AI in Retail: 6 Ways Generative AI is Being Used for Consumer and Market Insights

Home improve­ment pro­fes­sion­als and com­pa­nies rely on core insights about con­sumer behav­iors and mar­ket trends to build suc­cess in the industry.

Iter­a­tions in tech­nol­o­gy, and access to that tech­nol­o­gy, has changed the abil­i­ty for orga­ni­za­tions to gath­er and ana­lyze data at scale.

One tech­nol­o­gy that is cur­rent­ly the sub­ject of a lot of buzz is gen­er­a­tive arti­fi­cial intel­li­gence (AI). In recent years, it’s become inte­grat­ed into the mar­ket research process in a vari­ety of ways to sup­ple­ment more tra­di­tion­al meth­ods and is unique­ly posi­tioned for use by strained cus­tomer insights teams to get gen­uine insights out of large data sets.

What is Generative AI?

Gen­er­a­tive AI, as a tech­nol­o­gy, has been around for decades. It was ini­tial­ly intro­duced in the form of chat­bots on web­sites. How­ev­er, since the 1960s, this tech­nol­o­gy has become more refined. Today, gen­er­a­tive AI can pro­duce rel­a­tive­ly authen­tic-seem­ing imagery, text, audio and syn­thet­ic data on its own, or through a self-gen­er­a­tion method. This is due to the imple­men­ta­tion of gen­er­a­tive adver­sar­i­al net­works (GANs) and Vari­a­tion­al Autoen­coders (VAEs) for train­ing neur­al net­works to pro­duce fresh data, along with the advent of trans­form­ers and recent advances in large lan­guage mod­els (LLMs).

As the inter­faces for tap­ping into gen­er­a­tive AI have become increas­ing­ly sim­pli­fied, the oppor­tu­ni­ties for its use have surged. This is the tech­nol­o­gy that you see being uti­lized by con­tent cre­ators to make innocu­ous deep­fakes on social media; by the film indus­try to pro­duce bet­ter dub­bing; and by devel­op­ers to gen­er­ate new pro­gram­ming code. These are just a few exam­ples of gen­er­a­tive AI and how it is used, and there are a myr­i­ad of oth­ers that range from ben­e­fi­cial to neu­tral to nefarious.

How is AI Being Used in Market Research for Retail Products?

Gen­er­a­tive AI is also mak­ing an appear­ance in mar­ket research and cus­tomer insights, where its var­i­ous appli­ca­tions are used to gen­er­ate con­sumer and mar­ket insights. Along with oth­er mar­ket research tools, the tech­nol­o­gy has enabled home improve­ment orga­ni­za­tions to both gath­er and ana­lyze more infor­ma­tion about their tar­get audi­ences and niche mar­ket seg­ment to inform changes in go-to-mar­ket strategies.

From design­ing more tar­get­ed sur­veys to iden­ti­fy­ing trends in qual­i­ta­tive and quan­ti­ta­tive data, here are some of the ways cus­tomer insights and oth­er pro­fes­sion­al teams are employ­ing gen­er­a­tive AI in mar­ket research:

1. Developing Research Hypotheses

For many home improve­ment orga­ni­za­tions, chat­bots are a famil­iar aspect of gen­er­a­tive AI, with Chat­G­PT, devel­oped by Ope­nAI, being released in Novem­ber 2022. This tech­nol­o­gy, trained using data up through June of 2021, mim­ics brain cells to gen­er­ate human-like respons­es about var­i­ous top­ics from inputs by customers. 

Cur­rent­ly, it is being imple­ment­ed to gath­er cus­tomer feed­back through chats. Cus­tomer and mar­ket insights teams can then use that data as a start­ing point for devel­op­ing hypothe­ses to use in fur­ther mar­ket research. They can even inte­grate a pos­si­ble hypoth­e­sis into a con­ver­sa­tion with con­sumers and then ana­lyze the responses.

2. Data Sifting and Assisting with Sentiment Analysis

Accord­ing to a sur­vey by Adobe, a major­i­ty, or 88 per­cent, of cus­tomer expe­ri­ence and mar­ket­ing pro­fes­sion­als feel that gen­er­a­tive AI will help them iden­ti­fy new audi­ences in their mar­ket seg­ment. Addi­tion­al­ly, these pro­fes­sion­als see use cas­es to bet­ter under­stand con­sumer sen­ti­ment and behav­ior pat­terns.

