surveying them, they'd struggle to establish how attendees had heard of the event first. For the majority of attendees there was no direct way to correlate ticket purchases with ad spend because almost no-one immediately buys a $1000+ ticket after clicking an ad :-)

jackgolding 9y ago
Market research is the easiest way to solve these problems!
- ChristianGeek 9y ago
  What if you provide a discount code in the ad that’s unique to the ad? Subsequent ads have slight lower discounts so there’s an incentive to use the code from the first ad they saw.
  - lozenge 9y ago
    Not being able to find a discount you saw can be pretty frustrating. Less important in this case as employers often pay for these conferences.
  - icebraining 9y ago
    Might work for some people, but the probability that I would even remember the first ad had a code - let alone the exact string - is minuscule.
  - petercooper 9y ago
    Discount codes was one method used, the only problem was that at the $1000+ ticket level, it's often another department buying the tickets and they frequently don't bother to use the codes :-D

libertine 9y ago

Precisely - "marketers" these days don't grasp the concept of brand building. Just because you can measure directly results it doesn't mean the goal is to get a direct conversion - specially from a platform like youtube where people are actively consuming content - not buying stuff. The idea of compound investment into brand building doesn't seem to exist. I don't know if these guys thought that decades of TV advertising campaigns were made by clueless people, with no education, with no market research, benchmarks, that dumped millions to build reach and frequency to try to fill a small spot in the mind of consumers.

DisruptiveDave 9y ago
I'd venture to say it's not the marketers acting on their own in many cases. Shit runs down hill. VCs want fast returns, CEOs want their investors to be happy about their CAC, marketers want to prove their value in this system. And EVERYBODY is obsessed with analytics, often as a poorly constructed crutch. You try pitching brand building in that environment.
- e_b 9y ago
  Exactly this.
- libertine 9y ago
  But I understand that, so maybe it's a matter of choice. What kind of brand you want, and set it right away from the beginning because it will define your media channel/production investment. I think that brands who had strong branding advertising are still well positioned in people's minds, and it's indeed an investment that should be made with a massive ROI. Now, it's a long term game. Doesn't suit the status quo like you well said.
  - e_b 9y ago
    The long term brand plan must be set by the CEO and in the example provided by the parent, the CEO only cares about keeping investors happy in the short term, alas no spend without immediately identifiable ROI.

jehlakj 9y ago

Indeed. People are subjected to feel more secure about a brand that is more familiar to them (e.g., Tide).

sfifs 9y ago

There actually are ways like market mix models and multi touch attribution models to account for impact of those "late conversions" across multiple channels but requires fairly sophisticated data integration and modeling and is generally worth it only if you are optimizing millions of dollars in ad spend. The tools Google/Facebook currently provide are not really sophisticated enough for such work. Source: I work on these techniques

KodiakLabs 9y ago
Could you possibly point to by resources on these models?
- sfifs 9y ago
  While details tend to be proprietary, MMMs are well established and lots of academic papers online. For instance https://web.archive.org/web/20120813035903/http://faculty.fuqua.duke.edu/~mela/bio/papers/Ataman_Mela_Van_Heerde_2006.pdf https://web.archive.org/web/20120813035903/http://faculty.fu... High quality multi touch attribution models are still fairly new and I've not seen much public literature on the designs. We've had to spend a fair amount of time in development ourselves vs. relying on existing methods.
istorical 9y ago
I work in analytics in the sales and marketing department at a late stage startup. I was recently tasked with moving from just reporting to working in modeling and prediction. A first step I thought of was to build some multiple variable models / apply multiple linear regression to try to measure the relative influence of our marketing campaigns and touches to a sale. With your experience in this area - are there are particular models you recommend trying to build as a first step for someone who's more of a beginner in modeling and predictive analytics? Or business questions you find go the furthest within this specific domain of sales and marketing?

ekianjo 9y ago

Indeed you mainly just buy awareness of your service. The road from awareness to trial can be quite long.

ggg9990 9y ago

Disagree. I had a startup selling a piece of software that immediately solved a major problem for a small niche of users. Paid for conversions on common search terms and they were ROI positive since each click converted at 25% to signup, and then the product spread around the companies where one user adopted. Sold the company for >$10m.

dylanha 9y ago
How did you find the niche? Any reading material I should see?
- ggg9990 9y ago
  I had a CS degree but was working outside the field in pharmaceuticals. I was having the problem myself and wrote a crazy Excel spreadsheet with tens of thousands of lines of poorly written copy pasted code to do it. I knew that there would be a market for it so I contacted an old classmate of mine who rewrote it much better as a web service. That is why I think CS and software engineering should be part of every college degree, so that people can see the opportunity for software where it exists. I wasn’t good enough to be a programmer but I had enough knowledge to see an opportunity whereas most people thought “hm, this is just how this is done.”
  - nathan_f77 9y ago
    Reminds me of this AMA answer by @patio11 [1] > Here's an exercise you can do: do you understand what a life insurance agent does all day every day? Make it your mission for a week to do so, well enough to explain it to a close friend who has no access to your sources. All you have to do to learn this is read and make conversations happen. (People are happy to talk to you!) I really need to take that seriously, and talk to people in a lot of different industries. [1] https://www.indiehackers.com/forum/im-patio11-patrick-mckenzie-creator-of-appointment-reminder-ama-fb96fd1f86?commentId=-L5qqQU1KexqFnGATD7I https://www.indiehackers.com/forum/im-patio11-patrick-mckenz...
282883392 9y ago

kraig911 9y ago

Uhh look up funnel marketing. There's a whole system in place that exercises and excels in exactly what you say we in advertising are inexperienced? I mean you can technically pay for views it's just not anything over 5% of impressions though... retargeting though works.