Search

Predicting an Influencer’s Results: Ad Value, Engagement Value & Sales Value

According to BigCommerce.com 61% of marketers agree it’s challenging to find the best influencers for a campaign, suggesting this problem is far from solved.


There are a lot of factors that feed into this issue - what is the best platform to use? What is the most efficient process to do so? Who should I be looking for?


And one of the most common - how can I predict the results that this influencer will bring me?


As time is passing, we are seeing influencer marketing really break down into it's own categories.


Phillip Brown (link), founder of the Influencer Marketing Academy based in London, UK - (who I like to vibe with!) has even argued that we should be distinguishing between "influencer marketing" and "influencer advertising" given how many brands are measuring influencer success primarily on advertising metrics, leading influencers to also treat relationships as "rented space" or glorified billboards. (You can view the full article here ).


While many brands take the "influencer advertising route", some brands have shifted their focus to using influencer marketing strictly for "engagement", while others still focus on the referrals, end sales and "actions" for a brand that can be generated through influencer marketing.


Before we even get into predicting if an influencer is going to be successful, we have to ask ourselves - on what premise? What am I looking to GET out of this campaign, partnership or program. How am I measuring it's success? (I got deeply into all of the different values you can measure here and here, if you need to play catchup!)


Once you establish the value you are looking for, it's actually pretty easy to predict whether an influencer partnership could be valuable to you so long as you can grab the right metrics - or keep industry benchmarks handy.



You can download this template, complete with formulas from Google Sheets to your preferred Spreadsheet Program HERE.


Let me take you through 3 lenses for the most popular ways to measure an influencer's impact: advertising value, engagement value, and sales value. I've developed this resource [download link] complete with benchmarks and formulas that you can download and utilize to help analyze your influencers. The formulas are embedded within the spreadsheet so all you have to do is plug-in your influencer data.

Note: At the moment this resource is best used for Instagram. While Advertising value can be adapted for other platforms, engagement and sales value rates and interactions differ heavily and will require their own resource. Interested in adaptions? Email me.



Using the Spread Sheet with BENCHMARK DATA (AS IS)


I have done a sample analysis of 2 influencers in the health and wellness category (this is a real example). The brand in question was looking to see what kind of value they may receive from partnering with two major accounts on a new retail distribution partnership.


In Column A, fill in the different configurations of posts. I like to separate feed and stories specifically for Instagram because they have different benchmarks for reach and distribution as well as different paths to sale. Separation is more accurate.


In Column B, you want to fill out the influencer's cost PER piece of content (Green Boxes). I included a bundle here looking strict at the FEED results prediction as you can look at each line item separately to assess both values (and to show the brand that option). You can use either the influencer's REAL rate, or what you expect to pay an influencer in this column.


In Column C, you want to fill out their total audience size (Green Boxes).


Avg. Cost Per Audience was OPTION (Column D), but for this client, since I knew the rates were low against other accounts this size, I plugged in the AVERAGE cost of what an influencer with this audience size costs to compare the two. The "quick" formula, especially from an "AD VALUE" perspective is $100 for every 10k followers, which comes out to roughly $10 CPM (the top of the industry average).


Column E and F auto-populate based on the following formulas:

  • (E) Estimated Impressions = Audience Size x Average Reach Rate [for stories we used benchmarks of 4%, the average being that 4-10% of an influencer's audience will see a story. This slides in scale based on the size of the influencer. This influencer was macro sized, so we used a lower average. For feed it's about 20-30% of an influencer's audience that sees a feed photo, we used 25%, the middle reading, as there is no intel as to whether the size of the influencer affects reach rate.]

  • (F) Cost Per Impression = Cost / Estimated Impressions

This influencer has high advertising value. They have competitive pricing with, let's say, a Facebook ad coming in at 0.01-.02 per impression. If a brand were measuring the ROI here similar to that of other ad methods, these accounts would have high value. Not only do they directly reflect the same benchmarks, we know that influencer marketing (over other ad forms) still has further benefit and value (it comes from 3rd party source, it's human, etc.)

Moving on to engagement, I took readings on some different categories that were most closely related to the interests of the brand.


The brand plays in the health and wellness space, so I looked at posts in fitness apparel, protein powder brands, bars, natural products - reading the engagement rate on these sample posts would help me understand how engaged an audience was on the topics related to the brand. This is Category Engagement Rate. On average, accounts with 1M+ following have an engagement rate of 1%-2%, meaning this influencer in question's overall engagement is simply average.

Category Affinity Comparison is how that engagement rate stacks up with other topics on their page. For instance, if engagement rate is higher on, let's say, HOME GOODS or CHILDREN, while the INFLUENCER may be a fan of health and wellness, the BULK of their audience may not be. This is how we can help distinguish between topics that interest INFLUENCERS and those that interest AUDIENCES to better predict whether there is engagement value. Higher category affinity - or higher engagement rates for the categories that best align with client categories, will produce better engagement value.


This specific influencer is average on engagement value, however it was worth noting to the brand that of all of their posts, natural products, bars and similar categories to the brand's products were some of the top engaging on that page. The bulk of their audience is most interested in healthy living / natural products.


Finally, sales value.


Sales value is a bit difficult to claim "perfection in formula", because there is extreme variation in an influencer's ability to sell [there are indicators, this is a convo for another day though!] so the benchmarked metrics may not be TOTALLY accurate and will give you a really wide potential range. But it still puts a quantifiable number or prediction against a rate to determine if this really makes sense for a brand, measuring on sales value, to pursue.