Although it’s viewed as less accu­rate than data gath­ered through empir­i­cal research, gen­er­a­tive AI can sift through large amounts of data — from social media, chats, reviews, and oth­er text — to iden­ti­fy pat­terns; look for con­nec­tions between par­tic­u­lar phras­es and words; and high­light par­tic­u­lar con­sumer atti­tudes toward a brand and its prod­ucts or ser­vices. From there, com­pa­nies can find ways to improve based on this sen­ti­ment analysis.

3. Identifying Themes and Understanding Trends

Gen­er­a­tive AI is also being applied in nat­ur­al lan­guage pro­cess­ing capac­i­ties on the report­ing side of mar­ket research: to iden­ti­fy themes and trends in respons­es gath­ered through qual­i­ta­tive inter­views or sur­veys. Rather than hav­ing to man­u­al­ly sort through each and every answer from respon­dents to look for themes and trends, insights teams are begin­ning to use gen­er­a­tive AI to do this work for them, which is viewed as a time-sav­ing mechanism. 

Addi­tion­al­ly, gen­er­a­tive AI is applied in a broad­er fash­ion in order to scout exter­nal datasets and dig­i­tal con­tent to iden­ti­fy and pre­dict mar­ket trends. It will scour large amounts of texts and cre­ate short sum­maries. By under­stand­ing how the mar­ket is chang­ing or how cus­tomers’ needs and pref­er­ences are evolv­ing, home improve­ment brands are able to keep pace with the speed of change among the mod­ern customer’s pur­chase sen­ti­ments and adapt their strate­gies accordingly.

4. Assisting with Pattern Detection

Mar­ket researchers are fre­quent­ly look­ing for pat­terns to gain bet­ter mar­ket and con­sumer insights, and that’s where gen­er­a­tive AI has found anoth­er use. It ana­lyzes datasets and detects pat­terns and con­nec­tions between mul­ti­ple vari­ables that have to do with con­sumer behav­iors and pref­er­ences; prod­uct and sales per­for­mance; and mar­ket trends.

This type of pat­tern recog­ni­tion, or pat­tern detec­tion, bet­ter enables home improve­ment com­pa­nies to explore ways to opti­mize their prod­ucts and adjust mar­ket­ing strate­gies based on these insights.

5. Creating Predictive Models

Some cus­tomer insights ana­lysts are find­ing ways to use Gen­er­a­tive AI as a pre­dic­tive mod­el­ing tech­nique to pro­vide insight into trends sur­round­ing pur­chas­ing deci­sions made by home improve­ment pro­fes­sion­als and cus­tomers.

The premise is using his­tor­i­cal sets of empir­i­cal data with pre­dic­tive mod­el­ing in order to make a pre­dic­tion about future behav­ior and pur­chase trends.

In the day-to-day work of cus­tomer intel­li­gence, this means insights teams are able to

• iden­ti­fy new cus­tomer journeys

• stay abreast of buy­ing and sell­ing decisions

• bet­ter per­son­al­ize cus­tomer experiences

• fore­cast cus­tomer demand and stock prices

• under­stand more about prod­uct per­for­mance in an indus­try segment

6. Generating Real-Time Insights

Mar­ket demand is shift­ing towards a desire for real-time ana­lyt­ics, which is why many orga­ni­za­tions are exper­i­ment­ing with gen­er­a­tive AI. Vir­tu­al assis­tants, chat­bots and avatars are used to har­vest live data via web­site traf­fic and to some degree, auto­mate ​“social lis­ten­ing” on social media. With this tech­nol­o­gy deployed orga­ni­za­tions are able to con­tin­u­al­ly col­lect real-time insights on a range of topics.

Optimizing Market Research for Improved Insights in the Home Improvement Industry

The tech­niques used to gath­er, syn­the­size, and ana­lyze mar­ket research will con­tin­ue to evolve as new tech­nolo­gies and tools, such as gen­er­a­tive AI, emerge and become avail­able for pub­lic use.

No mat­ter what method­olo­gies, tech­niques and process­es you pre­fer, it’s impor­tant to uti­lize indus­try-spe­cif­ic data and insights around con­sumer behav­iors, mar­ket trends, and prod­uct performance.

One option is becom­ing a mem­ber of the Home Improve­ment Research Insti­tute (HIRI). You will increase your organization’s access to a wealth of valu­able research and on-demand data about home improve­ment prod­ucts cat­e­gories specif­i­cal­ly, to help you and your teams make informed deci­sions for the future of your company.