Columns I, J and K are also autofilled on the following formulas:


  • (I) Estimated SwipeUps = Estimated Impressions x SwipeUp Rate [For this example we used the average swipe-up rate of 10%. The range is between 10-20%, we chose a lower reading.]

  • (J) Estimated Sales Potential = SwipeUps x Conversion to Sale Rate [Conversion metrics can vary, but for this benchmark we used .5% of swipe-ups that could covert to sale. Average conversion rates are around 1%, we took a smaller reading because this was not an e-commerce based transaction.]

  • (K) Cost Per Acquisition = Influencer Cost / Estimated Sales . This will give you how much you paid the influencer PER referral, which can be balanced against your target Cost Per Acquisition Ranges.


This particular influencer would have a high CPA to start. While it's possible that they sell more than predicted (especially given we used lower metrics to estimate) it still may only drop CPA slightly. If this number comes close to an acceptable target for you, this influencer may have high potential sales value.


Keep in mind the ease or difficulty of audience obtaining your product - e-commerce brands have the greatest ease, while retail brands have a disconnect between online consumption and POS. Influence has still been known to affecting purchasing decisions but your retail footprint will play a large role in determining the followthrough.

Using the Spread Sheet with REAL INFLUENCER DATA


Can we gain a more accurate prediction? Yes! Influencers and talent managers should be able to provide you with an influencer's own Reach Rate, Swipe-Up Rates or Average Conversion Rates - metrics currently allow for this. Save Rates may also help you predict conversion on feed photos.


If you'd like to use REAL influencer data, you will simply need to change the excel formulas for each analysis you do. Here's what you can change.


  • In Cell E6 = Swap 0.04 (Benchmark) for the influencer's own Reach Rate % (expressed as 0.04 = 4%)

  • In Cell E8 = Swap 0.25 (Benchmark) for the influencer's own Reach Rate % (expressed as 0.25 = 25%)

  • In Cell I6 = Swap .10 (Benchmark) for the influencer's own Swipe Up Rate %

  • In Cell I8 = Include the formula '=E8*0.XX' where 0.xx is the influencer's save rate. This will give you a reading on the rate in which their content is bookmarked

  • In Cell J6 = Swap .005 for an influencer's conversion rate (if they know it). It can be rare to find an influencer who can provide this, but I do teach this to influencers (running tests using Calls-to-Action) so they may have some familiar data that can give you a better idea of whether they covert on the low end (0.005) or higher end (0.02)


An influencer's own data can be way more accurate and help you make individual predictions over what kind of value your partnership may bring to your brand.


E-commerce-based brands also have the luxury of being able to plug in their own conversion data - if for instance, they know the percent of traffic they typically will convert from a similar kind of referral. This may give a more tailored reading of expected conversions and conversion costs on influence.

While it may not be feasible to run this deep analysis on every influencer you are considering (it could be way time consuming!), not feasible for large scale programs and not as valuable for more complex goals - it definitely has value to help predict results of one-off partnerships, small-scale campaigns and potential long-term relationships and programs - as sales opportunities can also grow over time.



FAQs from People Using This Sheet

(if you have a question to add, shoot us an email!)


Q: For Influencer A you estimated swipe-ups/traffic just off the one story with the swipe-up but it looks like you have a bundle with a story there too. Is there a reason you didn’t include estimated sales off that? Or is this sheet not fully completed and you were just sharing the one line as an example? I also assume you are just estimating sales off clicks here as I don’t see sales correlating to the giveaway or IG feed post. Thoughts on how you might estimate sales off that?


Lot's of great questions. To break it down...


  • I didn't include estimated sales off of FEED posts because, unlike stories, there is not a direct line to a sale since feed captions cannot contain hyperlinks. The path is disconnected as Instagram does not yet have the capability for influencers to have shopable products (unless you completely isolate influencer timing or we are working distinctly in fashion with Liketoknow.it) There ARE workarounds that many brands use such as bit.ly links within captions (easily re-typeable, just so the option is there) but given the inconvenience its not as reflective as what "total" sales end up being from feed-created awareness. You are missing that ability to really trace and attribute search, those who may flow through a brand's own social media assets, but originated from an influencer, and more measurement points that can be difficult to isolate. There's also not as much data on feed conversion to sale as there is on SwipeUp paths.

  • Because it's difficult to estimate sales off feed, anywhere there is FEED, I have left off sales (though I did include a method above in relation to bookmarks!). Even though the package contains 1 story, the CPA would be misleading because the cost of that package is higher (because it includes feed). In this instance I would simply advise the brand to consider that the 1x story aspect of the package will be a BASELINE for what can be expected for sales.

  • I do not provide sales value on giveaways because most of the time, giving away product is counter productive to trying to drive sales. Giveaways have high ADVERTISING and ENGAGEMENT (aka 'Brand Awareness') value because their impressions and engagements are typically higher than what is predicted in this sheet, but lack followthrough when people try to "win".

stay connected

navigate

  • Instagram
  • Facebook
  • Pinterest
  • Twitter
  • LinkedIn

© 2020 Social Creates Impact 

All Rights Reserved.

design: by me  ///  hosted:  wix.com

photography:  Lauren Nicole   ; Various Friends

for brands

for creators